ISCB Africa ASBCB Conference on Bioinformatics 2017


All posters will remain up throughout the conference.  Odd numbered posters will present on Wednesday, October 11 and Even number posters will present on Thursday, October 12.

Poster - A-001
A Computational Pipeline for Unraveling Complex Traits in plants (A Case Study of CMD in Cassava)

Presenting Author: Toyin Abdulsalam, University of Ibadan, Nigeria
Toyin Abdulsalam, University of Ibadan, Nigeria
Andreas Gisel, International Institute of Tropical Agriculture, Nigeria
Seun Opeloye, International Institute of Tropical Agriculture, Nigeria
Livia Stavolone, International Institute of Tropical Agriculture, Nigeria
Short Abstract: Genome-wide transcription profiling is an important and powerful tool leading to the generation of testable hypotheses for novel processes not yet characterized at the molecular level. RNA sequencing (RNA-Seq) provides visibility to previously undetected changes occurring in organisms, under different environmental conditions and across a broad range of other study designs.

However, researchers interested in studying and constructing transcriptome, especially for non-model species, face the challenge of either a poor Reference annotation or none or incomplete reference genome.

Here we present a Computational pipeline for functional annotation of de novo assembled transcriptome using Trinity and Trinotate for none model plants. In our case study we report the de novo transcriptome assembly of CMD infected Cassava using Trinity. Trinotate was used to carry out functional annotation of the assembled transcript and the transdecoder generated peptide sequence along with BLASTX, BLASTP and HMMER searches and loaded into a Sqlite3 database. Each of the original sequence files were aligned to the assembled transcriptome using Bowtie and then RSEM was used to estimate expression values based on the alignment. EdgeR was used to calculate differential expression between the two conditions. We found that a subset of such transcripts was consistently either up-regulated or down-regulated in both TMEB117 and TMS-4(2)425. The gene ontology enrichment demonstrates that these gene products are involved in processes such as metabolic processes, transport and response to stress; they cover functions such as catalytic/transporter activities and nucleic acid binding, and they are mainly associated or integral components of membranes
- return to Poster LIst -

Poster - A-003
Development of algorithm for gene clustering using expression and ontology data: kRCC

Presenting Author: Mahmoud Abu Azoum, Institut Pasteur de Tunis, Tunisia
Adnen Saadaoui, Institut Pasteur de Tunis, Tunisia
Ines Abdeljaoued-Tej, Institut Pasteur de Tunis, Tunisia
Dorra Louati, Institut Pasteur de Tunis, Tunisia
Amira Kebir, Institut Pasteur de Tunis, Tunisia
Slimane Ben Miled, Institut Pasteur de Tunis, Tunisia
Maher Moakher, Ecole Nationale d'Ingénieurs de Tunis, Tunisia
Alia Benkahla, Institut Pasteur de Tunis, Tunisia
Short Abstract: Background: Clustering of gene expression is an unsupervised procedure commonly used to explore the classification of genes into groups. It leads to the profiling of genes sharing similar patterns (transcript/protein expression, gene regulation). The results of such clustering is important in the formulation of hypotheses on the molecular function, biological processes and/or metabolic pathways in which genes are involved. Adding gene ontology information to guide gene clustering would increase the accuracy and stability of clusters and indicate co-function and co-regulation between genes. However classical genes clustering algorithms usually use a single data type: numerical data. Integrating genes’ ontology annotation (binary) side by side with gene expression (numerical) requires an algorithm that is able to cope with a mixture of data types. Method: The k-Real Centroid Clustering algorithm (kRCC), that is introduced here combines the use of the Euclidian distance together with a new metric distance for binary data and a new strategy for the selection of centroids. The validation of kRCC was done by applying a gene set enrichment analysis of produced clusters on interacting proteins. Result: The developed algorithm, kRCC was tested on data concerning 303 genes, extracted from 3 experiments, related to Colorectal cancer and downloaded from ArrayExpress. The gene ontology annotation of those genes was extracted from the ENSEMBL database using the BioMart data mining tool. Applying kRCC on this mixed dataset (gene expression and gene ontology) gives stable and homogeneous clusters of genes in terms of gene expression and of gene function.
- return to Poster LIst -

Poster - A-005
Molecular Isolation and Characterization of Soil Fungi- Trichoderma Koningii From Osun Osogbo Grove

Presenting Author: Nusrah Afolabi-Balogun, Fountain University, Osogbo, Nigeria
Aisha Salihu, Fountain University, Osogbo, Nigeria
Nafisah Oyebode, Fountain University, Osogbo, Nigeria
Azizah Oni, Fountain University, Osogbo, Nigeria
Mutiat Mosobalaje, Fountain University, Osogbo, Nigeria
Falilat Oseni, Fountain University, Osogbo, Nigeria
Uzwat Diekola, Fountain University, Osogbo, Nigeria
Tajudeen Ganiyu, Fountain University, Osogbo, Nigeria
Aminat Lawal, Fountain University, Osogbo, Nigeria
Short Abstract: Materials and Method: Soil samples obtained from Osun Osogbo Groove were cultured on Potato Dextrose Agar (PDA) media; pure colonies after several subculture were subjected to genomic DNA extraction, which was taxonomically and toxicologically characterization using primers specific for ITS 1 and 2, Tri5, coding key enzyme trichodiene synthase as well as Tri3 and Tri4, which are also components of Tri5 cluster around the sesquiterpene cyclase gene.
Results: Nucleotide-Nucleotide blast of PCR sequence on revealed a 449bp ITS gene found to have at least 98% similarity with Trichoderma spp. on the database. Multiple alignment and phylogenetic analysis on shows isolate is most related to Trichoderma koningii and subclass megablast reveal it’s most associated with Hypochea koningii. Alignment against ex-type strain sequences from TrichoBlast and established Trichoderma taxonomy shows sequence contains 449bp nucleotide on which both ITS 1 and 2 were found, three anchors denoting first, second and third genus-specific hallmark (GSH) were located in positions 11, 32 and 175 respectively, anchors for four and five were however not located. Trichokey confirmed sequence as ITS 1 with a short 5.8s RNA sequence and ITS 2 region of Hypocrea / Trichoderma. Toxicological evaluation of the genomic DNA using Tri5, coding the key enzyme trichodiene synthase as well as Tri3 and Tri4, which are also components of Tri5 cluster around sesquiterpene cyclase gene revealed organism lack trichothecene gene.
Conclusion: Organism isolated from soil of Osun Osogbo Grove is Trichoderma koningii, non-toxin producing, hence a potential organism for biotechnological exploit in biocontrol.
- return to Poster LIst -

Poster - A-007
Motif Discovery in DNA Sequences Using an Improved Gibbs Sampling Algorithm

Presenting Author: Usman Lamidi, University of Ibadan, Nigeria
Angela Makolo, University of Ibadan, Nigeria
Short Abstract: Motifs are repeated patterns of short sequences usually of varying lengths between 6 to 20 bases. Within Deoxyribonucleic Acid (DNA) sequences, these motifs constitute the conserved region of most common signatures for recognizing protein domains that are relevant in it evolution, function and interaction. The Gibbs sampling is a Markov Chain MonteCarlo(MCMC) algorithm which has been applied in the past to discover motifs in DNA sequences. A problem with this technique is the profusion of iterative operations in the sampling process because it progressively chooses new possible motif positions from a continuous randomize sampling in DNA sequences. We applied an Improved Gibbs (IGibbs) sampling algorithm on Breast Cancer human disease DNA sequences to overcome this unwieldy iteration by altering the processes to obtain a reduced runtime and also achieve an accurate satisfactory motif result. The methodology applied in IGibbs algorithm takes an input of .gbk or .fasta DNA file and creates a list of all nucleotides to predict a random sampling starting position. It applies motif length, lesser iterative value and further computes the probability and position ranking scores using Position Weight Matrix (PWM). The algorithm was implemented using Python,Python(x,y) and Biopython. The IGibbs algorithm was evaluated using varying motif lengths of 12, 18 and 24 on different base lengths of 5,000, 10,000 and 15,000. The result showed that the IGibbs returned a better average runtime of 7, 10 and 23 seconds respectively compared to 12, 32 and 60 seconds respectively in the existing Gibbs sampling algorithm found at
- return to Poster LIst -

Poster - A-009
H3ABioNet​ ​ node​ ​ assessment​ ​ exercise:​ ​ the​ ​ CBSB​ ​ experience

Presenting Author: Azza Ahmed, University of Khartoum, Sudan
Rehab Ahmed, University of Khartoum, Sudan
Somia Mohammed Somia, University of Khartoum, Sudan
Faisal​ ​ M.​ ​ Fadlelmola, University of Khartoum, Sudan
Short Abstract: Recent technological advancements has moved the field of Genomics from purely wet lab based approaches to being one of the most demanding computational and data sciences. In particular, variant calling in Next Generation Sequencing (NGS) experiments aiming at the discovery of disease causing mutations are one such example where more emphasis is now put on the computational analysis of biological samples and the interpretation of resulting variants​ called.

By designing node assessment exercises, H3ABioNet is providing a platform and foundation for its participating nodes to focus on developing computational and analytical skills within its scientists, to empower their own research, and also gauge their potential as collaborators in the​management​ and​ ​analysis​ ​of​ ​datasets​ ​generated​ ​within​​ H3Africa​ ​projects​ and​ ​beyond.

As the Sudan Node within the H3ABioNet, we carried out the variant calling for exome sequencing exercise successfully by taking advantage of training opportunities in the form of workshops and internships of some of the participating scientists. We amended these skills learned by assembling a local team of researchers who took further advantage of the the Standard Operational Procedures and datasets provided by the Node Assessment team within H3ABioNet. In this poster, we reflect on our experience in the accreditation exercise and​ ​the​ ​ lessons​ ​ learned​ ​ from​ ​ participation.
- return to Poster LIst -

Poster - A-011
Tannase obtained from a novel Bacillus pumilus strain AJ-2 MF083685 from Bambara (Vigna subterranea (L) Verde) nuts *Ajayi A.A., Egunjobi, A.K., Olasehinde, G.I., Anosike, S.O., Onibokun, A.E., Department of Biological Sciences, College of Science

Presenting Author: Adesola Ajayi, Covenant University, Ota, Ogun State, Nigeria, Nigeria
Grace Olasehinde, Covenant University, Nigeria
Adeyinka Egunjobi, Covenant University, Nigeria
Adeola Onibokun, Covenant University, Nigeria
Selina Anosike, Covenant University, Nigeria
Short Abstract: Tannases are hydrolytic enzymes that can be obtained from fungal and bacterial sources and they are very important with diverse uses in industries. There is the need for cost effective production tannase. DNA extraction was carried out on tannase producing bacterial isolate using the Jena Bacteria DNA Extraction kit according to manufacturer’s instructions. The purity and concentration of the extracted DNA was evaluated using a Nanodrop (ND 1000) Spectrophotometer (Thermo Scientific, USA). Polymerase chain reaction was carried out to amplify the 16SrRNA gene of the bacteria using the primer pair 27F- 5′- AGAGTTTGATCCTGGCT CAG -3’, and 1492R 5’-GGTTACCTTGTTACGACTT -3’. All PCR products were purified with spin column based PCR purification and sent to Epoch Life science (USA) for Sanger sequencing. Molecular identification of the tannase-producing bacteria revealed Bacillus pumilus strain AJ-2 and it has been deposited in the NCBI GeneBank (Accession number MF083685). The DNA yield was between 5ng - 25ng. It was optimally pure showing A260/A280 between 1.60 - 1.80. Amplification of 16SrRNA fragment of genomic DNA produced a single amplicon of 1474bp for the organism. The organism was found to have 95% sequence similarity with Bacillus saffensis strain NBRC 100820 (GenBank Accession Number giI631252747INR 113945.1).This research established tannase production from a novel Bacillus pumilus Strain AJ-2 obtained from deteriorated bambara (Vigna subterranea (L) Verde) nuts.

Keywords: Tannase, Bacillus pumilus, kinetic characterization, purification, 3D structure, homology modelling
- return to Poster LIst -

Poster - A-013
Comparing and Evaluating of the Existing CRISPR Editing Tools: Towards Better Design of the Next-generation CRISPR Editing Algorithms

Presenting Author: Mohamed AlFaki, Neelain University, Sudan
Short Abstract: Clustered regularly interspaced short palindromic repeats (CRISPRs) are important genetic elements in many bacterial and archaeal genomes, which act as a natural defense mechanism against invading phage and plasmid nucleic acids. The CRISPR system has also been engineered to introduce site-specific mutations in a variety of organisms turned the system from a bacterial shield into a gene-editing tool. To ensure target specificity and guide activity of the CRISPR system, researchers depend on intelligent guide RNA design tools to predict guide RNA behavior. Several algorithms have already been developed which use guide RNA sequences as predictors of both on and off-target activity based on sequence composition. Additional tools focus on GC content, homopolymers and other features. Existing online web tools frequently offer one or combine a few design considerations, but rarely aggregate all of these parameters in one place. This forces investigators to spend time comparing across multiple websites in order to guarantee optimal guide RNA design. To address these problems, we explored and compared seven existing and popular CRISPR design algorithms and databases. We used Breast cancer 1 (BRCA1) gene as a constant input or query to present the similarities as well as the areas of significant disagreement and highlight the shortcomings of existing tools. This study may provide guidance on the development of the next-generation CRISPR editing computational tools.
- return to Poster LIst -

Poster - A-015
Sequence variation in the mitochondrial cytochrome oxidase subunit I among Sudanese prostate cancer patients

Presenting Author: Mojahed Alsafi, University of Khartoum, Sudan
Mai Masri, University of Khartoum, Sudan
Lenda Mohamed, University of Khartoum, Sudan
Short Abstract: Background: Transformations in mitochondrial DNA (mtDNA) have been suspected to be part of the carcinogenesis process of prostate cancer. Understanding the potential impact of somatic mutations of mitochondrial genome will facilitate potential therapeutic applications. Approach: In the current study sequencing of partial Cytochrome oxidase subunit I gene (mt-COI) “the central gene for mitochondrial oxidative metabolism and ATP synthesis” was attempted in prostatic tissue biopsies obtained from 22 Sudanese patients undergoing prostatectomy. Sequence variations on gene structure and function on the corresponding protein was predicted using different bioinformatics software packages (Mutpred, SIFT, SNP, GO and PyMOL). Results: The study revealed 30 COI somatic mutations (60% missense, 20% neutral and 20%silent) obtained from two specific tissue types (tumor and benign). The frequency of mutations was distributed almost equally among tumor and benign tumors, however, transition mutations were significantly higher in tumor samples. Two of the identified mutations (A136G and G246A) were reported in several studies to have a role in prostate carcinogenesis. Out of all identified mutations, six mutations were predicted to affect protein function by SIFT, SNP & GO software. 12.5% are suggested to diminish the protein function by substituting amino acids into stop codons. Conclusion: mt-DNA mutations might be an early indicator of malignant transformation in prostate tissue. Mutations in COI affecting its functionality and consequently leading to tumorigenic growth were demonstrated. The predicted impact of selected mutations by different bioinformatics prediction software on protein structure of mt-COI and how they relate to cancer formation was also discussed.
- return to Poster LIst -

Poster - A-017
Gene Expression Profiles of Hodgkin Lymphoma (HL) and Anaplastic Large Cell Lymphoma (ALCL) cell lines

Presenting Author: Nihad Alsayed, Centre for Bioinformatics and Systems Biology, Faculty of Science, University of Khartoum, Sudan
Faisal Fadlelmola, Centre for Bioinformatics and Systems Biology, Faculty of Science, University of Khartoum, Sudan
Kamal Mohamed, Radiation and Isotope Centre of Khartoum, Sudan
Diponkar Banerjee, Division of Anatomical Pathology, Eastern Ontario Regional Laboratory Association and The Ottawa Hospital, Canada
Faisal M Fadlelmola, Centre for Bioinformatics and Systems Biology, Faculty of Science, University of Khartoum, Sudan
Short Abstract: Background: Hodgkin’s lymphoma (HL) and Anaplastic large cell lymphoma (ALCL) are subtypes of human malignant lymphomas. The borderline is not well characterised in the context of morphologic, immunophenotypic, and clinical characteristics between the two diseases, as well as they both express CD30. The use of gene expression microarray in diseases is a genomic technology that improves and provides more accurate diagnosis. Aims: In the current study TM4 suits software was used to analyse gene expression microarray data for four haematopoietic cell lines comprised KMH2 and L428 of human HL-derived cell lines, and DEL and SR786 of ALCL-derived ones. Results: An overall of hundred and twenty one genes were showed to have a significant differential expressed patterns in all cell lines, among which five of them were considered to be a diagnostic biomarker candidate. Furthermore the candidate differentially expressed genes were subjected to pathway and protein-protein interactions analysis. The study showed that three of them were found to be involved in different 83 pathways and the majority were shown to have interactions with many other proteins. Conclusion: Significant findings have showed a resemblance to other symmetrical studies in the literature. In contrast some genes have exhibited different levels of expression than has been reported. On the other hand a part of the findings have not been previously reported in Lymphoma literature but it has been involved in other types of cancer. Further downstream analysis with tissue microarray and protein array is needed for the validation and confirmation of these results.
- return to Poster LIst -

Poster - A-019
Estimating relative fitness of rpoB gene mutants of multidrug-resistant Mycobacterium tuberculosis isolates from Uganda using Computational approches.

Short Abstract: Background
Mutations in the 81-bp region of the rpoB gene in Mycobacteria tuberculosis results in RNA polymerase enzyme of varying capabilities that eventually affects the fitness of the bacteria. This enzyme is responsible for production of primary bacteria RNA transcripts (proteins) hence central for the bacteria survival. Furthermore, this will provide great insights for therapeutic advances & elucidating the evolutionary mechanisms shaping rifampicin resistance in M. tuberculosis.

Multidrug resistant clinical isolates of Mycobacteria tuberculosis will be sub-cultured at the Mycobacteriology (BSL3) Laboratory, College of Health Sciences Makerere University. This will be followed by DNA extraction and subsequently targeted sequencing (MiSeq sequencer) of the rpoB gene of Mycobacterium tuberculosis to identify the mutations.

Analysis plan
Identification of non-synonymous mutations by aligning the rpoB gene onto a reference rpoB gene from H37Rv. Consequently, the relative fitness of the mutants will be determined using Computational approaches (SIFT) that enable the sorting of intolerant from tolerant amino acid substitutions and predicts whether an amino acid substitution in a protein will have a phenotypic effect. SIFT is a multistep procedure that (1) searches for similar sequences, (2) chooses closely related sequences that may share similar function to the query sequence, (3) obtains the alignment of these chosen sequences, and (4) calculates normalized probabilities for all possible substitutions from the alignment. Positions with normalized probabilities less than 0.05 are predicted to be deleterious, those greater than or equal to 0.05 are predicted to be tolerated.
- return to Poster LIst -

Poster - A-021
Adaptive Genomics

Presenting Author: Agostinho Antunes, CIIMAR, University of Porto, Portugal
Short Abstract: Agostinho Antunes

(1) CIIMAR/CIMAR, Interdisciplinary Centre of Marine and Environmental Research, University of Porto, Terminal de Cruzeiros do Porto de Leixões, Av. General Norton de Matos, s/n, 4450–208 Porto, Portugal.
(2) Department of Biology, Faculty of Sciences, University of Porto. Rua do Campo Alegre, 4169-007 Porto, Portugal.

(*) This email address is being protected from spambots. You need JavaScript enabled to view it.

The completion of the human genome sequencing in 2003 opened a new perspective into the importance of whole genome sequencing projects, and currently thousand of species are having their genomes completely sequenced, from simple organisms, such as bacteria, to more complex taxa, such as mammals. This voluminous sequencing data generated across multiple organisms provides also the framework to better understand the genetic makeup of such species and related ones, allowing to explore the genetic changes underlining the evolution of diverse phenotypic and adaptive traits. Here, recent results from our group retrieved from comparative evolutionary genomic analyses of varied species will be considered to exemplify how gene novelty and gene enhancement by positive selection might have been determinant in the success of adaptive radiations into diverse habitats and lifestyles. The findings pinpoint unique molecular products of critical relevance in species evolution, diversification and conservation, but also highlight genomic novelties of importance for environmental and biomedical research.
- return to Poster LIst -

Poster - A-023
Common molecular signature in sickle cell disease patients: Microarray meta analysis and GWAS study

Presenting Author: Cherif Ben Hamda, Institut Pasteur de Tunis, Tunisia
Alia Benkahla, Institut pasteur de Tunis, Tunisia
Raphael Sangeda, Muhimbili University of heath and alied science, Tanzania
Amel Ghouila, Institut pasteur de Tunis, Tunisia
Evarist Msaki, Muhimbili University of heath and alied science, Tanzania
Sharon Cox, Muhimbili University of heath and alied science, Tanzania
Bruno Mmbando, Muhimbili University of heath and alied science, Tanzania
Ozlem Tastan Bishop, Rhode University, South Africa
Nicola Mulder, University of Cape Town, South Africa
Julie Makanie, Muhimbili University of heath and alied science, Tanzania
Kais Ghedira, Institut pasteur de Tunis, Tunisia
Short Abstract: Sickle cell disease(SCD) is a blood monogenetic disorder caused by a point mutation of the beta globin gene leading to multiple severe pathologies. Although the large SCD biopathology are explained essentially by the chronic inflammatory state that take place and a part due to the abnormally altered pro and anti-apoptotic agents. The principles dysregulated actors in these major pathways and their mechanisms still to far to be fully identified and elucidated. Jointly combine gene expression and GWAS data analysis represent a novel strategy appraoch to refine the identification of key mediators and functions in complex disease. We performed a gene expression meta-analysis of five independent publicly available microarray data related to homozygote SCD patients in order to identify a consensus SCD transcriptomic profile and overlapping metabolic pathways. Meta analysis was conducted using MetaDE R package based on combining p-value(maxP) approach, we identified a total of 335 differentially expressed genes(DEGs) (224 overexpressed and 111 underexpressed). TRANSPATH database was used for functional gene set analysis which emphasized the importance and the triggered state of the innate immune system, erythrocyte development, response to stress and hemostasis pathways. Advanced anslysis of GWAS data generated within this study using atSNP R package and SIFT tool identified respectively 80 regulatory single nucleotide polymorphysm(rSNPs) occuring in the promoter of 29 DEGs and a deleterious SNP affecting CAMKK2 protein function. Furhtermore, the present study has generated a novel database of candidate genetic markers, transcription factors and rSNPs associated with SCD for public reuse, which might aid furhter investigation for therapeutic target discovery.
- return to Poster LIst -

Poster - A-025
Improving Proteins Selection Accuracy During Target-Decoy Approach Using Stochastic Models

Presenting Author: Taoufik Bensellak, National School of Applied Sciences of Tangier, Morocco
Short Abstract: Target-decoy approach first goal is estimating false positives and false discoveries in peptide-spectrum matches during proteomic database search. Multiple strategies have been set up such as decoy database construction methods and searching combinations (separate or composite). Many existing works did tackle the effects of decoy construction and concluded that stochastic/statistic methods were more precise than simple sequence reversing or shuffling methods.

In the present study, we implemented (JAVA) new decoy generation methods using stochastic models such as Hidden Markov, Conditional Random Fields (CRFs) and Semi-Supervised Sequence prediction. Implemented methods generate decoys sequences respecting digestion sites (Or not) and can be peptide or protein based. Comparisons were made using ABRF sPRG2006 standard MS/MS data sets and a similarity scoring filter (filter-Xcorr), different search engines were used.

False discovery rate reached smaller values compared to deterministic methods. Accuracy found improved due to two reasons, first for being stochastic approaches and second for taking in count target database initial structure and all amino acids apparition probabilities. The current implementation will serve in the two search stages of an existing open source and JAVA based software (ProteoAnnotator) for proteogenomics annotation.
- return to Poster LIst -

Poster - A-027
Enabling the processing of bioinformatic workflows where data is located through the use of cloud and container technologies

Presenting Author: Eugene De Beste, University of the Western Cape, South Africa
Alan Christoffels University of the Western Cape, South Africa
Antoine Bagula, University of the Western Cape, South Africa
Short Abstract: The use of “big data” to inform biomedical decisions poses complex problems of storage, privacy and data security. This is especially true for fields such as e-health which deal with human health records. Organisations holding such data need to be able to assure regulators and patients of the security of their data storage and handling. In addition, when dealing with large datasets, movement of data for processing can pose a challenge. Many applications that are used to process various types of data have strict software package dependencies, imposing competing requirements on the administrators of institutional computing platforms. Software containers are a lightweight and generally better performing, albeit less diverse alternative to virtual machines. The advancement and increase in adoption of these container technologies have resulted in adaption for use in a variety of scenarios and fields. This allows researchers to replace the shipment of data with shipment of code by packaging their software into containers. Researchers are allowed to define their own toolchains and workflows to do analysis with rather than being limited to what has been allowed by the organisation managing the data set. Utilizing the growing cloud environment ecosystem, with platforms such as OpenStack, it is possible to provide researchers with an easy to use interface to execute custom workflows remotely, without the hassle of software dependency management and direct technical knowledge and reducing the need to send potentially large data sets from one location to another.
- return to Poster LIst -

Poster - A-029
Genetic Dating and Pattern of Admixture in Modern Human Evolution

Presenting Author: Joel Defo, University of Cape Town, South Africa
Nicola Mulder, University of Cape Town, South Africa
Chimusa Rugamika Emile, University of Cape Town, South Africa
Short Abstract: Genetic variation is shaped by admixture between populations in an evolutionary process. The mixture dynamic between groups of populations results in a mosaic of chromosomal segments inherited from multiple ancestral populations. The distribution of ancestral chromosomal segments and the recombination breakpoints in an admixed genome provide information about the time of admixture. Studying populations with particular ancestries has become a major interest in population genetics because of medical and evolutionary impacts of the patterns of single nucleotide polymorphisms. It provides a better understanding of the impact of population migrations and helps us uncover interactions between several populations. Here, we assess various admixture dating tools which estimate the time of admixture between two parental populations. We do so by performing various simulations assuming a particular number of generations and use these to evaluate the tools. We apply the top three assessed methods to some admixed populations from the 1000 Genomes project. We developed a custom script which pertains to estimate the date of admixture in a 3-way admixture scenario accounting for the effect of other admixtures. Despite MALDER shows improvement and produces reasonable date estimates over other current methods, the results from both simulation and real data suggest that dating ancient admixture events accounting for the effect of other admixtures remains a challenge. Our results suggest the need for developing a new approach to date ancient and complex admixture events in multi-way admixed populations.
- return to Poster LIst -

Poster - A-031
A systematic study on the structure and function of Plasmodium falciparum Triosephosphate isomerase as a therapeutic option in antimalarial drug design

Presenting Author: Olanrewaju Durojaye, University of Nigeria, Nigeria
Innocent Okagu, University of Nigeria, Nigeria
Odiba Solomon, University of Nigeria, Nigeria
Samuel Cosmas, University of Nigeria, Nigeria
Robert Igomu, University of Nigeria, Nigeria
Short Abstract: Malaria is a major global public health problem. Plasmodium falciparum happens to be the most virulent among the malarial parasites. The development of drug resistance in Plasmodium falciparum strains has built up a great interest in the search for new antimalarial drugs and drug targets. As part of a program to detect metabolic enzymes as potential drug targets, the 3-dimensional structure and ligand binding sites of Plasmodium falciparum triosephosphate isomerase was determined. The focus on glycolytic enzymes in the malaria parasite results from the observation that in the asexual stage of the parasite in the human red blood cells, the energy requirements of the organism is almost exclusively met by glycolysis. Moreover, significant difference in homology between Plasmodium falciparum TIM and human (Homo sapiens) TIM makes it a suitable candidate for drug therapy.
- return to Poster LIst -

Poster - A-033
Improving the protein identification for a better theragnostics approach

Presenting Author: Sara El jadid, University Ibn Tofail, Kenitra, Morocco
Ahmed Moussa, National School of Applied Sciences of Tangier, Morocco
Short Abstract: Proteomics has made a massive advance and reached enough fullness previously reserved only for genomics. Therefore, it generates a flood of data that present significant challenges for their interpretation. With the emergence of mass spectrometers that allow the study of proteins in a large-scale, the analysis of complex protein samples becomes possible. This proteomic strategy is applied to “theragnostics”, the association of biomarker discovery and drug design. This approach describes the combination of diagnostic tests and therapeutic intervention. Until now, oncology has benefited the most from this strategy, but this fact is likely to change and spread out in all medicine fields as the demand for personalized medicine is constantly increasing. The most fortunate theragnostic study is to be faced to a biomarker and drug target being the same molecular entity.
During the biomarker discovery study, the researcher gets a list of proteins that may be potential targets. However, a part of this list represent false positives.

Here we review the actual bioinformatics tools for the analysis of mass spectrometry data in the context of theragnostics. We focus on aspects of protein identification using MS data, a field that combines four basic approaches: descriptive, stochastic, probabilistic and interpretative. We also discuss the challenge between sensitivity and specificity, as they are the key factor to control the growth of false positive.
- return to Poster LIst -

Poster - A-035
Association of a common rs9939609 variant in the fat mass and obesity-associated (FTO) gene with obesity in Sudanese children

Presenting Author: Alyaa Elbadawi, Centre for Bioinformatics and Systems Biology, Faculty of Science, University of Khartoum, Sudan
Faisal M Fadlelmola, Centre for Bioinformatics and Systems Biology, Faculty of Science, University of Khartoum , Sudan
Short Abstract: Background: Obesity disease develops as a result of interactions between the individual's genetic component and his exposure to environmental risk factors such as lifestyle and food habits. It is well known that the Fat mass and obesity associated gene (FTO) is responsible of the body mass index (BMI) and the risk of obesity. Aims: The study objective was to assess the association of FTO gene rs9939609 polymorphism with BMI in a sample of Sudanese children. Methods: Twenty seven subjects with ages ranging between 7-14 years were sampled from different locations of Khartoum State during the period of January to March 2016. Anthropometric measurements were taken from the subjects and then they were categorized into cases and controls based on their age, gender and BMI. Buccal swab samples were collected and the genomic DNA was extracted using guanidine chloride extraction method. PCR was performed using FTO rs9939609 forward and reverse primers. PCR-RFLP was performed using the Sca1 restriction enzyme. Results: Out of twenty seven samples, eleven participants show the homozygous wild type TT genotype with 41%, five participants show heterozygous TA genotype with 18% and eleven participants show homozygous mutant AA genotype with 41%. The results showed that the polymorphic A allele had the highest risk of obesity with p-value
- return to Poster LIst -

Poster - A-037
Using DIAMOND-MEGAN Pipeline to Explore the Impact of Antibiotics on Gut Microbiome based on High-throughput shotgun Metagenomic Sequencing Data

Presenting Author: Mohamed El-Hadidi, Nile University, Egypt
Daniel Huson, Tübingen University, Germany
Matthias Willmann, Tübingen University, Germany
Benjamin Buchfink, Tübingen University, Germany
Short Abstract: We explored the impact of using Ciprofloxacin antibiotic on the structure of the human gut microbiome using a fast and sensitive pipeline. Our pipeline incorporates two main tools; first, DIAMOND; a high-throughput DNA-to-protein alignment software faster than BLASTX 20,000 times and second, MEGAN Community Edition (CE); an interactive software for taxonomic and functional classification of metagenomic data. We analyzed more than 800 million reads in 12 samples from two subjects (six-time series samples from each subject). Changes in taxonomic profiles on species level before, during and after antibiotic treatment were visualized using PCoA plots based on Bray-Curtis distance. The top five bi-vectors (species) were shown. Interestingly, PCoA plots showed different individual responses to antibiotics uptake, represented by different vector magnitude. One subject had partial microbiome construction after stopping antibiotics, whereas the other subject showed almost full reconstitution. Generally, in both subjects, during the course of antibiotic treatment, taxonomic profiles shifted further away from the baseline, then moved back close to the baseline after stopping antibiotics. Abundance profile tables were exported from MEGAN and R software was used to visualize the microbial community evolution over the course of ciprofloxacin treatment. The phylogenetic dynamics of the five most abundant phyla were visualized using Stream Plots, while the most abundant 26 genera were visualized using Horizon Plots. More details about MEGAN CE were previously published in, and the detailed study of ciprofloxacin treatment was published in
- return to Poster LIst -

Poster - A-039
Comparative Analysis of Corynebacterium pseudotuberculosis Genomes and association with their hosts.

Presenting Author: Mahmoud Elhefnawi, National Research Center, Egypt
Elsayed Hegazy, Nile University, Egypt
Dr Mahmoud Elhefnawi, National Research Center, Egypt
Short Abstract: Background: Using the power of sequencing data analysis. We can do multiple bacterial genome analysis for the same species to enrich our understanding of the variation across hosts. This include detection of presence and absence of specific genes.
Methods: PHAGE detection and identification performed by PHASTER, replication origins detection performed by Ori-Finder, Comparative analysis performed by MetaCYC server, Antibiotic Resistance Gene detection performed by CARD server and Phylogenetic analysis were performed using REALPHY 1.10 an online tool from Swiss Institute of Bioinformatics.

Results and conclusion: The comparative analysis for C. pseudotuberculosis five isolate reveals many information for example genome stability, antibiotic resistance and unique genes and proteins. We found that all isolates share the same ability of antibiotic resistance except horse isolate lack tet38 gene this means host effect is limited to antibiotic resistance. Number of unique genes and proteins in Camel is 1997 this means the Camel as a host may force the C. pseudotuberculosis to adapt and create many variations. GC content is so close for all isolates this means no effect from host regarding GC content and the stability of bacterial genome. Also we found the ability of replication shows that Goat, Buffalo, Horse, Camel and Sheep respectively sorted descending from the high ability to low ability which mean the highest replication ability is Goat isolate. PHAGE region detection we found that the Buffalo shows less infection rates by PHAGE.
- return to Poster LIst -

Poster - A-041
Investigation of Genetic variations involved in cardiovascular diseases and atherosclerosis: Literature review

Presenting Author: Zakaria ELYAZGHI, Institut Pasteur du Maroc / University Mohamed V Rabat, Morocco
Khalid Sadki, University Mohamed V Rabat, Morocco
Loubna El Yazouli, Institut Pasteur du Maroc / University Mohamed V Rabat, Morocco
Fouzia Radouani, Institut Pasteur du Maroc, Morocco
Short Abstract: Atherosclerosis is a condition where the veins become noticeably narrowed and solidified because of an intemperate development of plaque around the artery wall. The disease disrupts the stream of blood around the organism, prompting genuine cardiovascular complexities. The reason for atherosclerosis is multifactorial. In fact, certain components incorporate hypertension, high levels of cholesterol, smoking, obesity, elevated amounts of sugar in the blood and infectious agents such as bacteria. Advances in strategies of molecular genetics have uncovered that genetic ground also influences susceptibility to atherosclerotic vascular diseases.

We led a comprehensive literature search in various databases including PubMed, Embase, Web of Science and Cochrane Library. we considered the experimental evidence that has aggregated in the course of the recent 27 years assessing the relationship between genetic variation with atherosclerosis, such as mutations or common SNPs in human genome.

We gathered and arranged most of candidate genes (APOB-100, PCSK9, ABCA1, APOE, CD36, SCARB1, 5LO, PAI-1, ACE… ), as they have been proved to be associated to modulation of a range of risk factors, impacting plasma lipoprotein levels, inflammation and vascular calcification.

Learning of all these variety in human genome will positively anticipate from an early age the risk to develop premature atherosclerosis, and additionally to encourage the disclosure of novel medication targets. Profound search and assessment of information is in process
- return to Poster LIst -

Poster - A-043
Polypharmacological screening of the drug bank reveals Tigycycline as a new HCV NS5B inhibitor

Presenting Author: Mohamed Fares, National Research Center, Egypt
Mahmoud ElHefnawy, National Research Centre, Egypt
Marc Windisch, Group Head & Project Leader Viral Hepatitis Drug Discovery Institute Pasteur Korea, Korea, Rep
Short Abstract: Polypharmacology and repositioning of certain drugs for a different use other than originally approved purpose has been proposed repeatedly. This process shortens the time required for the development of new drugs for pressing emerging problems by eliminating the time required for studying the pharmacokinetic and pharmacodynamics for already in use drugs. Hepatitis C virus RNA dependent RNA polymerase has been recognized as a potential drug target for direct acting antivirals acting against hepatitis C virus. Here in this study in order to find effective and economical inhibitors for HCV RNA dependent RNA polymerase (ns5b), we implemented structure based virtual screening approach to evaluate the off-label use of drug bank compounds as HCV inhibitor. Evaluation of the antiviral activity of resulted hit Tigecycline in HCVcc & replicon system was performed. Tigecycline shows activity in HCV systems (selectivity index (SI) > 6.5). No 100% inhibition was observed using 100mM compound (Imax 70%). This result suggests more investigation on the potential of Tigecycline usage as HCV inhibitor in combination with other inhibitors like ribavirin and interferon for providing economical treatment protocol for HCV infected patients.
- return to Poster LIst -

Poster - A-045
Environment, health, and sustainable development: Evaluating the impacts of invasive aquatic weeds on the environment of water bodies and the riparian health - by means of remote sensing

Presenting Author: Dalia Farghaly, Leuphana university of Lueneburg, Germany
Emad Elba, Leuphana university of Lueneburg , Germany
Brigitte Urban , Leuphana university of Lueneburg , Germany
Short Abstract: Of all invasive weed species, aquatic weeds are perhaps the most pernicious. The thick weed mats are the reason of tremendous environmental and socio-economic problems , since they disrupt water transport, irrigation systems and hydroelectric schemes, hinder fishing, deteriorate water quality and thereby cause the spread of waterborne diseases. Lake Kyoga which is a shallow depression consisting of number of arms, many of which are filled with swamp vegetation, namely water hyacinth, water lettuce, papyrus and others. These weeds caused the presence of the huge areas of floating mats and islands leaded to tremendous environmental damage in year 1999 where the floating weeds accumulated together in Lake Kyoga outlet and hindered the flow to Nile Kyoga, increased consequently the water levels in lake Kyoga that caused flooding of all villages and communities along the lake shores. The water quality was deteriorated not only due to the dense mats of water hyacinth that reduce oxygen level in water but also due to the mixing of sanitation water from the flooded villages with the lake water which resulted in numerous victims due to the spread of waterborne diseases such as malaria, typhoid, and bilharzias. The riparian societies are very poor, and having only for fishing or agriculture activities without any industrial activities due to lack in the energy resources (no electric network). This lake is wide and inaccessible therefore satellite image are used in studying this area. Landsat images covering the period from 1974 till 2009 and ASTER images for 2008 and 2009 were used.
- return to Poster LIst -

Poster - A-047
Characterizing the Male Urogenital Tract Microbiome using 16S rRNA Gene Analysis

Presenting Author: Kirsty Garson, University of Cape Town, South Africa
Enock Havyarimana, University of Cape Town, South Africa
Katie Lennard, University of Cape Town, South Africa
Heather Jaspan, University of Cape Town, South Africa
Nicola Mulder, University of Cape Town, South Africa
Short Abstract: Dysbiosis of the male urogenital tract microbiome, i.e. the presence of several anaerobic microbes at high abundances, has been linked to an increased likelihood of HIV acquisition. In addition, the relative abundances of these taxa were found to be decreased in participants who had undergone medical male circumcision, at a one-year follow-up.

Males scheduled to undergo medical male circumcision were recruited into a longitudinal study to further elucidate the effects of circumcision. Penile swabs were taken for microbiome analysis immediately prior to circumcision, as well as at 2 weeks, 12 weeks and 24 weeks after circumcision. For the 16S rRNA gene analysis of the penile swabs, de novo OTU clustering was performed at 97% similarity using USEARCH and taxonomic assignment was done using RDP classifier. Downstream analyses were carried out in R using packages such as vegan, phyloseq and metgenomeSeq, to correlate the composition data with available participant data.

Differential abundance testing revealed a post-circumcision community composition that was distinct from baseline, by two weeks post-circumcision. In a principal coordinates analysis, samples clustered by circumcision status, suggesting that circumcision status may be an important determinant of male urogenital tract microbiome composition. An indicator species analysis identified several taxa that characterized pre- and post-circumcision samples, including Lactobacillus species and Peptoniphilus species, respectively. These results correlated well with those generated using a random forest classifier.

Further insight into the male urogenital tract microbiome may enable the development of strategies to reduce susceptibility to HIV and other STIs, by modulating microbiome composition.
- return to Poster LIst -

Poster - A-049
Exploration of the impact of delivery mode and mother socioeconomic status on Tunisian Newborns microbiota

Presenting Author: Mariem Hanachi, Institute Pasteur of Tunis, Tunisia
Alia Benkahla, Institute Pasteur of Tunis, Tunisia
Oussama Souiai, Institute Pasteur of Tunis, Tunisia
Meriem Baouendi, Institute Pasteur of Tunis, Tunisia
Short Abstract: Background:
Microbiota colonisation is a lifetime fluctuant process and is closely related to individual health status. The composition of newborns’ gut microbiota is conditioned by multiple factors including the delivery mode, a subject to controversy. Some studies reported differences between cesarean and vaginally-delivered neonates, suggesting a correlation with the higher risks of being diseased in cesarean delivered infants. However, more recent studies either suggest that gut microbiome is not associated with the mode of delivery or that the differences observed between vaginally delivered versus cesarean newborns disappear within 3 years. The high rate of cesarean practicing in Tunisia makes the investigation of this mode of delivery on newborns’ microbiota a public health priority.

Herein, we propose a comparative analysis of meconium extracted from Tunisian newborns through Shotgun sequencing analysis. We will fill the gap of the previous studies by considering the missing metadata such the mother health record parameters (treatments followed during the pregnancy or her smoking status, or other ‘effective’ environmental factors) and their socioeconomic factors. We will also investigate the effect of delivery mode on the diversity of bacteriophages known to play a key role in shaping gut microbiota.

So far, we have (1) optimized a protocol for sample collection and DNA extraction based on a bibliographic research as well as (2) a quality control step and a downstream bioinformatic pipeline on shotgun public data.
A statistical analysis will be achieved in order to highlight the role of the mentioned factors on bacteria and phages community in newborns gut microbiota.
- return to Poster LIst -

Poster - A-051
Computational approaches for the analysis of protein interactomics

Presenting Author: Sridhar Hariharaputran, National Centre for Biological Sciences. Indian Institute of Science, Bangalore and Bharathidasan University, Tiruchirappalli, India, India
Short Abstract: Interactomics is the study of interactions, the whole set of molecular interaction in cells involving physical interaction and indirect interactions among molecules and genes. Protein interactome can provide more information about the functional links and the functional context of the proteins which are not apparent from protein sequence analysis or protein expression analysis. The significance of this information for systems biology and drug development rises with the increase in the size of protein interaction data and with the availability of protein interactomes. Domain-domain and protein-protein interactions are being analyzed from a biochemical and evolutionary perspective.

Work involves developing computational methods, algorithms to understand protein interactions. Work is centered on the alignment of distantly related domains and the role of outliers (which are the extremely deviant members in a superfamily). Work focuses on the analysis of functional annotation and biological relevance of outliers using Gene Ontology and other resources. On-going work about a map or a network of domain interactions will be introduced for a better understanding of the structural interactome and in predicting the off-site interactions of drug candidates related to Mycobacterium tuberculosis. Also, the work focuses on the analysis of a protein superfamily and domains involved multiple biochemical functions.
- return to Poster LIst -

Poster - A-053
Constructing and interrogating a network of functional host-pathogen protein interactions to improve understanding of HIV and TB co-infection

Presenting Author: Alexa Heekes, University of Cape Town, South Africa
Nicola Mulder University of Cape Town, South Africa
Short Abstract: Tuberculosis (TB) and HIV co-infection is a lethal symbiosis, TB being responsible for almost half of HIV related deaths in sub-Saharan Africa. For individuals infected with Mycobacterium tuberculosis (Mtb), the pathogen that causes TB, HIV infection is the strongest risk factor for developing active pulmonary TB, and reciprocally infection with TB can increase the rate of progression of HIV infection. Host-pathogen protein interaction networks provide an opportunity to study the complexities of the relationship between the host and the pathogen during infection. This analysis has produced a comprehensive and high confidence functional protein-protein interaction network, containing inter- and intra-species interactions between human, HIV and Mtb proteins, as well as drug target interactions. Using the network, we identified 28 human proteins that functionally interact with both pathogens. Compared to other human proteins in the network, the 28 human proteins had significantly higher network centrality measures. In addition, they were enriched for gene ontology terms related to immunological functions and they had shorter path distance to Major Histocompatibility Complex (MHC) proteins, which are known to play an important role in immunity. This suggests that MHC proteins may be playing an important role in facilitating host pathogen interactions during co-infection. None of the 28 human proteins were targeted by existing HIV or TB drugs, however the existing drug target proteins also had significantly higher centrality measures compared to other proteins in the network, suggesting that the 28 human proteins could be potential drug targets.
- return to Poster LIst -

Poster - A-055
VarPhen: Web based tool for genotype-phenotype association.

Presenting Author: Elsayed Hegazy, Nile University, Egypt
Elsayed Hegazy, Nile University, Egypt
Dr Mahmoud Elhefnawi, National Research Center, Egypt
Short Abstract: Summary: Personalized medicine and the highly attention of next generation sequencing increase the demand of turning the genotype data into meaningful phenotype data. VarPhen is a web based tool used to do such thing. It’s written in C# code it’s based on using RefSeq SNPs ID as a genotype to retrieve the relevant phenotype. VarPhen use ClinVar database as the source of clinical information and phenotypes relevant to specific variant.

Poster - A-057
PopulaPy: a workflow and visualization toolkit for Population stratification and admixture

Presenting Author: Andreas Henschel, Khalifa University of Science and Technology, United Arab Emirates
Short Abstract: We here present PopulaPy, an Open Source Python based toolkit to facilitate population structure analyses
while making use of popular tools such as PLINK and ADMIXTURE.

It is common to visualize of multilocus/whole genome SNP genotype datasets in a way that enables inspection of systematic differences in allele frequencies between subpopulations. Techniques include dimensionality reduction methods like Principal Component Analysis (PCA) and Multidimensional Scaling (MDS), resulting in two or three dimensional graphs that reveal such stratifications.

Additional information such as admixture, phenotypic information and even individual chromosome painting can be incorporated in such plots in order to investigate population stratification and admixture simultaneously.

A central tool in PopulacePy is therefore the automated generation of PCA/MDS graphs as scatter-pie charts, where the scatter reflects the differences of individuals along principal components, while pie-charts represent individual represented as pie charts showing individual admixture.
In a similar fashion we also combine Hierarchical clustering and admixture, such that admixture bar charts are arranged in alignment with a dendrogram derived from PLINK's clustering.

In addition, PopulaPy comes with a number of parameterizable workflows that provide a Python interface to popular tools such as PLINK and ADMIXTURE.
- return to Poster LIst -

Poster - A-059
Accelerating Pairwise Local Alignment based on Fragmentation using Particle Swarm Optimization

Presenting Author: Mohamed Issa, University of Zagazig, Egypt
Ahmed Alzohairy, University of Zagazig, Egypt
Short Abstract: Sequence alignment algorithm is used for aligning biological sequences to measure degree of similarity. It used in a lot of applications such as DNA fragment assembly, searching database for similar sequences and Phylogenetic trees construction. Pairwise alignment concerns with measuring alignment between two DNA/protein sequences and there is two kinds, the first is global alignment which concerns for finding the alignment over the entire length of sequences. The second is local alignment which concerns locating the matching regions of similarity between the sequences.
Alignment algorithms based on dynamic programming (DP) technique produce the accurate alignment but consumes execution time proportional with cubic of length of the aligned sequences. Hence, sequences with long length consumes large execution time. This work concerns with accelerating pairwise local alignment based on using the trend of stochastic algorithms.

Stochastic algorithms are used to optimize scientific problem. It based on mimic strategy of nature creatures such as birds, whale such as particle-swarm-optimization (PSO), whale-optimization algorithm or biological process such as chromosomes modifications such as genetic algorithm for reaching optimal solution.

The idea of developed local alignment algorithm based on fragmentation of long lengths and local alignment is used to align all short lengths randomly and the fragments that deliver the maximum similarity degree are kept. PSO is used to keep the search into the region that deliver maximum similarity.
The developed alignment outperform the classical alignment that based on DP. The developed algorithm produce local alignment with success of 97 % with speed up 22.
- return to Poster LIst -

Poster - A-061
Evaluating HIV-1 Compartmentalization between Blood and Male Genital Tract using Deep Sequencing

Presenting Author: Samuel Kariuki, University of Cape Town, South Africa
Philippe Selhorst, University of Cape Town, South Africa
Kevin Rebe, University of Cape Town, South Africa
Kevin Arien, Institute of Tropical Medicine, Belgium
jeffrey Dorfman, University of Cape Town, South Africa
Short Abstract: HIV-1 compartmentalization, i.e. restricted movement of the virus between anatomical sites, is likely to shape sexual transmission and present challenges in designing an effective vaccine. The aim of our study was to use deep sequencing to evaluate compartmentalization of HIV-1 between blood and the male genital tract which has not been shown before. Blood and semen samples were collected from 10 HIV-1 positive, therapy naïve men. Deep sequencing of either the HIV-1 env V3 or C3V4C4V5 was done using the paired-end Miseq Illumina platform. PrimerID approach was used to discern which individual RNA transcript each individual PCR product originated from. An in-house Bioinformatics pipeline was used to process the raw reads. Multiple sequence alignment was done using MAFFT before piping the alignment to FastTree for phylogenetic tree construction. Wright's Measure of Population Subdivision(FST) was used to evaluate compartmentalization. Eight (80%) did not show evidence of compartmentalization with FST ratios ranging from 0.004 to 0.03. One of the sample pair was fully compartmentalized with FST of 0.8 while the other had FST of 0.24 which was interpreted as partial compartmentalization, which could also be seen on the phylogenetic tree. In did, after evaluating compartmentalization using another method, Slatkin-Maddison test (SM) on these participants, confirmed evidence of compartmentalization with minimum migratory events.This data shows evidence of high HIV-1 trafficking between blood and male genital compartment. Understanding the dynamics of HIV-1 populations in individuals and genital compartments is important in designing an effective vaccine and other HIV-1 intervention strategies at the mucosal sites.
- return to Poster LIst -

Poster - A-063
The Genome Centre at the Uganda Virus Research Institute (UVRI) – supporting clinical diagnoses by metagenomic next-generation sequencing.

Presenting Author: Jonathan Kayondo, Uganda Virus Research Institute, Uganda
Hanna Jerome, MRC- University of Glasgow Centre for Virus Research (CVR), University of Glasgow, United Kingdom
John Kayiwa, Uganda Virus Research Institute, Uganda
Timothy Byaruhanga, Uganda Virus Research Institute, Uganda
Annet Nankya, Uganda Virus Research Institute, Uganda
Irene Ataliba, Uganda Virus Research Institute, Uganda
Sreenu Vattipally, MRC- University of Glasgow Centre for Virus Research (CVR), University of Glasgow, United Kingdom
Emma Thompson, MRC- University of Glasgow Centre for Virus Research (CVR), University of Glasgow, United Kingdom
Julius Lutwama, Uganda Virus Research Institute, Uganda
Short Abstract: Background: UVRI hosts a National Reference Laboratory for highly pathogenic viruses. Metagenomic next-generation sequencing (MNGS) has emerged as an additional diagnostic tool for virus detection and discovery. To pilot MNGS at our facility, we analyzed five serum samples from patients with VHF symptoms that showed positive zika virus IgMs and ambiguous or borderline serological test results for yellow fever, west nile and dengue viruses.

Methods: RNA was extracted from plasma or blood (RNAdvance Blood kit, Beckman Coulter), transcribed into double stranded cDNA (SuperScript® III Reverse Transcriptase, Invitrogen, Second Strand cDNA Synthesis, New England Biolabs) and illumina adaptors were ligated to the DNA fragments (Library preparation kit, Kapa Biosystems) for sequencing on the Illumina MiSeq platform. Bioinformatic analysis of the datasets has been done by deNovo assembly using dipSPAdes v3.7.0, the exploration tool anwesh and the mapping software tanoti, which are all specialized tools to pull out short and highly variable viral genomes.

Results: Each sample resulted in 1.1 – 1.3 million reads. 2 – 32 % of the reads corresponded to non-human sequences.
In sample ARB176 > 18,000 reads mapped to a yellow fever virus strain from Kenia (JN620362.1) across 99 % of the genome; ARB169 had 36 reads mapping to hepatitis A virus genome (KP879217.1) at 43 % coverage; ARB248 had 16 reads annotating to dengue virus 3 (JF504679.1) across 17 % of the genome. We did not detect reads that annotate to zika virus in any sample.

Conclusion: UVRI Genome Center shows ability to clear ambiguities with MNGS
- return to Poster LIst -

Poster - A-065
Feature Engineering in Prediction of Human Malaria Resistance from Systems approaches using Deep Learning

Presenting Author: Ali Kishk, H3abionet, University of Zagazig, Egypt
Short Abstract: Small data Systems Biology allow for new insights, not usually revealed by systematic integration of the big data. Feature engineering involves filtering and selecting the most relevant features in Machine Learning and can be applied in any discipline. Genomic data usually contains hundreds of gene measures per sample which makes applying Machine Learning challenging. We highlight results showing how integrating Systems approaches in the analysis of small scale datasets can help with feature selection in Genomic data.

Two Systems approaches were used independently to identify important features that might predict Plasmodium falciparum drug sensitivity from genomic data. In the first approach, drug perturbation signatures from the Library of Network-Based Cellular Signatures (LINCS) database were used to identify genes that respond to perturbation with artemisinin in human cell lines. Phylogenetic Profiling was used to infer a list of orthologous gene from human to malaria. In the second approach, a Pharmacophore search was used to find possible proteins that might bind to artemisinin. Additional genes with high number of mutations in P falciparum in Southeast Asia populations were used.

In total, 237 genes were identified and used as features for constructing a Deep neural network to predict artemisinin sensitivity from microarray gene expression data [1] (0.605 cross-validation R2, 0.458 test R2). Knowledge-based feature engineering might solve the feature per sample challenge in applying Machine Learning in Genomic data.

References : [1] Mok, S., et al. "Population transcriptomics of human malaria parasites reveals the mechanism of artemisinin resistance." Science 347.6220 (2015): 431-435.
- return to Poster LIst -

Poster - A-067
Identification of priority genes for Sanger sequencing in autism spectrum disorders using bioinformatics approaches

Presenting Author: Modibio Kouyate, Bioinformatics, Mali
Modibo Kouyate faculty od Sciences and techniques, Mali
Short Abstract: Autism spectrum disorders (ASD) is characterized by impairment of reciprocal social interaction, verbal and non verbal communication along stereopyted and repetitive behaviors. Genetic predisposition is found in 10-15% of autistic children. To support molecular genetic research on ASD in Mali, we carried out a bioinformatics study in which we compiled all the single nucleotide mutations and copy number variations in ASD related genes from the literature. We used a set of criteria to identify which mutations should be given priority in the context of gene by gene mutation screening in developing countries. The type of mutation and its impact for theoretical protein expression level in the brain and 3-D structure were used among many other criteria. We found 14 gene mutations that we have been using to screen 100 ASD patients and families as well as hundreds of healthy controls.Our overarching goal is to validate a panel of ASD gene mutations the clinicians can screen their patients for in case of high probability of genetic involvement in causation of ASD. This strategy can be applied to many other childhood neuropsychiatric disorders (bipolar disorder, schizophrenia, depression, and substance abuse) in Mali and elsewhere where there is scarcity in qualified human resources for whole exome sequencing and research funding. Our study highlighted the importance of bioinformatics in human genetic research as well.
- return to Poster LIst -

Poster - A-069
Molecular Insights of Peptide Folding Propensities for Cancer Drug Target Improvisation and Anti Microbial Peptide Library

Presenting Author: Praveen Kumar, Bharathiar University, India
Murugesh Easwaran, Bharathiar University, India
Shanmughavel Piramanayagam, Bharathiar University, India
Short Abstract: Peptide plays characteristic role in Drug Discovery, Development and Drug Improvisation purpose. Experimentally, the peptide has different monomeric domains to occupy the structure which is inevitably making a peptide functional. Molecular cause of a peptide secondary structure profile has more than one function when it interacts with target protein or act as a linker or anchored one. Solvent property influence the parameter of a peptide structure eg: Temperature, pH, ratios of solvent volume, multimeric composition in the solvent. Before to design druggable molecule and to study drug likeliness disease, the properties of peptide and functions should be studied. But there is no proper computational annotations for peptide nature and behavior in different solvent. So this study targets to annotate functional monomeric peptides that could target "dis-ease" and "dis-order" profile. To probable short sequence of residues, Microbial Surface proteins were retrieved from Meta-Protein sequence Database. Library of Protein Sequence were subjected to variate differentially in length and functional features. There are 1809 peptide sequence were retrieved and analyzed features of is same. Totally six features were adopted from obtained peptide sequence such as length, residual classification, secondary structure properties, conformational features (co-ordinates), volumes of peptide structure and functional features. Output of this study concentrates on making a peptides confined to make use for Cancer Drug Discovery purpose and Library Construction.
- return to Poster LIst -

Poster - A-071
Comparative genomic analysis of a sub-lineage of multi-drug resistant non-typhoidal Salmonella Typhimurium ST313 that has recently emerged in Blantyre, Malawi

Presenting Author: Benjamin Kumwenda, University of Malawi, Malawi
Rocio Canals Alvarez, Institute of Integrative Biology, University of Liverpool, United Kingdom
Sian Owen, Institute of Integrative Biology, University of Liverpool, United Kingdom
Alistair Darby, University of Liverpool, United Kingdom
Dean Everret, University of Liverpool, United Kingdom
Robert Heyderman, University College London, United Kingdom
Melita Gordon, University of Liverpool, United Kingdom
Chisomo Msefula, University of Malawi, College of Medicine, Malawi
Jay Hinton, University of Liverpool, United Kingdom
Carsten Kroger, Moyne Institute, Ireland
Short Abstract: Salmonella enterica serovar Typhimurium is a major cause of bloodstream infections among immune-compromised individuals such as young children and HIV-positive adults in sub-Saharan Africa. Invasive non-typhoidal Salmonella Typhimurium accounts for 388, 000 deaths annually in Africa. Multi-drug resistant S. Typhimurium ST313 lineage II strains have been isolated from patients in Malawi since 2002. Surveillance data have revealed a new phylogenetic group of strains that emerged in 2006 and belongs to a sub-lineage designated IIa. The lineage IIa non-typhoidal Salmonella isolates are now causing significant levels of bloodstream infections in Malawi, and share the same multi-drug resistance profile as lineage II strains. This work focused on two representative isolates of S. Typhimurium ST313, namely D23580 and D37712. Finished PacBio-generated genomes were used to investigate whether genetic differences account for the emergence of lineage IIa strains in Malawi. The two strains were investigated for SNPs, unique accessory genes, and differences in the plasmid, phage and pseudogene profiles. Comparative genome analysis between strains D23580 and D37712 revealed differences in plasmid profile, gene composition and SNPs. There were 27 SNPs identified, including 14 non-synonymous SNPs in genes responsible for metabolism, energy production, cell membrane and transport. Both D23580 and D37712 strains share a similar prophage profile and carry two identical plasmids pSLT-BT and pBT3. However, the D23580-associated plasmids pBT1 and pBT2 were absent from the lineage IIa strain D37712. The strain D37712 carried two novel plasmids not present in D23580. The new plasmids share high levels of homology with Salmonella and E. coli.
- return to Poster LIst -

Poster - A-073
Blood collection tubes significantly impact expression of activated CD4+ and CD8+ T cells

Presenting Author: Samuel Terkper Ahuno, Kwame Nkrumah University of Science and Technology, Ghana
Alexander Kwarteng Kwame Nkrumah University of Science and Technology, Ghana
Ginikachukwu Uzoekwe, Kwame Nkrumah University of Science and Technology , Ghana
Alexander Kwarteng, Kwame Nkrumah University of Science and Technology, Ghana
Short Abstract: Quantitative and qualitative analysis of T-lymphocyte subsets (CD4 and CD8) via flow cytometry are important for understanding immune response and dysregulation during infections. However, little attention is given to quality control of flow-cytometry. We investigated the impact of time (>1hr and 4hrs) and collection tubes (EDTA and heparin) on the frequency and activation status of CD4+ and CD8+ T cells among healthy Ghanaian individuals. Blood samples were taken from nine (9) healthy individuals into EDTA and heparin tubes followed by PBMC isolation and whole cell staining. In addition, the frequency and activation status of CD4 and CD8 T cells using CD69 action markers were assessed with flow cytometer. Statistical analysis was performed with GraphPad Prism and paired/unpaired t-test or Mann-Whitney–U test were performed to compare the two groups after normality test. P-values of 0.05 or less were considered significant. We observed that the type of collection tube (EDTA or heparin) has no effect on the expression pattern of CD4+ and CD8+ T cells. Comparing activation status of CD4+ and CD8+ T cells in EDTA and heparin tube, a significant difference in activation status was seen in <1hr, however, after 4hrs, we failed to observe significant differences in both T cells. Of note, CD8+ T cell activation frequency was seen to be consistently higher than that of CD4+ T cell at the various study time points and collection tubes. Time lapse and the type of blood collection tubes are key factors to consider in phenotypic characterization of activated immune markers
- return to Poster LIst -

Poster - B-002

Presenting Author: Tawakalt Oduwole, Fountain University, Nigeria
Nusrah Afolabi-Balogun Fountain University, Osogbo, Nigeria
Sholadoye T.T, Ahmadu Bello University Teaching Hospital, Nigeria
Aliyu H.O, Ahmadu Bello University Teaching Hospital, Nigeria
Afolabi-Balogun N.B, Fountain University , Nigeria
Falilat Oseni, Fountain University Osogbo, Nigeria
Short Abstract: Hirschsprung disease (HSCR) is the major cause of chronic constipation in children with a complex pattern of inheritance, sometimes associated with mutations in genes of receptor tyrosine kinase (RET), and endothelin receptor B (EDNRB) signaling pathways, which are crucial for development of enteric nervous system. In this study, samples from seven Nigerian patients with HSCR enrolled in Ahmadu Bello University Teaching Hospital, Shika, Zaria were used for evaluation of the genetic basis of HSCR for the very first time in Nigeria. Analysis involved Six (6) colon samples from previously operated HSCR patients with both short and Long segment aganglionosis and one (1) blood sample from a pre-operation patient. Male (71%) and female (29%), diagnosis age was between 5month - 8years. Genomic DNA (gDNA) isolation from formalin stored colon samples yielded no result using commercial tissue DNA extract kits. However, 134ng/μl gDNA with A260/A280 ratio of 2.132 was isolated from 200ul blood sample. Polymerase chain reaction at 54.50C annealing temperature revealed 648bp and 634bp electrophoresis bands representing three EDNRB-coding variants. Deduced FASTA sequence for EDNRB1 was named ABUTH_TAWAKALT_EDNRB_S1 and ABUTH_DR.TT_EDNRB_S3 respectively, blastn of ABUTH_TAWAKALT_EDNRB_S1 on revealed ABUTH_TAWAKALT_EDNRB_S1 is homologous with nucleotide in database. It has 96% identity, covering 77% Query of 780 max score of a total score of 780 with nucleotide with Accession no. NG11630.2, NM_001122659.2, NM_003991.3 and AY547312.1. Multiple Sequence Alignment of ABUTH_TAWAKALT_EDNRB_S1 on reveal ABUTH_Tawakalt_EDNRB1 consist of frameshift, deletion, insertions as well as substitution most of which are reflected in encoded protein.

Keyword: Hirschsprung’s disease, EDNRB
- return to Poster LIst -

Poster - B-004

Presenting Author: Clement Likhovole, Maseno University, Kenya
Prof. Collins Ouma, Maseno University, Kenya
Dr. John Vulule, Kenya Medical Research Institutr, Kenya
Albert Okumu, Kenya Medical Research Institute, Kenya
Jeremiah Khayumbi, Kenya Medical Research Institute, Kenya
Wilfred Murithi, Kenya Medical Research Institute, Kenya
Susan Musau, Maryland Global Initiatives Corporation, Kenya
Short Abstract: In 2015, 10.4 million people worldwide developed tuberculosis (TB) and 1.4 million died from the disease. In Kenya, in 2014, TB cases with multi-drug resistant (MDR-TB) were 2.2% and 14% in new and retreatment cases, respectively. Chemotherapy with effective anti-tuberculosis drugs is used for the treatment of TB. The spread of mono-resistant and MDR-TB has been enhanced by delays in the identification of resistant strains. The objectives of the current study was to determine the proportion of drug resistant Mycobacterium tuberculosis in sputum isolates. Early morning sputum samples were collected and cultured on Mycobacteria growth indicator tubes (MGIT) and incubated at 37°C. Drug susceptibility testing (DST) was done on Ziehl-Neelsen (ZN) smear positive MGIT tubes and line probe assay (LPA) performed to identify specific mutations on the rpoB, katG and inhA genes. Mutations on discordant samples were confirmed by the BigDye® Terminator v3.1 Cycle Sequencing Kit. ZN microscopy was done on sputum samples with confirmed mutations during second visits to monitor treatment by determining the sputum smear conversion rate. The proportion of MDR-TB, RIF mono-resistant(RMR) TB and INH mono-resistant (INHMR) TB as estimated was as follows: MDR-TB 0.95%, 1.53%;RMR-TB 0.88%, 0.66%; INH mono-resistant TB 1.83%, 1.97%; respectively. Binary logistic regression, indicated that RMR-TB is associated with HIV status (P = 0.025). The smear conversion rate for participating patients was as follows; RMR-TB = 100%, INHMR-TB = 60% and MDR-TB = 67%. Research findings showed that there is a potential association between RMR-TB and HIV infection.
- return to Poster LIst -

Poster - B-006
Faster Growth With Shorter Antigens Explains A VSG Hierarchy During African Trypanosome Infections: A Feint Attack By Parasites

Presenting Author: Dianbo Liu, University of Dundee/Harvard Medical School/MIT/The Broad Institute, United Kingdom
Luca Albergante, Institute Curre , France
David Horn , University of Dundee, United Kingdom
Tim Newman, University of Dundee, United Kingdom
Short Abstract: The parasitic African trypanosome, Trypanosoma brucei, evades the adaptive host immune response by a process of antigenic variation that involves the clonal switching of variant surface glycoproteins (VSGs). The VSGs that periodically come to dominate in vivo display a hierarchy, but how this hierarchy arises is not well-understood. Combining publicly available genetic data with mathematical modelling, we report a VSG-length-dependent hierarchical timing of clonal VSG dominance in a mouse model, revealing an inverse correlation between VSG length and trypanosome growth-rate. Our analysis indicates that, among parasites switching to new VSGs, those expressing shorter VSGs preferentially accumulate to a detectable level that is sufficient to trigger an effective immune response. Subsequent elimination of faster-growing parasites then allows slower parasites with longer VSGs to accumulate. This interaction between the host and parasite is able by itself to explain the temporal distribution of VSGs observed in vivo. Thus, our findings reveal a length-dependent hierarchy that operates during T. brucei infection, representing a “feint attack” diversion tactic utilised during infection by these persistent parasites to out-maneuver the host immune system.
- return to Poster LIst -

Poster - B-008
The population dynamics of haemoglobins F, A2 and S in the context of the haemoglobinopathies HbS and ? thalassaemia among Kenyan children 3-12 months of age

Presenting Author: Alexander Macharia,, Kenya
Thomas Williams, Kemri/Wellcome trust, Kenya
Johnstone Makale, Kemri/Wellcome trust, Kenya
Metrine Tendwa, Kemri/Wellcome trust, Kenya
Short Abstract: While the pattern of haemoglobin switching has been well described in many populations, few studies have been undertaken in Africa where the presence of haemoglobinopathies including sickle cell anaemia (HbSS) and ? thalassaemia can complicate the picture. In the current study, we have used the BioRad Variant HPLC method to document the patterns of production of the common haemoglobin variants HbA, HbA2 and HbS by age and ? thalassaemia genotype among 15,301 children 3-12 months who were recruited to a cohort study from within a defined area on the Kenyan Coast. As expected, we found that HbF% was highest and declined most slowly among HbSS children. Consistent with previous studies we found no significant association between HbF% and gender in children with HbSS but we did find an association in those with HbAA, in whom HbF% was consistently higher in females than in males beyond the 5th month of life. While HbA2% did not vary significantly within the age-range studied, we confirm that HbA2 measurements using the BioRad Variant instrument are unreliable in HbAS and HbSS subjects. Our study provides a rare description of the dynamics of Hb production in a large population of Kenyan children.
- return to Poster LIst -

Poster - B-010
Structural analysis of the functional cavities of penicillin-binding proteins and ß-lactamases for rational antibiotic design

Presenting Author: mame Mbaye, Free University of Brussels and Cheikh Anta Diop University of Dakar, Senegal
Short Abstract: The class of ß-lactam antibiotics has proven to be very efficient in targeting the family of bacterial penicillin binding proteins (PBP), leading to their inactivation and the blocking of the bacterial cell wall synthesis, with a deleterious, often lethal, effect. However, the beneficial effect of these drugs is limited, given that the bacteria have developed resistance mechanisms, the most widespread being due to the bacterial ß-lactamase (ßLACT) enzymes that inactivate ß-lactam antibiotics. Here we perform a structural study of PBPs and ßLACTs that occur in enterobacteria, a family of bacteria that is the cause of many true and opportunistic infections. We focused on the subset of PBPs whose inactivation has a lethal effect, and on class A, C and D ßLACTs which have a PBP-type catalytic mechanism. The comparison of the sequence and structure of the active cavity of ßLACTs and PBPs shows a high spatial resemblance of the functional cavity with an almost perfect similarity of the catalytic site, despite the shuffling of the cavity residues along the sequence. Some interesting differences in the functional cavity appear, further away from the catalytic site. Some of these residues are good candidates to be targeted in designing new ligands with increased affinity and specificity for PBPs, or that specifically inhibit ßLACT proteins. The insights gained by these structural analyses will be exploited to design new antibiotic molecules that are specific to either PBPs or ßLACTs and hence overcome drug resistance.
- return to Poster LIst -

Poster - B-012
Functional host genetic loci associated with paediatric HIV/AIDS disease progression in sub-Saharan Africa

Presenting Author: Gerald Mboowa, Makerere University, Uganda
Graeme Mardon, Baylor College of Medicine, United States
Neil Hanchard, Baylor College of Medicine, United States
David Kateete, Makerere University , Uganda
Moses Joloba, Makerere University , Uganda
Harriet Groom, University of Cambridge, United Kingdom
Effrossyni Gkrania-Klotsas, University of Cambridge, United Kingdom
Gabriel Anabwami, Botswana-Baylor Children's Clinical Centre of Excellence , Botswana
Short Abstract: HIV/AIDS is a leading cause of morbidity and mortality among children in sub-Saharan Africa with majority of them acquiring the virus perinatally from their infected mothers during pregnancy, labor and delivery, or breastfeeding. The goal of this project is to identify critical genetic loci, pathways, and mechanisms important in pediatric AIDS disease progression and developing markers to distinguish HIV/AIDS rapid progressors (RPs) from long-term non-progressors (LTNPs). This would be an important advance in the management of HIV infected individuals, enabling prioritizing of HIV antiretroviral treatment to RPs and minimizing toxic effects of this treatment to pediatric LTNPs. These genetic screening markers are also important in HIV therapeutics & vaccine development. Using exome sequencing data of 314 individuals (including 141 LTNPs & 173 RPs from Botswana & Uganda) & Bioinformatics, we applied single locus association testing for dichotomous case/control phenotype using Fisher's exact test and identified, at a statistically suggestive level, several Single Nucleotide Variants (SNVs) that may be biologically relevant in pediatric AIDS progression. Furthermore, applying gene-based SNV aggregation methods that utilize only exonic nonsynonymous SNVs, we identified four different exonic SNVs in the CCR6 chemokine receptor whose ligand is macrophage inflammatory protein 3 alpha (MIP-3 alpha), regulates the migration and recruitment of dendritic and T cells during inflammatory and immunological responses. A naïve P-value of 0.035 was obtained under a gene-based testing approach. However, the sample size was underpowered making all the above variants both common and rare fail to stand Bonferroni false discovery rate correction for multiple testing
- return to Poster LIst -

Poster - B-014
The Genetic variation between Nilo-Saharan and Niger-Congo sub-Saharan African populations

Presenting Author: Julius Mulindwa, Makerere University, Uganda
Harry Noyes, University of Liverpool, United Kingdom
Enock Matovu, Makerere University, Uganda
Short Abstract: The TrypanoGEN consortium has sequenced the genomes of 298 individuals from six Niger-Congo populations (Uganda, DRC, Zambia, Cameroon, Cote D’Ivoire and Guinea) and a Nilo-Saharan group from Uganda ( Their population structure revealed four clusters which correlated with ethno-linguistic group and geographical latitude, that is, West African Niger-Congo A, Central African Niger Congo, East African Niger-Congo B and the Nilo-Saharan. We observed a spatial distribution of positive natural selection signatures in genes associated with AIDS, Tuberculosis, Malaria and Human African Trypanosomiasis among these TrypanoGEN samples. Having observed a marked difference between the Nilo-Saharan Lugbara and Niger-Congo populations, we identified genes by EHH, XPEHH, Fst (iHS –log p > 3.0, Rsb –log p > 3.0, Fst > 0.1 bonferroni p > 1.8x10e4)], which are highly differentiated between the two ethnic groups and under positive selection in the Lugbara population.
- return to Poster LIst -

Poster - B-016



1 Department of Biochemistry, Jomo Kenyatta University of Agriculture and Technology (JKUAT), P. O. Box 62000 - 00200 Nairobi, Kenya.
2 Department of Biochemistry and Biotechnology, Technical University of Kenya (TUK), P. O. Box 52428 - 00200 Nairobi, Kenya.
3 Department of Evolutionary Neuroet
hology, Max Planck Institute for Chemical Ecology, Hans-Knoell Str. 8, 07745 Jena, Germany.

* Corresponding author:This email address is being protected from spambots. You need JavaScript enabled to view it.

Tsetse flies, the sole vectors of deadly African trypanosomiases, attract little study interest compared to vectors of similar economic importance. Like other insects, tsetses detect chemical stimuli in their environment by sense of smell and taste to locate mates, larviposition sites, and resting sites. Quick detection involves chemosensory receptors and soluble proteins, while stimuli integration in inner brain involve numerous neurotransmitters and their receptors (NTRs) classes including gamma aminobutyric acid, nicotinic acetylcholine, serotonin, dopamine and tyramine/octopamine receptors. of these groups, NTRs remain relatively unknown. We aimed to computationally identify NTRs in published tsetse genomes including Glossina morsitans, G. brevipalpis, G. austeni, G. fuscipes, G. palpalis, and G. pallidipes. Tsetse genomes and transcriptomes were retrieved from vectorbase, while known NTRs homologs in Drosophila melanogaster were retrieved from flybase. Respective NTRs were searched in tsetse sequence-data using stringent blastn and tblastn algorithms. Positive blast hits were retrieved and screened in BLAST-2-GO pipeline before being manually curated in ARTEMIS genome viewer tool. Members of each NTR class identified were assigned unique names and identities in keeping with names found in non-redundant NCBI Protein Database. Respective novel
- return to Poster LIst -

Poster - B-018

Presenting Author: Norah Namatovu, Makerere University College of Health Science, Uganda

Since the 1970’s, the Sanger method is the gold standard for sequencing. However, its low throughput and relative low sensitivity have called for change to new paradigms in Uganda. Next generation- sequencing (NGS) is used for both clinical care and clinical research with benefits of lower costs, increased work flow speeds and enhanced sensitivity in mutation targeted analysis. Its validation in TB related studies in a resource limited setting will provide multiple benefits including; deeper understanding of the genetic basis of diseases and improved identification of putative therapeutic targets. Overall validation of NGS assays will follow the same basic principles that have been established for validating most of the other complex molecular diagnostic procedures. Through quality controls at every stage of the pipeline, we aim to both test and calibrate the accuracy of current NGS data obtained and processed in a resource limited TB lab in Uganda.

AIM: To assess the quality of next generation sequencing data obtained in a resource limited lab and use the results to calibrate and standardize DNA extraction, Library preparation and MiSeq sequencing in Uganda.

Quality scores obtained from FastQC software run on linux platform on the reads will be compared to those of collaborators in a resource rich setting (The University of Georgia, USA).
- return to Poster LIst -

Poster - B-020
Saffold virus contributes to measles-like illness in Uganda

Presenting Author: Prossy Namuwulya, Uganda Virus Research Institute, Uganda
Josephine Bwogi, Uganda Virus Research Institute, Uganda
Pontiano Kaleebu, Uganda Virus Research Institute, Uganda
Henry Bukenya, Uganda Virus Research Institute, Uganda
Brian Kigozi, Uganda Virus Research Institute, Uganda
Jonathan Kayondo, Uganda Virus Research Institute, Uganda
Emma Thomson, Centre for Virus Research, United Kingdom
Shirin Ashraf, Centre for Virus Research, United Kingdom
Marc Niebel, Centre for Virus Research, United Kingdom
Chris Davis, Centre for Virus Research, United Kingdom
Weronika Witkowska, Centre for Virus Research, United Kingdom
Suleman Sabir, Centre for Virus Research, United Kingdom
Jesus Salazar, Uganda Virus Research Institute, Uganda
Short Abstract: Introduction
Measles-like illness (defined as a rash plus fever in the absence of positive serology for measles and rubella) is a common presentation in Ugandan children and has been associated with deaths in this population. -Several infections have been implicated including dengue virus, Parvovirus B19, HHV6 and enteroviruses. We aimed to detect evidence of viral infection in serum samples obtained from children with measles-like illness using metagenomic next generation sequencing (MNGS).

Five samples from patients with measles-like illness (MLI) were available for analysis. RNA extraction, cDNA synthesis, adaptor ligation and sequencing on MiSeq platform were carried out following standard protocols. Mapping and de novo assembly were done to identify viral causes of MLI.
Maximum likelihood phylogenetic analysis was carried out in MEGA 7.0 program using the GTR modeland bootstrap values of 500 replicates.

One sample contained a near full viral genome most closely related to Saffold 3 virus previously isolated from a patient in Pakistan.
Phylogenetic analysis of the full genome is shown in Figure 1.

Figure 1; Molecular Phylogenetic analysis by Maximum Likelihood method

This is the first report of Saffold virus (SAFV) in East Africa. Further work should be done to establish whether SAFV is a common cause of MLI in Uganda and is a candidate cause of severe clinical presentation including death.
 - return to Poster LIst -

Poster - B-022
Comparing Machine Learning and Bayesian Modelling Classification of Genomic Microarray Data

Presenting Author: Richard Newton, Cambridge University, United Kingdom
Lorenz Wernisch, Cambridge University, United Kingdom
Short Abstract: Streptococcus pneumoniae is an important human pathogen comprising more than 90 strains (serotypes). A test for serotypes using a genomic microarray is monitoring the impact of a worlwide vaccination program. We developed a Bayesian model to identify serotypes from raw data from the microarray. With few samples available at the time, a model driven approach was required. Now, however several thousand samples are available, providing an opportunity to investigate serotype classification by machine learning.

We compare the Bayesian model with two machine learning algorithms: Gradient Boosting Machines and Random Forests. Despite the availability of thousands of serotyping arrays, a problem encountered was the lack of training data containing mixtures of serotypes. Most of the available data comprises samples containing only one serotype. To overcome this we implemented an iterative analysis, creating artificial data of serotype mixtures by combining raw data from single serotype arrays.

With this enhanced training set the machine learning algorithms out performed the Bayesian model. However, for rarely occurring serotypes, currently with insufficient training data, the optimal implementation was a combination of results from the Bayesian Model and the Gradient Boosting Machine. As well as being an effective method for classifying biological data, machine learning can be used as an efficient method for revealing subtle biological insights, which we illustrate with an example. The work demonstrates a generic strategy whereby a preliminary probabilistic model may be complemented or replaced by a machine learning classifier once enough training data are available.
- return to Poster LIst -

Poster - B-024
Bioinformatics in sub-Saharan Africa: Is Cameroon lagging behind?

Presenting Author: Jean Claude Njabou Katte, University of Yaounde 1, Cameroon
Short Abstract: Although sub-Saharan Africa is thought to hold an enormous genetic diversity, genomic studies and bioinformatics applications were absent until recently. Much of the progress made in bioinformatics in this region are from South Africa, Kenya, and Nigeria. South Africa started with the establishment of the South African National Bioinformatics Institute and the now defunct National Bioinformatics Network. Scientists in Kenya have also been making progress in developing intellectual capacity for bioinformatics through research and training activities conducted with collaborating institutions. In Nigeria, bioinformatics research groups have been constituted, and bioinformatics techniques have been applied to address some local research questions. Cameroon’s progress in bioinformatics is recent and still in its embryonic stages. There have been a number of short-term bioinformatics training programmes offered in Cameroon to provide foundational skills. The eBioKit Bioinformatics workshop organized in October 2014 at the University of Buea in collaboration with the Swedish University of Agricultural Sciences (SLU). The African Centre of Excellence in Information and Communication Technologies (CETIC) of the University of Yaounde 1 do offer regular shorterm trainings in Bioinformatics. Presently, no Cameroonian university offers bioinformatics as a degree programme. Bioinformatics is taught as one of the courses in the module of comparative genomics in the Public Health Biotechnology professional Master of Science programme in the Department of Biochemistry, Faculty of Sciences of the University of Yaounde 1. Some of the challenges for the progress of bioinformatics in Cameroon are scarce research funding and absent and/or inadequate training opportunities.
- return to Poster LIst -

Poster - B-026
Cerebrospinal fluid markers to distinguish bacterial meningitis from cerebral malaria in children

Presenting Author: James Njunge, KEMRI-Wellcome Trust Research Programme, Kenya
Ian Oyaro, KEMRI-Wellcome Trust Research Programme, Kenya
Nelson Kibinge, KEMRI-Wellcome Trust Research Programme, Kenya
Martin Rono, KEMRI-Wellcome Trust Research Programme, Kenya
Symon Kariuki, KEMRI-Wellcome Trust Research Programme, Kenya
Charles Newton, KEMRI-Wellcome Trust Research Programme, Kenya
James Berkley, KEMRI-Wellcome Trust Research Programme, Kenya
Evelyn Gitau, Alliance for Accelerating Excellence in Science in Africa , Kenya
Short Abstract: Few hospitals in high malaria endemic countries in Africa have the diagnostic capacity for clinically distinguishing acute bacterial meningitis (ABM) from cerebral malaria (CM). As a result, empirical use of antibiotics is necessary. A biochemical marker of ABM would facilitate precise clinical diagnosis and management of these infections and enable rational use of antibiotics. We used label-free protein quantification by mass spectrometry to identify cerebrospinal fluid (CSF) markers that distinguish ABM (n=37) from CM (n=22) in Kenyan children. Fold change (FC) and false discovery rates (FDR) were used to identify differentially expressed proteins. Subsequently, potential biomarkers were assessed for their ability to discriminate between ABM and CM using receiver operating characteristic (ROC) curves. The host CSF proteome response to ABM (Haemophilus influenza and Streptococcus pneumoniae) is significantly different to CM. Fifty two proteins were differentially expressed (FDR<0.01, Log FC≥2), of which 83% (43/52) were upregulated in ABM compared to CM. Myeloperoxidase and lactotransferrin were present in 37 (100%) and 36 (97%) of ABM cases, respectively, but absent in CM (n=22). Area under the ROC curve (AUC), sensitivity, and specificity were assessed for myeloperoxidase (1, 1, and 1; 95% CI, 1-1) and lactotransferrin (0.98, 0.97, and 1; 95% CI, 0.96-1). Myeloperoxidase and lactotransferrin have a high potential to distinguish ABM from CM and thereby improve clinical management. Their validation requires a larger cohort of samples that includes other bacterial aetiologies of ABM.
- return to Poster LIst -

Poster - B-028
Structural insights into the inhibitory mechanism of Mycobacterium tuberculosis cytosine monophosphate kinase: An in silico study

Presenting Author: Anati Nkaule, University of the Western Cape, South Africa
Mohd Shahbaaz, University of the Western Cape, South Africa
Alan Christoffels, University of the Western Cap, South Africa
Short Abstract: Cytosine monophosphate kinase (Rv1712), responsible for Mycobacterium tuberculosis cell wall proliferation is an enzyme which catalyzes the transfer of ATP to cytidine diphosphate (CDP) and deoxycytidine monophosphate (dCDP), then Cytidine monophosphate (CMP) and Deoxycytidine monosphate (dCMP) within the cytosolic environment. Therefore, a comprehensive understanding of protein interactions with the inhibiting molecules is of great interest as it provides opportunities for understanding the Rv1712 function and therapeutic intervention. In the present study, the homology models for Rv1712 were developed using the multiple crystal structures of phosphotransferases (PDB ID - 3R20, 3R8C and 3AKC). The alignment of target and template proteins was performed using JALVIEW software for the identification of conserved regions, showed the accuracy of 75.35%. The generated structures were refined using MOE-DYNAMIC tool and accuracy of the predicted structural elements was assessed using PROCHECK, Verify3D and ERRAT programs. In order to understand the inhibitory mechanism of Rv1712, the molecular docking was performed using Molecular Operating Environment (MOE) with a library of 1290 putative ligands in the predicted active site of Rv1712. The Tacrine-ester (TE) derivative showed highest scores of free energy of binding of -8.26 Kcal/mol and the predicted inhibition constant of 888.54 nM. Furthermore, this docked complex was immersed in the explicit water environments and subjected to 50 ns Molecular Dynamics (MD) simulation using GROMACS package, verifying the interaction preferences of Rv1712-TE complexes. Finally, all the accumulated results verified the biological structure and function of the target protein to be a phosphotransferase, an unstable and acidic protein.
- return to Poster LIst -

Poster - B-030
A genome wide association study to identify gene variants associated with chronic kidney disease among participants of a rural general population cohort

Presenting Author: Rebecca Nsubuga, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Onesmus Kamacooko, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Robert Kalyesubula, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Gershim Asiki, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Lawrence Lubyayi, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Sylvia Kiwuwa-Muyingo, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Robert Newton, MRC/UVRI Uganda Research Unit on AIDS, Uganda
Short Abstract: Introduction: Genome-wide association studies (GWAS) have improved our understanding of the genetic basis of kidney function and disease. Population-based studies, used to investigate traits that define chronic kidney disease (CKD), have identified >50 genomic regions in which common genetic variants associate with estimated glomerular filtration rate (eGFR) or urinary albumin-to-creatinine ratio.

We perform a GWAS to identify gene variants associated to CKD among participants of a rural general population cohort (GPC) in south-western Uganda.

Methods: Within the GPC at MRC/UVRI, we have GWAS, creatinine and sociodemographic data. We perform a quality control (QC) on the GWAS data using Mendelian checks such as family structures, lower MAF, Hardy-Wenberg equilibrium, per SNP missingness and sample quality. We then perform a case-control analysis, by categorising the eGFR, computed from the creatinine levels, into cases and controls with eGFR< 90mls/min deemed cases and test the differences of allele frequencies in the cases/controls using logistic regression.

Results: 1851 (1203 females) individuals had both the GWAS and creatinine data; out of the original 2230258 variants, 320472 passed QC. For the case-control analysis, adjusted for sex, 6 SNPS had p-values < 10-5 with the smallest p-value of 1.72x10-6 associated with SNP at position kgp2159143 (rs759370129); with a MAF of 9.8%.

Conclusion: Our initial analyses have yielded SNPs not considered obvious candidates for association with CKD. We will investigate further if these associations are authentic.
- return to Poster LIst -

Poster - B-032
Genome-wide survey of copy number variation in populations endemic to Human African Trypanosomiasis (HAT)

Presenting Author: Oscar Nyangiri, Makerere University, Uganda
Harry Noyes, University of Liverpool, United Kingdom
Julius Mulindwa, Makerere University, Uganda
Enock Matovu, Makerere University, Uganda
Short Abstract: Background
Copy Number Variants (CNVs) are segments of repeated or deleted DNA sequence > 1000 bp. They constitute 5-10% of the human genome and may contribute to disease susceptibility. Whole genome sequences from five populations residing in HAT endemic areas of Uganda (Nilotic, UGN and Bantu, UGB), Democratic Republic of the Congo (DRC), Ivory Coast (CIV) and Guinea (GUI) (N=233) were analyzed for CNV as part of TrypanoGEN.

We aimed to identify and annotate CNVs relevant to HAT and test if HAT CNVs were disproportionately distributed between HAT cases and controls.

Genomic DNA was sequenced on the Illumina HiSeq platform. CNVs were identified using Genomestrip (Broad institute) and Cn.mops algorithms. We used bedtools to find overlaps with known CNVs in DGV, genes in ensembl and genes in the KEGG pathway during HAT infection. Chi-square test was used to compare the numbers of individuals with CNVs in cases and controls.

Preliminary Results
We detected 13676 CNVs, including unreported CNVs. We identified CNVs in HPR, part of trypanolytic factor 1; genes involved in parasite traversal of the blood brain barrier (LAMININ-4, GNAQ, PLCB-1, PRKCA) and host immune genes (TLR-6 and the MAP-K pathway). Individuals with at least one HPR CNV in the DRC population had disproportionately lower HAT cases compared to individuals without (P=0.03), suggesting the CNV may protect against HAT.

Future work
Tagging SNP haplotypes to copy numbers, control for the effect of population structuring on apparent association with disease and conduct CNV analysis for all H3Africa populations.
- return to Poster LIst -

Poster - B-034
Expression Levels of Odorant Receptor Genes in the Savanna Tsetse Fly, Glossina morsitans morsitans

Presenting Author: Cyrus TAREH, Jomo Kenyatta University of Agriculture and Technology, Kenya
Steven Nyanjom Jomo Kenyatta University of Agriculture and Technology, Kenya
Fred Wamunyokoli, Jomo Kenyatta University of Agriculture and Technology, Kenya
George Obiero, The Technical University - Kenya (TU-K), Kenya
Steven Nyanjom, Jomo Kenyatta University of Agriculture and Technology, Kenya
Short Abstract: Tsetse flies (Genus: Glossina), are the main vectors of African Trypanosomiasis. Olfaction play critical role in tsetse flies behaviour including larviposition, seeking host to fed on and mate for reproduction. Odorant receptors (ORs) are important in insect chemoreception as they bind volatile odorants and transport them to Odorant Receptor Neurons (ORNs) to elicit behavioural response. To better understand Glossina chemoreception, we used real-time quantitative (qPCR) to examine the expression level of ORs in female and male Glossina morsitans morsitans antennae and legs. Results showed that G. m. morsitans ORs codes for transmembrane domain and are involved in odorant binding. The ORs are quite diverse in sequence and the reduced number of tsetse ORs could be linked to its restricted blood feeding diet. The OR genes were highly expressed in antennae than the legs with GmmOR3 and GmmOR45 transcript level being high in the female and male respectively while GmmOR26 and GmmOR20 being high in female and male G. m. morsitans legs respectively. These findings show that expression of OR genes in female and male G. m. morsitans could be conserved in function with the antenna being the main olfactory organ. Our results open avenues to determine functional roles of tsetse ORs which are potential molecular target that can be used to control the vector based on disruption of chemosensory system and response to ligands that can be used to improve traps and baits.

Keywords: Glossina morsitans morsitans, Odorant receptors, Expression profile
- return to Poster LIst -

Poster - B-036
On the trail of a killer: A multilocus sequence typing approach to characterizing Deformed Wing Virus strains

Presenting Author: Faith Obange, University of Nairobi, Kenya
Jandouwe Villinger, International Centre of Insect Physiology and Ecology, Kenya
Michael Lattorff, International Centre of Insect Physiology and Ecology, Kenya
Christine Adhiambo, University of Nairobi, Kenya
Short Abstract: Deformed wing virus (DWV) is the deadliest honeybee virus. Accurate detection of the lethal DWV strain is complicated by its recombination with commonly occurring non-lethal DWV strains; impairing detection of the true pathogen in circulation. Conventional DWV detection involving amplification of a single genomic locus via qPCR could prove inadequate if the locus falls within a recombination site. This necessitates development of a multilocus sequence typing (MLST) approach for detection and identification of potentially lethal DWV strains. This study aimed to develop a high-resolution melting (HRM)-based MLST method for identifying and characterizing different DWV strains and their recombinants within honeybee populations. We sampled five honeybees each from ten hives in each of two apiaries in Karura Forest, Kenya. We extracted viral RNA, reverse transcribed it to cDNA then amplified it by PCR. We performed high-resolution melting (HRM) analysis of PCR amplicons to identify divergent DWV sequences from their sequence-specific HRM profiles. We then sequenced PCR products with unique HRM profiles and performed bioinformatics analyses using Geneious® software for MLST comparisons across loci distributed across the DWV genome. Results indicated the presence of markedly diverse strains of lethal and non-lethal DWV. Therefore, this method increases detection accuracy of DWV and can detect bottlenecks in DWV viral diversity; a phenomenon found to result in emergence of a dominant DWV strain associated with colony collapse. This methodology could serve as an early warning tool to predict colonies at risk of collapse and give beekeepers the opportunity to put in place appropriate preventative measures.
- return to Poster LIst -

Poster - B-038
HPC pipeline for metagenomics

Presenting Author: Trust Odia, Covenant University Bioinformatics Research Group, Nigeria
Ezekiel Adebiyi, Covenant University Bioinformatics Research Group, Nigeria
Short Abstract: NextGenerationSequencing (NGS) has provided an opportunity for metagenomics studies, by making sequencing cheaper and easier. It is now possible to sequence whole genome and exome rRNA. Metagenomics microbial data analysis can be challenging because of the large nature of the data, when several samples are involved . It becomes more complex when several memory inefficient tools are required for down stream analysis. To overcome this, we provide a pipeline/SOP that alleviates the memory demand and uses few tools.

Here, we show how to overcome the above challenge by relabelling sequence identifiers and achieves the following objectives: (1) assign taxonomy (2) compare phylogenetic relationship among sequences (3) predict genes and functions. This SOP focuses on overcoming memory cap issues with USEARCH 32-bit version.

We applied this pipeline to 7 samples sequenced from different marine regions in southern Nigeria. We were able to: provide taxonomy, identified novel micro-organisms/species among our samples and measured the abundance of organisms per sample.
- return to Poster LIst -

Poster - B-040

Presenting Author: Olufemi Ojelabi, National Biotechnology Development Agency., Nigeria
Otunola Adedayo, Babcock University, Nigeria
Moses Efuntoye, Olabisi Onabanjo University, Nigeria
Olatunde Odusan, Olabisi onabanjo University Teaching Hospital, Nigeria
Short Abstract: Background: Antibiotic resistance has become one of the most important clinical challenges today, so also is diabetes mellitus. The immune and metabolic deficiency state in diabetic patients predisposes them to lots of nosocomial and community acquired infections which may remain asymptomatic and the causative organism may already possess antibiotic resistance genes at that level.

Aim: Isolation and Identification of uropathogens from type II diabetes outpatients with significant asymptomatic bacteriuria, antibiotic sensitivity profiling of the uropathogens to cepahalosporins, molecular typing of extended spectrum beta lactamase producing uropathogens and genes conferring resistance.

Methods: isolates from samples with significant bacteriuria were identified and subjected to antibiotic susceptibility test by Kirby-Bauer disk diffusion method. Multidrug resistant isolates were screened for beta lactamase production by the starch iodometric method and double disk synergy test respectively followed by Multiplex PCR for genotyping for ESBL genes.

Result: Out of the 200 patient screened for significant bacteriuria, 29 (14.5%) presented with significant bacteriuria, 2 were symptomatic while 27 were asymptomatic. Prevalent uropathogens isolated were Escherichia coli (44%), Proteus mirabilis (19%) and Klebsiella pneumoniae (15%). 100% resistance was recorded for cefepime and ceftazidime, 41% for ceftriaxone, 37% for cefuroxime and 4% for imipenem. Electrophoresis of PCR products confirmed that the uropathogens carried blaCTX and blaSHV genes.

Conclusion: Cephalosporins being used as front line agents in treating urinary tract infections in Type II diabetic patients are presently being rendered useless by antibiotic resistance genes. Continuous genetic study of antimicrobial resistance in pathogens is crucial for successful antimicrobial therapy
- return to Poster LIst -

Poster - B-042
Phylogenetic Analysis Pol gene of HIV- 2 in Some West Africa Countries

Presenting Author: Elijah Kolawole OLADIPO, Adeleke University, Nigeria
Awoyelu E. H, Ladoke Akintola University of Technology, Nigeria
Afolabi A. Y., Ladoke Akintola University of Technology, Nigeria
Oyawoye O.M., Adeleke University, Ede, Osun State, Nigeria
Ajibade O.A., Adeleke University, Ede, Osun State-, Nigeria
Oluremi A.S., Ladoke Akintola University of Technology, Nigeria
Short Abstract: Background: The presence of infection by human immunodeficiency virus type 2 (HIV-2) in West Africa has been previously documented. However the phylogenetic relationship of the strains that circulate in West Africa countries is of great importance.
Objectives: The present work constitutes the phylogenetic relationship of HIV-2 from selected countries in West Africa.

Methods: The HIV-2 pol gene sequences obtained from NCBI database were analyzed for the construction of a phylogenetic tree with reference sequences of HIV-2 from Nigeria, Togo, Cote d lvoire, Burkina Faso and Ghana.

Results: Analysis of pol gene from selected HIV-2 sequences showed two lineages. The phylogenetic analysis showed that Nigeria and Burkina Faso HIV-2 sequences formed different clusters, while few sequences from Burkina Faso and Cote d lvoire formed a cluster. Ghana and Cote d lvoire alongside Togo and Ghana formed a cluster. Ghana has the most recent isolate of HIV-2 from the evolutionary tree.

Conclusion and Recommendation: The results of the study will reinforce the program on the epidemiological surveillance of the infection in West Africa and make possible further evolutionary studies. This study has also provided epidemiological insights not readily obtained through standard surveillance methods.
- return to Poster LIst -

Poster - B-044
A Comparative Computational Genomics of Different Strains of Ebola Disease Viruses: In-silico Insight for Ebola Control

Presenting Author: Olaitan Awe, University of Ibadan, Nigeria
Oluwagbemi Olugbenga Federal University Lokoja, PMB 1154, Lokoja, Nigeria
Olaitan Awe, University of Ibadan, Nigeria
Short Abstract: The spate of Ebola Virus Disease among West African countries in the year 2014 resulted to fatality cases. After curtailing the disease, few years after, Ebola Virus Disease has re-emerged and it has been confirmed in DRC Congo on the 11th of May, 2017. The disease has killed some inhabitants of a remote village in the Northeastern region of DRC Congo.

The updated versions of the complete sequenced genomes of five (5) different strains of Ebola virus was obtained from the NCBI database. Clustal X (version 2.1) software was used to perform a complete multiple sequence alignment on the five Ebola virus strains. We also performed the phylogenetic analysis. UPGMA clustering algorithm and NJ clustering algorithm were applied during the analyses.

The five Ebola strains have the amino acids (Cys, Val, Ile, Pro, Phe, Tyr, Met, Trp) in common in the large green regions of the alignments; this result also reveals the hydrophobic residue type of regions. The blue regions reveal the positively charged residue type of the regions with amino acids (Lys, Arg). The yellow regions reveal the small non-polar residue type. The phylogeny results revealed that the Bundibugyo and Sudan Ebola virus strains are closely related. Reston and Tai forest Ebola virus strains are also closely related. However, the Zaire Ebola virus strain is the strongest in our Phylogeny results.

Insight gained from these results can form a good foundation for research that will lead to the production of multi-protective and multi-treatment vaccines for different Ebola virus strains.
- return to Poster LIst -

Poster - B-046
Harnessing the potentials of fuzzy logic and voice-computing in the development of anti-malaria drug informatics software

Presenting Author: Olaitan Awe, University of Ibadan, Nigeria
Olugbenga Oluwagbemi Federal University Lokoja, Nigeria
Folakemi Oluwagbemi, Salem University, Nigeria
Olatunji Fagbore, TechPRO Systems and Solutions Limited, Plot 21 A, Molade Okoya Thomas Street, Victoria Island, Lagos, Nigeria
Short Abstract: Malaria is a public health menace consistently inherent in many Sub-Sahara African countries. With respect to medications; there are many issues of concern. These include: wrong diagnosis and wrong dosage administration of anti-malaria drugs on affected patients; Many complications have resulted ranging from severe headaches, stomach and body discomfort, blurred vision, dizziness, hallucinations, and in extreme cases, death. Various informatics software have been developed to support different infectious and communicable disease diagnoses, but not certain of any yet, that have been specifically designed as a voice-enabled application to diagnose and translate malaria patients’ symptomatic data for pre-laboratory screening and correct prescription of correct dosage of the appropriate medication. Malaria voice-enabled computational fuzzy informatics software for correct dosage prescription of anti-malarial drugs (Malavefes, henceforth) was developed using a cocktail of programming languages (Visual Basic.NET. and Java). This work is currently impacting the fields of drug informatics, malaria control, and bioinformatics. It has sensitized and still sensitizing the public to the (i) dangers of self medication of anti-malaria drugs (ii) wrong prescription of anti-malaria drugs by fake pharmacists; (iii) it has elucidated insight into the integration of voice-computing into informatics and bioinformatics tools to promote public health (iv) it has elucidated a new research direction in the development of voice-based/voice-enabled bioinformatics and informatics tools in indigenous African languages for promoting public health (v) it has opened up new opportunities for resource development in bioinformatics, and computational biology especially for malaria researchers. The user-experience and performance evaluation of Malavefes software was impressive
- return to Poster LIst -

Poster - B-048
Characterization of potential drug targeting folate transporter proteins from Eukaryotic Pathogens

Presenting Author: Benson Otarigho, Edo University, Iyamho, Edo State, Nigeria, Nigeria
Short Abstract: Medically important pathogens are responsible for the death of millions every year. There are limited options for therapy and resistance to commonly used drugs is fast emerging. Therefore, there is need to explore an alternative means to control these eukaryotics disease causing protozoan pathogens. The availability of genome sequences of eukaryotic microbes is providing critical biological information for understanding parasite biology and identifying new drug and vaccine targets. We developed automated search strategies in the Eukaryotic Pathogen Database Resources (EuPathDB) to construct a protein list and retrieve protein sequences of folate transporters encoded in the genomes of 200 eukaryotic microbes. The folate transporters were categorized according to features including mitochondrial localization, number of transmembrane helix, and protein sequence relatedness. We identified 234 folate transporter proteins associated with 63 eukaryotic microbes including 48 protozoa, 13 fungi and others belong to algae and bacteria. Phylogenetic tree revealed major and minor clades with 219 and 15 proteins, respectively. All the folate transporter sequences from the malaria parasite, Plasmodium, belonged to the major clade. The identified folate transporters include folate-binding protein YgfZ, folate/pteridine transporter, folate/biopterin transporter, reduced folate carrier family protein and folate/methotrexate transporter FT1. About 60% of these identified proteins have not been known before now. Phylogeny computation shows the similarity of the proteins identified. The findings in this study offer new possibilities for potential drug development targeting folate-salvage proteins in eukaryotic pathogens.
- return to Poster LIst -

Poster - B-050
Characterizing the adult gut metaproteome using de novo sequence tags.

Presenting Author: Matthys Potgieter, University of Cape Town, South Africa
Nicola Mulder University of Cape Town, South Africa
Andrew Nel, University of Cape Town, South Africa
David Tabb, Stellenbosch University, South Africa
Jonathan Blackburn, University of Cape Town, South Africa
Nicola Mulder, University of Cape Town, South Africa
Short Abstract: The characterization of complex metaproteomic samples is a new and challenging field, offering potential novel insights into the pathogenesis of human disease. Here we describe a novel open-source algorithm used to characterize the adult human gut metaproteome using a de novo sequence tag search of the UniProt knowledge base, followed by database compaction and export using normalized spectral abundance factor quantification at the protein and species level. We present a novel data processing pipeline for the identification and correction of sequence tag errors leveraging protein BLAST. Using our pipeline, we were able to identify 4928 protein groups and 23089 peptides from 16 human stool samples that were analysed using a Q-Exactive mass spectrometer. Mean precursor ion intensity was used to quantify taxa identified using UniPept lowest common ancestor analysis, with significant differences in dietary and microbial diversity identified between disease states. The pipeline was validated by searching de novo sequence tags of a set of tryptic digests of human neural cell line proteins against the entire UniProt database, and comparing the performance of the exported sequence database against the human reference proteome using a target-decoy approach. Our results shared 17051 out of 25920 peptides identified using the human reference proteome, and 1897 peptides were only identified using our pipeline. Analysis of variance of posterior error probability score distributions showed a statistically significant difference between non-reference PSMs and decoy hits, indicating potential novel human peptide identifications.
- return to Poster LIst -

Poster - B-052
In Silico disign of an antigenic peptide against leishmaniasis

Presenting Author: Sibiry Samake, UNIVERSITY OF BAMAKO, Mali
Seydou Doumbia, University of Bamako, Mali
Short Abstract: Up to now, many promising vaccine candidates against leishmaniases have been developed and published. An anti leishmania vaccine which does not mimic natural transmission model with sand fly bites may not work efficiently. Sand fly transmits leishmania to human through different salivary proteins with multiple effects on the human immune response. Phlebotomus papatasi salivary protein 15 (PpSP15/SL1) is one of those salivary proteins known to confer protection against leishmania infections in animal models.

We will use bioinformatics tools to collect all the available PpSP15/SL1 sequences from NCBI, find a consensus sequence and evaluate the antigenicity of this molecule.

A new molecule was designed using PpSP15-like/SL1 protein from differentes geographical sandflies species.
This molecule alone or combined with others candidate leishmania multi-epitope peptide like « LEISHDNAVAX », may be a good option to help control leishmaniases diseases. The same molecule may be used as a molecular marker to measure the risk of exposure to sandflies bite.
- return to Poster LIst -

Poster - B-054
Designing new kinase inhibitor derivatives as therapeutics against Mycobacterium tuberculosis: A docking and molecular dynamics simulation study

Presenting Author: Mohd Shahbaaz, University of Western Cape, South Africa
Anati Nkaule, University of Western Cape, South Africa
Alan Christoffels, University of Western Cape, South Africa
Short Abstract: Inorganic polyphosphate (PolyP) plays an essential role in bacterial virulence and drug tolerance. The genome of Mycobacterium tuberculosis encodes for two polyphosphate kinases (Rv2984 and Rv3232c) and polyphosphatases (Rv0496 and Rv1026) for maintenance of intracellular PolyP levels. The mapping of metabolic pathways indicated Rv2984 as an essential drug target involved in the drug resistance of M. tuberculosis. Consequently, a library of 1290 compounds was designed by altering the scaffolds of known inhibitors and were subjected to the virtual screening against Rv2984 using MOE modules. The top three scoring inhibitors were selected and docking scores were validated using Autodock. The Sulfonatate (SN) derivative showed the highest interaction against Rv2984 with predicted free energy of binding of -8.27 kcal/mol and inhibitor constant of 866.29 nM. Similarly, for furano-di-Sulfate (FS) derivative, the parameters were calculated to be -8.99 kcal/mol (255.33 nM), while cytosine-furan-tri-sulfate (CFS) derivative showed the combined scores of -8.56 kcal/mol (528.83 nM). These observations indicated the high binding affinity of the selected inhibitors against Rv2984. Furthermore, these docked complexes were further analyzed using 50 ns Molecular Dynamics (MD) simulations in explicit water conditions. Through the assessment of obtained trajectories, the interactions patterns between the protein and the inhibitors were evaluated using MM/PBSA technique. The total interaction energies of Rv2984 with SN, FS and CFS complexes were calculated to be -100 kJ/mol, -500 kJ/mol and -1000 kJ/mol respectively. In conclusion, the CFS inhibits the activity of Rv2984 more efficiently and the outcomes will be validated using experimental inhibition studies.
- return to Poster LIst -

Poster - B-056
Transcriptomics-based Profiling Identifies Signaling Pathways Associated with Treatment Outcomes in Pancreatic Cancer

Presenting Author: Musalula Sinkala, University of Cape Town, South Africa
Nicola Mulder, University of Cape Town, South Africa
Darren Martin, University of Cape Town, South Africa
Short Abstract: Title: Transcriptomics-based Profiling Identifies Signaling Pathways Associated with Treatment Outcomes in Pancreatic Cancer

Author: Musalula Sinkala, Nikola Mulder, Darren Martin

Affiliation: University of Cape Town - Computational Biology Division

Genome-wide expression measurements allow us to simultaneously determine how many components in a cell change in response to perturbations: this opens prospects to the understanding of systems behaviour and personalised therapy. Employing network analyses, we integrated mRNA expression data with clinical data and multiple databases to reveal relationships between different signalling pathways, subcellular processes and patient treatment outcomes in the context pancreatic adenocarcinoma. We applied unsupervised hierarchical clustering to identify three major tumour clusters—refined using an anomaly detection algorithm and validated by support vector machines. Kaplan-Meier survival analysis revealed different overall survival and disease-free progression periods for the clusters. Further analysis demonstrated that groups with poorer overall survival and progression free survival correspondingly had worse treatment outcomes. Also, Cluster-based differential gene expression profiles were markedly distinctive. Pathway analysis based on Gene Set Enrichment Analysis revealed Gene Ontologies molecular function and biological process overrepresented by enhanced kinase signalling. Lastly, using prior knowledge-based models, we identified signalling via epidermal growth factor receptor and transforming growth factor beta receptor as candidate cellular pathways associated with poor prognosis in pancreatic cancer.
- return to Poster LIst -

Poster - B-058
Pasteur_galaxy: An open and sustainable Galaxy instance for NGS data analysis

Presenting Author: Oussama Souiai, Institut Pasteur de Tunis, Tunisia
Mohamed Alibi, Institut Pasteur de Tunis, Tunisia
Ines Touiri, Institut Pasteur de Tunis, Tunisia
Kais Ghedira, Institut Pasteur de Tunis, Tunisia
Benkahla Alia, Institut Pasteur de Tunis, Tunisia
Short Abstract: Background
The exponential growth of Omics and more specifically the NGS data has raised up a great technical challenge to experimentalists who are unused to bioinformatics skills and do not dispose of sufficient computing power.

Material and Methods
Through, the invaluable financial support provided by H3ABioNet, we settled and managed a 72 core, 512 Gb of RAM and 12T of storage server. To facilitate the access to the server we settled the Pasteur Galaxy server (Pasteur_Galaxy), an open free web-based platform for integrative analysis of NGS data.

Pasteur_Galaxy is based upon Galaxy, the most popular bioinformatics workflow management systems, which is considered as a standard for sharing bioinformatics data, tools and results. As a Galaxy instance, Pasteur_Galaxy aims at providing a large range of bioinformatics tools for the analysis of various types of NGS data. Galaxy supports reproducible computational research by providing an environment for performing and recording bioinformatics analyses.

The Pasteur_Galaxy project has the following main objectives:
1) Provide the academic scientific community an open and sustainable powerful Galaxy instance with a guaranteed availability.
2) NGS Workshops for the scientific community(2 workshops already organized in IPT).
3) Provide the possibility for H3ABioNet nodes and more broadly the H3Africa community to share their data and results.
4) Developing of new tools and services for Galaxy (wrappers and/or toolshed packages).
- return to Poster LIst -

Poster - B-060
Establishing the framework for an African Genome Archive

Presenting Author: Jamie Southgate, University of the Western Cape, South Africa
Alan Christoffels, University of the Western Cape, South Africa
Short Abstract: The generation of biomedical research data on the African continent is growing, with studies realizing the importance of African genetic diversity in discoveries of human origins and disease susceptibility. The archiving of such rich data is a growing concern for researchers (who may be unfamiliar with data governance) placing their trust in 'Western' partners' for the security of their data. There is also a great willingness amongst African researchers to collaborate and promote data discovery whilst maintaining ownership of data. The proposal of an African Genome Archive, enhanced with data curation, federated database access and quality control can enhance data retrieval.
- return to Poster LIst -

Poster - B-062
microbiomeSeq: An R package for analysis of microbial communities in an environmental context.

Presenting Author: Alfred Ssekagiri, Makerere University Immunity and Infection Research Centre of Excellence, Uganda
Short Abstract: Tremendous advancement in high-throughput sequencing has lead to great improvement in studying microbial communities. The major goal is to understand composition and functional diversity. Bioin- formatics pipelines in combination with statistical methods are used to evaluate and analyse community data so that it can be turned into biological insights. microbiomeSeq is built on existing tools to enhance statistical analysis and visualization of 16S ribosomal RNA sequencing data. The main features of the package include alpha diversity which is measured using common diversity indices and compared among experimental conditions by analysis of variance. Beta diversity is explored using multivariate statistics methods such as ordination techniques and permutation analysis of variance. Differentially abundant taxa among conditions are detected using Kruskal-Wallis test and DESeq package, these are assigned importance using random forest classifier. Group based differential analysis uses correlation to group taxa into sets which are then tested for differential expression using distance based score test. Sub communities are detected under specified environmental conditions basing on pairwise co-occurrence correlation between a given pair of taxa. These are assigned topological roles depending on their linkage within respective sub communities and the entire community. Relationships between microbial community and environmental traits by specified correlation coefficients and using fuzzy set ordination to test effects of perturbation in environmental variables to community structure. The package functionality is tested using a dataset which was generated by 16S ribosomal RNA sequencing of various latrines from Tanzania and Vietnam at different depths.
- return to Poster LIst -

Poster - B-064
Identification of Metabolic Pathways Using Atomic Group Tracking: A Constraint Programming Approach

Presenting Author: salihu suberu, University of Ibadan, Nigeria
Chinwe Ekenna, University of Albany, United States
Angela Makolo, University of Ibadan, Nigeria
Segun Fatumo, Wellcome Trust Sanger Institute, United Kingdom
Short Abstract: Identification of Metabolic Pathways Using Atomic Group Tracking: A Constraint Programming Approach

Suberu Salihu, Chinwe Ekenna, Angela Makolo, Segun Fatumo

Abstract:  Pathfinding methods using atom tracking have been widely used to find biochemically relevant pathways. While most of these methods require the users to define the atoms to be tracked which may lead to failing to predict potentially relevant pathways, a recent method AGPathFinder went further into finding alternative pathways by tracking the movement of atomic groups through the metabolic network and using the combined information of reaction thermodynamics and compound similarity to guide the search towards more feasible pathways and better performances. While the AGPathFinder was successful in finding meaningful biochemical pathways, it did not put into consideration the different reaction conditions such as pH and temperature when estimating thermodynamics.

In this work, we propose implementing constraint programming with atomic group tracking in searching for meaning pathways while putting into consideration the different reaction conditions. A methodologically challenging task has been developing a pathfinding method that can be adjudged to be universal by its ability to traverse the metabolic network of different organisms to find metabolic pathways that are biologically meaningful.
We will implement some of the existing atomic tracking algorithms with constraint programming algorithm. We will also implement the atomic group tracking (AGPathFinder) algorithm with constraint programming algorithm. The two results would be tested for time overhead and their ability to find the best pathways that could aid in better drug target identification in different organisms.
- return to Poster LIst -

Poster - B-066
Coupling a climate-based model with an agent based model for understanding and forecasting Rift Valley Fever Epidemics in Ferlo Region (Senegal).

Presenting Author: Paul Tandong, Cheikh Antao diop university, Senegal
Alassane Bah, University Dakar, Senegal
Short Abstract: The dynamics of Rift Valley Fever infection depend on complex interactions between animal hosts, mosquito vectors that are influenced by climatic factors. In this paper, we analyze and model with an agent based formalism the interactions between climate factors, animals, water ponds, Aedes vexans and culex poicilipes vectors and Rift Valley Fever outbreaks in Ferlo region (Senegal), and build a software platform that can help to provide a warning system. A multi-agent platform called CORMAS was used for building the various software agents (hosts, mosquitoes, water ponds) with their behaviors and the climate database containing daily temperature, daily precipitation, and daily humidity. We carried out several simulations of the model and results showed that the occurrence of outbreaks in Ferlo region (Senegal) was strongly influenced by climate during each year. Sensitivity analysis was conducted to estimate the yearly risk of an outbreak as a function of the meteorological variables. Local threshold values of maximal temperature and relative humidity were identified and correlated with animal infections. The outbreak and propagation of Rift Valley Fever in Ferlo were essentially driven by climate factors like daily temperatures and daily precipitation during each year. A software platform that we have developed will enable health authorities to predict the outbreak risk of Rift Valley Fever. The open code source of the model could be used to improve the vector-borne disease management in other world regions.

Poster - B-068
A Computational Biology Integrative Approach for Analysis of DNA-TF Interactions associated with epigentic modification

Presenting Author: Mohammad Tarek, Armed Forces College of Medicine, Egypt
Short Abstract: DNA protein interactions are of great biological interest for their role in regulation of gene expression. In this in-silico study we tried to investigate the key transcription factors responsible for regulation of LC3Av1 which may have a functional impact on gene silencing through epigentic methylation which is associated with tumorgeneis. Position weight matrix was applied in order to predict potential motifs of TRANSFAC database that may be responsible for TFs interacting with the upstream sequence of LC3Av1. GC content was analyzed computationally for methylation at predicted sites , then we applied a machine learning approach using WEKA to extract and rank characterizing features of predicted motifs according to their predicted scores. Bioinformatics modelling techniques for structural analysis were implemented to correlate interaction at structural level to motif sequence where the results suggested a strong correlation for disordered regions and a weak one for ordered structures. trying to investigate the DNA-TF complexes high-throughput DNA shape prediction were applied followed by a docking approach for the most potent TFs predicted trying to reveal the binding affinities and score them according to predicted free energies. the top scoring predicted Tfs were networked using Cytoscape to investigate the PPI networks which were analyzed with the aim to predict which methyl transferase may be responsible for silencing LC3AV1. PPI docking algorithm was performed to virtually investigate binding of methyl transferases DNMT1 and DNMT3a to Top scoring TFs which was also suggested to be interacting both methyl transferases after a literature mining approach.
- return to Poster LIst -

Poster - B-070
Bioinformatics Master Program at the African Center of Excellence (ACE) in Bioinformatics at University of Bamako, Mali

Presenting Author: Mamadou WELE, University of Bamako, Mali
Seydou Doumbia, USTTB, Mali
Mahamadou Diakite, USTTB, Mali
Amadou Kone, USTTB, Mali
Abdoulaye Djimde, USTTB, Mali
Aoua Coulibaly, USTTB, Mali
Cheikna Cisse, USTTB, Mali
Adama D Keita, USTTB, Mali
Christopher Whalen, NIH/NIAID, United States
Darrell Hurt, NIH/NIAID, United States
Michael Tartakovsky, NIH/NIAID, United States
Short Abstract: The African Center of Excellence in Bioinformatics (ACE) in Bamako was created in 2015 through collaboration between USTTB and NIAID to address a significant gap in computing infrastructure and training needed to support bioinformatics research in West Africa. The program receives support from NIH-Welcome Trust initiative for Human Heredity and Health in Africa (H3Africa).

NIAID and University of Cape Town (UCT)’s NIH funded Pan African Bioinformatics network (H3Abionet) have trained local trainers in order to Strengthen local teaching capacity in bioinformatics and research. The faculty are composed by USTTB’s researchers and professors in addition to the international faculty members of H3ABioNet and Scientist from NIAID/NIH.

This Master program is composed of four semesters curricula including course works, practical works at ICER-Mali and other Malian research institutions. The fourth semester is focused on thesis dissertation.

Currently, the first cohort of 9 students has defended their Master thesis on different subjects related to Plasmodium falciparum metabolomics, proteins involved in bacterial resistance, malaria drug development, TB immunogenomics and leishmaniasis vaccine. They come from different background such as biology, medical and pharmaceutical sciences. The second cohort of 13 students was recently recruited in 2017 and some of them received support from Fogarty training grants and Welcome Trust funded DELGEM program (Malian and Gabonese students).
This ACE program is well established and will contribute to the training of the next generation of African scientists capable of using bioinformatics and genomic approaches in order to address key research questions essential on both communicable and chronic diseases in Africa.
- return to Poster LIst -

Poster - B-072
Maternal HIV status, gut microbiome and BCG immunogenicity in South African infants

Presenting Author: Jerome Wendoh Milimu, University of Cape town, South Africa
Bryan Brown, Duke University, United States
Kattie Lennard, University of Cape Town, South Africa
Donald Nyangahu, University of Cape Town, South Africa
William Cameron, Ottawa Hospital Research Institute, Canada
Alashle Abimiku, Institute of Human Virology, Nigeria
Nicola Mulder, University of Cape Town, South Africa
Clive M Gray, University of Cape Town, South Africa
Heather Jaspan, University of Cape Town, South Africa
Short Abstract:
HIV-exposed-uninfected (HEU) infants have higher morbidity and mortality compared to HIV-unexposed (HU) infants. The intestinal microbiome is important for immune system development. HIV-infected patients have an altered gut microbiome. HEU infants and mothers receive antibiotics for Pneumocystis pneumonia prophylaxis. Hence, we hypothesize that HEU infants have an altered gut microbiome which may influence the infants’ immune responses to vaccines.

To describe the intestinal microbiome of HEU infants and relate that to immune responses to childhood vaccines.

HIV-infected mothers with uninfected infants were recruited from Khayelitsha, Cape Town. Stool DNA was extracted by MoBio-Power-Fecal DNA kit, and Illumina sequencing of 16srRNA gene was performed. Data were pre-processed using QIIME and analysed by R. Infant whole blood was incubated with vaccine antigens. Proliferation and cytokine expression by T-cells was measured using multi-parameter flow cytometry.

16SrRNA sequencing was done on 452 stool samples from 210 infants. Birth samples showed significant differences in the microbial profiles between the HEU and HU infants by alpha and beta diversity measures. Furthermore, after the introduction of breastfeeding, these differences between the HEU and HU reduced substantially. Differences in the gut microbiome by feeding mode were more pronounced at later time points. Still, there were no significant differences observed in vaccine responses with regards to HIV exposure.

The impact of maternal HIV status on infant gut microbiome profile potentially begins in utero. The output of this study could potentially inform policy developers in the care of HEU infants.
- return to Poster LIst -

- top -