Accepted Posters
Attention Conference Presenters - please review the Speaker Information Page available here.
If you need assistance please contact submissions@iscb.org and provide your poster title or submission ID.
Category J - 'Pathogen informatics'
Short Abstract: Plants have evolved different defense mechanisms to fight pathogens.According to life styles and infection strategies, pathogens can generally be divided into biotrophs and necrotrophs. Salicylic acid (SA)-dependent signaling pathway acts against biotrophic pathogens, while ethylene (ET) and jasmonic acid (JA) -dependent signaling pathway acts against necrotrophic pathogens. To investigate plant defense response to these two different pathogen types, we use systems biology approach to compare Arabidopsis gene expression after Erysiphe orontii (biotroph) and Botrytis cinerea (necrotroph) infection. Biotrophic- and necrotrophic-specific networks were constructed through integrating gene expression data and Arabidopsis protein-protein interaction network. We found that hub proteins in these two specific networks tended to be defense-responsive proteins. In particular, transcription factors (TFs) and hormone related proteins were enriched in these hub proteins. Module analysis on these two specific networks revealed that 50 functional modules were common to biotrophic and necrotrophic infections. In addition, 221 and 234 functional modules were specific to biotrophic and necrotrophic infections, suggesting that Arabidopsis utilizes different functional modules when response to these two different pathogens. Our preliminary analysis may provide some new hints in deciphering the molecular mechanism of plant immunity.
Short Abstract: Breast cancer has been linked with a number of risk factors including genetic risk factors, family history and others with the underlying mechanisms not fully determined. Viral infections have been a cause of carcinogenesis in various types of cancers including cervical cancer. So far, the studies that aimed to relate a viral infection with breast carcinogenesis have yield conflicting results. In the past, various experimental methods that require the knowledge of the sequence in search (like PCR) were used; however since the rise of next-generation sequencing, a more independent approach for the detection of viral sequences in human samples has been developed. In order to search for expressed viral sequences in human samples, we sequenced the transcriptome of 65 breast cancer samples. A number of different computational tools and methods were applied including the alignment on the reference genome, de novo assembly and a data partitioning bioinformatics tool. We were able to correctly detect viral sequences using a test set, showing the capacity of our methods for detecting viral sequences in human samples. The analysis of our breast cancer samples resulted in the identification of sequences belonging to viruses that have been linked with other types of cancer, however their very small number does not support an association with breast cancer.
Short Abstract: Antibodies are adapter molecules that identify and link pathogens to effector mechanisms for their destruction. Thought to play a critical role in vaccine-mediated protection, antibodies can combat HIV infection by directly binding to and neutralizing HIV viral particles. Alternatively, antibodies can facilitate the destruction of bound viral particles by activating the complement cascade or recruiting innate immune effector cells. This antibody-dependent effector cell response is both highly complex and “tunable” in that many biophysical characteristics influence the immune recruiting capabilities of antibodies. Antibody subclass, antigen-binding affinity, epitope-specificity, and the presence of specific glycan sugar structures all influence interactions that trigger effector cell functions against HIV. Here, we investigate the protective functions of antibodies that stimulate effector cells of the innate immune system. Applying unsupervised and supervised machine learning techniques, we explore the relationship between the biophysical properties of antibodies, potency of antibody-dependent effector cell responses, and clinically defined class of HIV infection. We integrate data from multiple experimental assay types conducted on samples from 200 HIV-infected subjects to create a rich, multidimensional feature set of 881 biophysical binding properties, 79 glycans, and 5 functional activities mediated by diverse effector cells. We show that classification and regression methods applied to these data robustly model antibody characteristics associated with one or more effector cell responses in different classes of HIV-infected subjects. Our results identify biologically meaningful properties of antibodies against HIV and provide insights into the interplay of protective antibody-dependent immune mechanisms that may help guide future HIV vaccine design.
Short Abstract: Objectives: High-risk Human Papillomaviruses (HPVs) cause cervical and head and neck cancers (HNC). HPV integrates into the human genome, an event associated with cancer progression, through an unknown mechanism. We aimed to determine features that make genomic loci prone to hosting HPV integration events. Additionally, HPV integration sites (n=578) were compared to those of 3 other viruses: Human Immunodeficiency Virus (HIV) (n=45,304), Hepatitis B Virus (HBV) (n=370), and Merkel Cell Polyomavirus (MCPyV) (n=35).
Methods: A catalogue of viral integration sites was constructed from the literature and 5 novel HPV integration sites identified by DIPS-PCR in oral cancers. Near each viral integration site, 275 Genomic Features that may affect viral integration were scored, including 8 Categories: genes, expression, fragile sites, open chromatin, histone modifications, protein binding, chromatin segmentation, and repeats. Genomic Feature scores were also calculated for random loci to determine whether features occur more or less frequently than expected near viral integration sites.
Results: 22 Genomic Features at HPV integration sites from cervical cancers and HNCs differed, with histone modifications scoring higher at HNC integrations. HPV integration sites were significantly associated with fragile sites, open chromatin, and activating histone marks, but not with the presence of cancer-related genes. The Genomic Features associated with each virus were largely distinct, although there was significant overlap between HPV and HBV.
Conclusion: We found that HPV integration is associated with multiple Genomic Features in 7 categories, including certain histone modifications. Histone modifications are drug targets that may prevent HPV infection from progressing to cancer.
Short Abstract: Due to recent innovations on sequencing technologies, the use of DNA/RNA sequencing is becoming a more rapid and affordable alternative for diagnosing pathogenic infections within a clinical setting. Our lab previously published PathoScope as a tool for identification of pathogen strain(s) through the isolation, sequencing, and filtering of pathogenic reads from the host. We now provide the framework for advanced functional annotation of identified strains through our modular pipeline known as PathoVar. PathoVar incorporates a user-friendly interface for identification of genomic variants such as Single Nucleotide Polymorphisms (SNPs) and Indels, and applies a set of intelligent defaults for filtering variant calls for quality. PathoVar then annotates these variants using a combination of powerful databases: Identification of translational changes using NCBI annotation information, BLASTing for associated pathways to identify antibiotic signatures existing within the DrugBank and Comprehensive Antibiotic Resistance Database(CARD) databases, and documenting changes in candidate immune response tags through the use of the Immune Epitope Database. We provide our results in both a variant-centric and gene-centric format, as well as in a rich browser application.
This information provides a guided explanation of the nature of both common and rare new strains of any identified organisms existing in the extensive NCBI database. Early testing has replicated previous findings that identify resistance signatures in a drug resistant Klebsiella pneumonia outbreak, and has identified and annotated small features of a Respiratory Syncytial Virus (RSV) strain.
Short Abstract: Bacterial proteins required for virulence, i.e., virulence factors, are a key component of bacterial pathogenicity, as they control and promote pathogenic infection and intracellular survival. Here, we present a combined in silico, in vitro, and in vivo strategy to identify and characterize novel virulence factors of Burkholderia mallei, an infectious intracellular pathogen and the causative agent of glanders. First, we used bioinformatics approaches to identify 49 putative virulent factors involved in B. mallei pathogenicity. Using yeast two-hybrid assays against normalized whole human and whole murine proteome libraries, we identified interactions between each of the putative virulent factors and host proteins. The analysis of these interactions helped us identify and characterize three novel B. mallei virulence factors, as well as host processes and pathways that can be exploited for drug and vaccine design. Finally, using murine aerosol challenge model experiments we verified that three novel virulence factors did indeed attenuate virulence.
Short Abstract: Bacteriocins are peptide-derived molecules produced by bacteria, which function as virulence factors, antibiotics, and signalling molecules. To date, over one thousand bacteriocins have been identified and classified. Recent discoveries have shown that bacteriocins are very diverse and suggest bacteriocins are widely distributed among bacterial species. However, many tools struggle with identifying bacteriocins due to the large sequence and structural diversity of bacteriocins. Bacteriocins are derived from their precursor via a pathway comprising several genes known as context genes. Although bacteriocins themselves are structurally diverse, context genes have been shown to be similar across unrelated species. Our goals are: (1) to identify new candidates for context genes which may clarify how bacteriocins are synthesized, and (2) to identify new candidates for bacteriocins which bear no sequence similarity to known toxins. To achieve these goals, we first find homolog clusters of genes that surround a bacteriocin, and then identify the larger clusters as possible candidates for context genes. In the next step, we use a Naïve Bayes classifier and several sequence homology tools to better train a method to identify bacteriocin-related genes, and look for unidentified bacteriocins near them. We show here initial experimental evidence that the enolase in Streptococcus pyogenes may be involved in streptolysin virulence.
Short Abstract: The increasing availability of sequence data for many viruses provides power to detect regions under unusual evolutionary constraint at a high resolution. Protein-coding regions in viral genomes often contain additional, overlapping functional elements, including regulatory elements, microRNAs, overlapping reading frames, and packaging signals. The synonymous substitution rate can be used as a signature to pinpoint genic regions encoding overlapping functional elements, since sites with synonymous substitutions selectively disfavored are characterized by excess synonymous constraint. We develop a phylogenetic codon model-based framework, FRESCO (Finding Regions of Excess Synonymous COnstraint), to identify regions of excess synonymous constraint in short, deep alignments, such as individual viral genes across many sequenced isolates. We demonstrate the high specificity of our approach on simulated data and apply our framework to the coding regions of ~30,000 open reading frames across 30 viruses species, identifying more than 200 regions of significant excess synonymous constraint. These regions include nearly all previously reported multifunctional elements in well-characterized viruses such as Hepatitis B virus, poliovirus, and West Nile Virus, often identified at a single-codon resolution. In addition, our framework predicts multiple novel functional elements within viral genes, including many novel elements with conserved, stable RNA secondary structure. The precision of FRESCO in identifying known multifunctional sites suggests that it is a useful approach for finding new overlapping functional elements in microbial genomes.
Short Abstract: Influenza A viruses (IAVs) infect a wide range of hosts, e.g. human, avian, swine, equine, canine, and sea mammals, through the binding formed by viral surface glycoprotein hemagglutinin (HA) and certain types of carbohydrate receptors on host cell membranes. Previous studies have shown that alpha 2,3-linked sialic acid (SA2,3Gal) in avian, equine, and canine, alpha 2,6-linked sialic acid (SA2,6Gal) in human, SA2,3Gal and SA2,6Gal in swine, are responsible for the corresponding host tropisms. However, more detailed and refined substructures determining host tropisms are still not clear. Since glycan microarray technique has offered high throughput characterization for glycoprotein-glycan binding affinities, in this study, we applied association rule mining on a set of glycan microarray data of 164 influenza viruses to identify sub-structural relationships among 610 glycans on the array and further characterize host specific substructures. Association rules were obtained from glycan array data for IAVs from five host groups, including human, canine, swine, avian (waterfowl), and avian (terrestrial). The results suggested that human IAV specific glycans share Neu5Aca2-6Galb1-, which is also associated with Neu5Aca2-6GalAcb1-, that swine IAV specific glycans share Neu5Aca2-6Galb substructure, and that canine IAV specific glycans share a unique Neu5Gca2-3Galb1 substructure. Among birds, waterfowl specific glycans are linked by Neu5Aca2-3Galb whereas terrestrial bird specific glycans could have both Neu5Aca2-3Galb and Neu5Aca2-6Galb branches. These observations could be limited by the number of structures on the glycan array, which do not necessarily reflect those in natural hosts. Nevertheless, this study shed some lights on molecular mechanisms for influenza host tropisms.
View Posters By Category
- A) Bioinformatics of Disease and Treatment
- B) Comparative Genomics
- C) Education
- D) Epigenetics
- E) Functional Genomics
- F) Genome Organization and Annotation
- G) Genetic Variation Analysis
- H) Metagenomics
- I) Open Science and Citizen Science
- J) Pathogen informatics
- K) Population Genetics Variation and Evolution
- L) Protein Structure and Function Prediction and Analysis
- M) Proteomics
- N) Sequence Analysis
- O) Systems Biology and Networks
- P) Other
| Search Posters: |