Accepted Posters

Category 'A'- Bioinformatics of Health and Disease'
Poster A01
Examining the proteomic similarity of different groups of bacteria to the human proteome
Brett Trost- University of Saskatchewan
Rolando Pajon (University of Calgary, Microbiology & Infectious Diseases); Anthony Kusalik (University of Saskatchewan, Computer Science);
Short Abstract: This work examines whether the proteomes of pathogenic bacteria have a different level of similarity to the human proteome than nonpathogenic bacteria, and whether bacteria causing chronic infections have a different level of similarity than bacteria causing acute infections.
Long Abstract:Click Here

Poster A02
Comparison of RNA Sequencing (RNA-Seq) Methods for Counting and Testing Differential Expression
Yongsheng Bai - The University of Michigan
Yongsheng Bai (The University of Michigan, Center for Computational Medicine and Bioinformatics); Justin Hassler (The University of Michigan, Department of Biological Chemistry); Randal Kaufman (The University of Michigan, Department of Biological Chemistry); Gilbert Omenn (The University of Michigan, Center for Computational Medicine and Bioinformatics); James Cavalcoli (The University of Michigan, Center for Computational Medicine and Bioinformatics);
Short Abstract: Few methods have been developed to test differential expression from RNA-Seq experiments. Here, we developed statistical methods for testing differentially expressed genes from Illumina technology and ERANGE RNA-Seq data and compared two approaches for their capability of identifying the number of overlapping genes between two mouse cell line experiments.
Long Abstract:Click Here

Poster A03
SIMBioMS: open source system for information management on collaborative projects.
Julio Fernandez Banet- European Bioinformatics Institute
Teemu Perheentupa (University of Helsinki, Institute for Molecular Medicine); Jani Heikkinen (University of Helsinki, Institute for Molecular Medicine); Huei-Yi Shen (University of Helsinki, Institute for Molecular Medicine); Juris Viskna (University of Latvia, Institute of Mathematics and Computer Science); Maria Krestyaninova (European Bioinformatics Institute, Microarray Group); Alvis Brazma (European Bioinformatics Institute, Microarray Group); Mikhail Gostev (European Bioinformatics Institute, Microarray Group);
Short Abstract: SIMBIOMS is an open source modular solution designed to assist high impact collaborative projects with a demand for a flexible data management system which is highly customizable and provides efficient and secure data sharing and storage capabilities that can handle the vast amounts of data generated during a study.
Long Abstract:Click Here

Poster A04
Multi-task hierarchical Bayesian models for predicting outcomes of HIV combination therapies
Jasmina Bogojeska- Max-Planck Institute for Computer Science
Thomas Lengauer (Max-Planck Institute for Computer Science, Computational Biology and Applied Algorithmics);
Short Abstract: We present a method that predicts outcomes of HIV combination therapies by taking the uneven therapy representation in the clinical datasets into account. We consider two adaptations of the multi-task hierarchical Bayesian modeling to the problem at hand that significantly improve the prediction results for therapies with few training samples.
Long Abstract:Click Here

Poster A05
Molecular Modelling and Dynamics of Prolyl Oligopeptidase to Study Conformational Changes and Ligand Binding
Swati Kaushik- National Centre for biological sciences
Ramanathan Sowdhamini (PI,National centre for biological sciences, Computational biology);
Short Abstract: Studies on ligand bound and unbound form of Porcine and Arabidopsis POPs have been carried out using theoretical tools to study conformational dynamics and ligand entry. A possible entry path for ligand has been reported along with the flexibility and energetic implications.
Long Abstract:Click Here

Poster A06
Analysis of Complex human disorders from the systems biology perspective
Sandhya Balasubramanian- The University of Chicago
Dina Sulakhe (Argonne National Lab - The University of Chicago, TCS); Eduardo Berrocal (The University of Chicago, Human Genetics); Conrad Gilliam (The University of Chicago, Human Genetics); Natalia Maltsev (The University of Chicago, Human Genetics);
Short Abstract: We present an approach and a supporting computational platform GEDI for analysis of common heritable disorders. This approach is based on a large-scale integration of experimental data; information from public databases and obtained by text mining. Analysis of large-scale GWAS experiments for patients with autism is used as an example.
Long Abstract:Click Here

Poster A07
Head ‘em off at the pass – Predicting Antigen Evolution
ALBINA RAHIM- University of Saskatchewan
Anthony Kusalik (University of Saskatchewan, Computer Science);
Short Abstract: Neisseria meningitidis is the bacterium that causes meningococcal meningitis in humans. This research aims to predict what the fHbp antigen of this bacterium will mutate into so that vaccines and therapies can be developed for the most likely mutants. As the research is not yet complete, results to date are reported.
Long Abstract:Click Here

Poster A08
A Novel Approach to Predict Cancer Outcomes Based on the Relationship between Protein Structural Information and Protein Networks
Kelvin Zhang- Ontarion Institute for Cancer Research
Francis Ouellette (Ontarion Institute for Cancer Research, Informatics and Biocomputing);
Short Abstract: We developed a novel approach to predict cancer outcomes based on relationships between protein networks and structural information. It
achieves accuracy of 86.8% if tested on a set of breast cancer patients. It is the first study to investigate network disruption caused by domain-domain interactions that might determine patient prognosis.
Long Abstract:Click Here

Poster A09
Development of a computer pipeline for next-generation RNA-Sequencing analysis
SHAOJUN TANG- University of Florida
No additional authors
Short Abstract: We are developing innovative computational tools to investigate global alterations of splicing patterns using RNA-Seq data. In particular, we are interested in accurately detecting alternatively spliced isoforms and their relative expression levels, and we plan to apply our method to transcriptome data from thymus samples in Mbnl1 mutant mice.
Long Abstract:Click Here

Poster A10
The Cancer Genome Atlas (TCGA) Data Coordinating Center and Data Portal
John Greene- SRA International, Inc.
Ari Kahn (SRA International, Inc., Health Research Technology Services/Informatics Systems); Robert Sfeir (SRA International, Inc., Health Research Technology Services/Informatics Systems); Jessica Chen (SRA International, Inc., Health Research Technology Services/Informatics Systems); Jon Whitmore (SRA International, Inc., Health Research Technology Services/Informatics Systems); Shelley Alonso (SRA International, Inc., Health Research Technology Services/Informatics Systems); David Nassau (SRA International, Inc., Health Research Technology Services/Informatics Systems); Namrata Rane (SRA International, Inc., Health Research Technology Services/Informatics Systems); Dominique Berton (SRA International, Inc., Health Research Technology Services/Informatics Systems); Carl Schaefer (NCI , Center for Biomedical Informatics and Information Technology);
Short Abstract: The Cancer Genome Atlas (TCGA) is a coordinated NIH effort to understand of the genetics of cancer, using genome analysis technologies. The center of TCGA informatics efforts is the Data Coordinating Center (DCC) and Portal, where information from samples of 20 human cancers are managed and entered into public databases.
Long Abstract:Click Here

Poster A11
De novo assembly of Corynebacterium pseudotuberculosis genomes using short reads obtained from mate-paired libraries
Louise Cerdeira- UFPA
Adriana Ribeiro (UFPA, Genetic); Rommel Ramos (UFPA, Genetic); Vivian D'Afonseca (UFPA, Genetic); Jerônimo Ruiz (Fiocruz/CPQRR, Parasitology); Vasco Azevedo (UFMG, Genetic); Maria Paula Schneider (UFPA, Genetic); Artur Silva (UFPA, Genetic);
Short Abstract: Corynebacterium Pseudotuberculosis was a pathogen the causative agent of Caseous LymphAdenitis disease, was veterinary importance. Three of these genomes was sequenced using SOLiD (Sequencing by Oligonucleotide Ligation and Detection) mate-paired libraries. The genomes assembly of short reads was performed using de novo approach.
Long Abstract:Click Here

Poster A12
High Frequency of PfCRT 76T in Two Malian Villages and Its Prevalence in Severe Relative to Non-Severe Malaria
Mamadou WELE- Malaria Research and Training Center, Department of Epidemiology of Parasitic Diseases, Faculty of Medicine, Pharmacy and Dentistry, University of Bamako
Abdoulaye A. Djimdé (Malaria Research and Training Center, university of Bamako); Aldiouma Guindo (Malaria Research and Training Center, university of Bamako); Abdoul habib beavogui (Malaria Research and Training Center, university of Bamako); Boubacar Sadou (Malaria Research and Training Center, university of Bamako); Zoumana Traore (Malaria Research and Training Center, university of Bamako); Ogobara Doumbo (Malaria Research and Training Center, university of Bamako);
Short Abstract: PfCRT 76T mutation was investigated in severe and non-severe malaria in Mali in case a control study. The mutation was present in 60.8% (n = 386) non-severe malaria cases and in 77.2% (n = 193) severe malaria cases (p <0.0001). PfCRT 76T was associated with malaria severity in this setting.
Long Abstract:Click Here

Poster A13
Consensus Measures for Genome-wide Association Studies – the PhenX Project
Helen Pan- RTI International
Jayashri Mehta (National Library of Medicine/National Institutes of Health, National Center for Biotechnology Information); Kimberly Tryka (National Library of Medicine/National Institutes of Health, National Center for Biotechnology Information); Isabel Fortier (University of Montreal Hospital Center , Research Center); Erin Ramos (National Institutes of Health, National Human Genome Research Institute ); Heather Junkins (National Institutes of Health, National Human Genome Research Institute ); Carol Hamilton (RTI International, Research Computing Division);
Short Abstract: Genome-wide association studies (GWAS) provide promise for the identification of genomic markers associated with complex disease, but require large sample sizes and replication studies. PhenX measures provide a common currency for collecting data, thereby facilitating cross-study analysis and increasing statistical power for identification of associations between genotypes, phenotypes, and exposures.
Long Abstract:Click Here

Poster A14
Bioinformatic and Comparative Analysis of Tumor Cardiolipin Remodeling Mechanisms Using Lipidomic Data
Lu Zhang- Boston College
Robert Bell (University of California San Francisco, Biomedical Sciences); Michael Kiebish (Washington University School of Medicine St. Louis, Internal Medicine); Thomas Seyfried (Boston College, Biology); Xianlin Han (Washington University School of Medicine St. Louis, Internal Medicine); Jeffrey Chuang (Boston College, Biology);
Short Abstract: We present novel computational approaches to mechanistically and functionally analyze high-throughput lipidomic data, focusing on cardiolipin. We built an equilibrium model to analyze tumorigenic and healthy lipidomic samples, whose parameters we optimized using machine learning. Cross-validations robustly predicted cardiolipin distributions, and provided evidence of abnormal lipid physiology in tumors.
Long Abstract:Click Here

Poster A15
Disease Portals at the Rat Genome Database: A Platform for Genetic and Genomic Research
Rene Lopez- Medical College of Wisconsin
Liz Worthey (Medical College of Wisconsin, Human and Molecular Genetics Center); Mary Shimoyama (Medical College of Wisconsin, Human and Molecular Genetics Center); Jennifer Smith (Medical College of Wisconsin, Human and Molecular Genetics Center); Rajni Nigam (Medical College of Wisconsin, Human and Molecular Genetics Center); Victoria Petri (Medical College of Wisconsin, Human and Molecular Genetics Center); Stan Laulederkind (Medical College of Wisconsin, Human and Molecular Genetics Center); Tim Lowry (Medical College of Wisconsin, Human and Molecular Genetics Center); Tom Hayman (Medical College of Wisconsin, Human and Molecular Genetics Center); Shur-Jen Wang (Medical College of Wisconsin, Human and Molecular Genetics Center); Jeff De Pons (Medical College of Wisconsin, Human and Molecular Genetics Center); Pushkala Jayaraman (Medical College of Wisconsin, Human and Molecular Genetics Center); Marek Tutaj (Medical College of Wisconsin, Human and Molecular Genetics Center); Weisong Liu (Medical College of Wisconsin, Human and Molecular Genetics Center); Diane Munzenmaier (Medical College of Wisconsin, Human and Molecular Genetics Center); Melinda Dwinell (Medical College of Wisconsin, Human and Molecular Genetics Center); Simon Twigger (Medical College of Wisconsin, Human and Molecular Genetics Center); Howard Jacob (Medical College of Wisconsin, Human and Molecular Genetics Center);
Short Abstract: Disease Portals at the Rat Genome Database provide a comprehensive platform for physiological genomics discovery through the integration of heterogeneous datasets into the context of the genome using multiple ontologies and sophisticated data mining and visualization tools
Long Abstract:Click Here

Poster A16
Automatic detection of metastatic cancer cells in the blood
Jesse Rodriguez- Stanford
Ashley Powell (Stanford, Surgical Oncology); Stefanie Jeffrey (Stanford, Surgical Oncology); Serafim Batzoglou (Stanford, Computer Science); David Paik (Stanford, Biomedical Informatics);
Short Abstract: Tumors shed cells into the blood to metastasize. These cells can be extracted with magnetic microbeads to study metastasis and improve prognostic accuracy. We present an automated computational method for identifying cells in a blood sample extracted by magnetic microbeads to facilitate a clinical device for microbead cell extraction.
Long Abstract:Click Here

Poster A17
cell.line.plot - Visualizing gene expression profiles of cell model systems
Vidal Fey- VTT Technical Research Centre of Finland
Rebecca Ceder (VTT Technical Research Centre of Finland, Medical Biotechnology); Roland Grafström (VTT Technical Research Centre of Finland, Medical Biotechnology); Merja Perälä (VTT Technical Research Centre of Finland, Medical Biotechnology); Olli Kallioniemi (University of Helsinki, Institute for Molecular Medicine Finland (FIMM));
Short Abstract: Cell.line.plot is a bioinformatics application for graphical exploration of gene expression profiling data in the context of alternative testing methods to facilitate the use of cell line models and high-throughput technologies. The outputs are conclusive multi-line plots depicting similarities of gene expression patterns over several cell lines or clinical samples.
Long Abstract:Click Here

Poster A18
pocketSNP: database of disease-related nsSNP on structural pocket
Byungwook Lee- KRIBB
Jin Ok Yang (KRIBB, KOBIC);
Short Abstract: None On File
Long Abstract:Click Here

Poster A19
PATHWAY RESOURCES AT THE RAT GENOME DATABASE: A DYNAMIC PLATFORM FOR INTEGRATING GENE, PATHWAY AND DISEASE INFORMATION
Weisong Liu- Medical College of Wisconsin
Mary Shimoyama (Medical College of Wisconsin, Human and Molecular Genetics Center); Jennifer Smith (Medical College of Wisconsin, Human and Molecular Genetics Center); Rajni Nigam (Medical College of Wisconsin, Human and Molecular Genetics Center); Victoria Petri (Medical College of Wisconsin, Human and Molecular Genetics Center); Stan Laulederkind (Medical College of Wisconsin, Human and Molecular Genetics Center); Tim Lowry (Medical College of Wisconsin, Human and Molecular Genetics Center); Tom Hayman (Medical College of Wisconsin, Human and Molecular Genetics Center); Shur-Jen Wang (Medical College of Wisconsin, Human and Molecular Genetics Center); Pushkala Jayaraman (Medical College of Wisconsin, Human and Molecular Genetics Center); Marek Tutaj (Medical College of Wisconsin, Human and Molecular Genetics Center); Diane Munzenmaier (Medical College of Wisconsin, Human and Molecular Genetics Center); Melinda Dwinell (Medical College of Wisconsin, Human and Molecular Genetics Center); Simon Twigger (Medical College of Wisconsin, Human and Molecular Genetics Center); Howard Jacob (Medical College of Wisconsin, Human and Molecular Genetics Center); Elizabeth Worthey (Medical College of Wisconsin, Human and Molecular Genetics Center); Jeff De Pons (Medical College of Wisconsin, Human and Molecular Genetics Center);
Short Abstract: This poster introduces pathway resources developed at RGD, such as the Pathway Ontology, pathway annotations and diagrams, web-based tools for pathway searching, browsing and visualization. Every pathway page provides links to related information available at RGD or other data sources. Users can also 'travel' through the pathway landscape.
Long Abstract:Click Here

Poster A20
Characterization of genomic structural variations in a neuroblastoma cell line using mate-pair sequencing
Valentina Boeva- Institut Curie
Bruno Zeitouni (Institut Curie, Inserm U900); Stéphanie Jouannet (Institut Curie, Inserm U830); Alex Cazes (Institut Curie, Inserm U830); Gudrun Schleiermacher (Institut Curie, Inserm U830); Emmanuel Barillot (Institut Curie, Inserm U900); Olivier Delattre (Institut Curie, Inserm U830); Isabelle Janoueix-Lerosey (Institut Curie, Inserm U830);
Short Abstract: Genome resequencing using mate-pair reads provides a more exhaustive and precise characterization of somatically acquired rearrangements in tumor cells as compared to conventional strategies. We applied this technique to sequence a neuroblastoma cell line. We developed an approach to identify somatic structural rearrangements and found genes producing broken or chimeric transcripts.
Long Abstract:Click Here

Poster A21
Markov Chain Monte Carlo Computational Analysis of Chromosome Conformation Capture Carbon Copy data
Mathieu Rousseau- McGill University
James Fraser (McGill University, Biochemistry); Josée Dostie (McGill University, Biochemistry); Mathieu Blanchette (McGill University, Centre for Bioinformatics);
Short Abstract: A Markov chain Monte Carlo approach to model and analyze three-dimensional chromatin structure from Chromosome Conformation Capture Carbon Copy (5C) interaction frequency data. Structure sample analysis for structure subfamilies and reliable substructures performed through hierarchical clustering and minimum-weight clique finding reveals a correlation between chromatin structure and gene expression.
Long Abstract:Click Here

Poster A22
Computational prediction of compound sensitivity with genomic signatures
Kavitha Venkatesan- Novartis Institutes for Biomedical Research
Michael Morrissey (Novartis Institutes for Biomedical Research, Oncology); Nicolas Stransky (Broad Institute, Computational Biology); Pichai Raman (Novartis Institutes for Biomedical Research, Developmental and Molecular Pathways); Dmitriy Sonkin (Novartis Institutes for Biomedical Research, Oncology); Michael Jones (Novartis Institutes for Biomedical Research, Oncology); Adam Margolin (Broad Institute, Computational Biology); Joseph Lehar (Novartis Institutes for Biomedical Research, Oncology); Christopher Wilson (Novartis Institutes for Biomedical Research, Developmental and Molecular Pathways); Sungjoon Kim (Novartis Institutes for Biomedical Research, Genomics Institute of the Novartis Foundation); Markus Warmuth (Novartis Institutes for Biomedical Research, Oncology); William Sellers (Novartis Institutes for Biomedical Research, Oncology); Jordi Barretina (Broad Institute, ); Giordano Caponigro (Novartis Institutes for Biomedical Research, Oncology); Levi Garraway (Broad Institute, Cancer Program); Robert Schlegel (Novartis Institutes for Biomedical Research, Oncology);
Short Abstract: Using data from an encyclopedia of ~1000 cancer cell lines profiled for genome-scale mRNA expression, gene copy number alteration, mutations and compound inhibitory effects, we have developed integrated computational models that can predict compound sensitivity and the associated predictive genetic features. Ultimately, this may aid cancer patient stratification.
Long Abstract:Click Here

Poster A23
Adding structural information to the von Hippel-Lindau (VHL) tumor suppressor interaction network
Emanuela Leonardi- University of Padua
Silvio Tosatto (University of Padua, Biology); Alessandra Murgia (University of Padua, Pediatrics);
Short Abstract: The von Hippel-Lindau (VHL) tumor suppressor gene is a protein interaction hub, controlling numerous genes implicated in tumor progression. Using structural information and computational analysis we have located three distinct interaction interfaces. We distinguish compatible and exclusive interactions by relating pVHL function to interaction interfaces and subcellular localization.
Long Abstract:Click Here

Poster A24
Evaluation of computational miRNA target predictions in human
Olivier Gevaert- Stanford University
No additional authors
Short Abstract: microRNAs are short RNAs that regulate expression through binding to the target mRNAs. An important shortcoming in current microRNA research is the lack of experimentally verified targets. In this contribution we report on the results when comparing the target predictions of seven often-used target prediction tools for human miRNAs.
Long Abstract:Click Here

Poster A25
Effector prediction in host-pathogen interaction based on a Markov model of a ubiquitous EPIYA motif
Chao Zhang- University of Missouri
Shunfu Xu (the First Affiliated Hospital of Nanjing Medical University, Department of Gastroenterology); Yi Miao (the First Affiliated Hospital of Nanjing Medical University, Department of General Surgery); Jianjiong Gao (University of Missouri, Department of Computer Science); Dong Xu (University of Missouri, Department of Computer Science);
Short Abstract: A hidden Markov model of five amino acids was built for the EPIYA-motif to search the non-redundant protein. Then by combining the EPIYA-motif and the adjacent SH2 binding motifs to built others HMMs with nine amino acids to predict many potential effectors for pathogens in bacteria and protista.
Long Abstract:Click Here

Poster A26
An Integrated High Content Analysis Pipeline for Alzheimer’s Disease Drug Discovery
Stephen Wong- Weill Cornell Medical College, The Methodist Hospital Research Institute
Peng Shi (Weill Cornell Medical College, The Methodist Hospital Research Institute, Ting Tsung and Wei Fong Chao Center for Bioinformatics Research and Neurosciences Imaging); Xiaobo Zhou (Weill Cornell Medical College, The Methodist Hospital Research Institute, Ting Tsung and Wei Fong Chao Center for Bioinformatics Research and Neurosciences Imaging); Marta Lipinski (Harvard Medical School, Department of Cell Biology); Alexei Degterev (Tufts University, Department of Biochemistry);
Short Abstract: An integrative bioinformatics tool is developed for high content screening of Alzheimer's disease drug discovery. In the analysis pipeline, based on neurite loss quantization with image quality control, hits selection for treating Alzheimer's disease is highly improved by integrating compound chemical structure with activity analysis of neurite outgrowth assays.
Long Abstract:Click Here

Poster A27
Molecular Re-Classification of renal disease through approximate graph matching, clustering and pattern mining
Ramakrishna Varadarajan- University of Wisconsin-Madison
Felix Eichinger (University of Michigan, Ann Arbor, University of Michigan Health System); Jignesh Patel (University of Wisconsin-Madison, Department of Computer Sciences); Matthias Kretzler (University of Michigan, Ann Arbor, University of Michigan Health System);
Short Abstract: Renal diseases are in need of a molecular disease definition. Renal biopsy mRNA-expression profiles and literature derived information were used to generate patient-specific transcriptional networks. Approximate graph-matching was used to cluster transcriptional networks. Distinct network motifs shared within clusters were identified to infer common molecular function.
Long Abstract:Click Here

Poster A28
Transcriptional and post-transcriptional regulatory networks in biliary atresia
Vivek Kaimal- Cincinnati Children's Hospital Medical Center
Bruce Aronow (Cincinnati Children's Hospital Medical Center, Biomedical Informatics); Anil Jegga (Cincinnati Children's Hospital Medical Center, Biomedical Informatics); Jorge Bezerra (Cincinnati Children's Hospital Medical Center, Division of Gastroenterology, Hepatology and Nutrition);
Short Abstract: We employ bioinformatics-based comparative and integrative genomics techniques supplementing gene expression analyses to identify the underlying transcription factor and microRNA based networks in Biliary Atresia. microRNA expression combined with gene expression further strengthens the findings.
Long Abstract:Click Here

Poster A29
Genome-Wide Primer and Probe Design Using PRIMEGENS
Garima Kushwaha- Christopher S. Bond Life Sciences Center
Gaima Kushwaha (Christopher S. Bond Life Sciences Center, Informatics Institute); Gyan Srivastava (Christopher S. Bond Life Sciences Center, Computer Science Department ); Dong Xu (Christopher S. Bond Life Sciences Center, Informatics Institute, Computer Science Department );
Short Abstract: PRIMEGENS is a tool for high-throughput primer and probe design by integrating Primer3 and BLAST in a pipeline using various efficient algorithms. It is unique compared to other related tools in automatically validating primer specificity against a genome or customized sequence database for large-scale sequences at a time.
Long Abstract:Click Here

Poster A30
Differential regulation of gene expression by copy-number alterations in cancer sub-types
yinyin yuan- cancer research uk
No additional authors
Short Abstract: We propose a deregulation network modelling method that integrates copy-number and expression data for modelling differential regulatory relationships between different disease subtypes. On a real breast cancer dataset we reveal both known and novel aspects of pathway deregulation in ER positive versus negative disease as well as crosstalk between pathways.
Long Abstract:Click Here

Poster A31
A computational pipeline for diagnostic biomarker discovery in the human pathogen Trypanosoma cruzi
Santiago Carmona- Universidad Nacional de San Martín
Paula Sartor (Universidad de Buenos Aires, Departamento de Microbiología); Susana Leguizamón (Universidad de Buenos Aires, Departamento de Microbiología); Oscar Campetella (Universidad Nacional de San Martín, Instituto de Investigaciones Biotecnológicas); Fernán Aguero (Universidad Nacional de San Martín, Instituto de Investigaciones Biotecnológicas);
Short Abstract: An integrative bioinformatic strategy was developed to identify peptidic antigens with low cross-reactivity in the pathogen's genome, based on a number of attributes such as protein disorder, tandem repeats, subcellular localization, codon usage, sequence similarity against human and related pathogens, etc. Peptide microarray technology was used to validate predictions
Long Abstract:Click Here

Poster A32
Systematic Evaluation of Disease Gene Identification in Molecular Interaction Networks
Jiao Li- National Library of Medicine
Zhiyong Lu (National Library of Medicine, National Center for Biotechnology Information); Xiaoyan Zhu (Tsinghua University, Department of Computer Science and Technology); Jake Chen (Purdue University, Department of Computer and Information Science);
Short Abstract: Recently, a number of methods have been proposed to identify disease genes by measuring distance to known disease genes in molecular interaction network. Here we systematically evaluated a set of factors in these methods: prior knowledge on disease-related genes, quality of molecular interaction network, and strategy for ranking candidate genes.
Long Abstract:Click Here

Poster A33
Genomic Analysis of Bacterial Pathogens to Identify Factors Contributing to Virulence
Shannan Ho Sui- Simon Fraser University
Raymond Lo (Simon Fraser University, Molecular Biology and Biochemistry); Rylan Fernandes (Simon Fraser University, Molecular Biology and Biochemistry); Jennifer Gardy (British Columbia Centre for Disease Control, Genome Research Laboratory); James Johnston (British Columbia Centre for Disease Control, Tuberculosis Control); Patrick Tang (British Columbia Centre for Disease Control, Genome Research Laboratory); Bob Johnsen (Simon Fraser University, Molecular Biology and Biochemistry); David Baillie (Simon Fraser University, Molecular Biology and Biochemistry); Steven Jones (Canada’s Michael Smith Genome Sciences Centre, Bioinformatics); Robert Brunham (British Columbia Centre for Disease Control, ); Fiona Brinkman (Simon Fraser University, Molecular Biology and Biochemistry);
Short Abstract: We are developing computational methods to enhance our understanding of bacterial pathogenesis. We describe an approach aiding (i) identification of anti-infective drug targets and drug leads and (ii) integration of genome sequence analysis with epidemiological investigation to yield insights into the origins and transmission dynamics within infectious outbreaks.
Long Abstract:Click Here

Poster A34
Phylogenetic methods for inferring tumor progression pathways from aCGH profiles of mixed cell populations
Ayshwarya Subramanian- Carnegie Mellon University
Russell Schwartz (Carnegie Mellon University, Biological Sciences);
Short Abstract: We examine character-based parsimony methods for inferring tumor phylogenies from computationally unmixed array comparative genomic hybridization (aCGH) data. The methods are designed to identify common tumor progression pathways and the major genetic abnormalities that characterize them in the presence of high uncertainty in reconstructions of tumor heterogeneity.
Long Abstract:Click Here

Poster A35
Next-Generation Sequencing workflow for Research and Clinical Applications
Sivakumar Gowrisankar- Partners Healthcare
Mollie Ullman-Cullere (Partners Healthcare, Partners Center for Personalized Genetic Medicine); Jordan Lerner-Ellis (Partners Healthcare, Partners Center for Personalized Genetic Medicine); Lisa Farwell (Partners Healthcare, Partners Center for Personalized Genetic Medicine); Oleg Iartchouk (Partners Healthcare, Partners Center for Personalized Genetic Medicine); Alison Brown (Partners Healthcare, Partners Center for Personalized Genetic Medicine); Sandy Aronson (Partners Healthcare, Partners Center for Personalized Genetic Medicine);
Short Abstract: Next-generation sequencing technology has been behind many of the research discoveries in the last few years. More recently many labs are forming clinical tests around this technology. We focus on a bioinformatics/workflow solution for next-generation sequencing that cater to both clinical as well as research needs in our core laboratory.
Long Abstract:Click Here

Poster A36
An accurate prediction of bacterial effectors using a feature-based supervised learning approach
Sneha Joshi- University of Missouri
Dmitry Korkin (University of Missouri, Informatics Institute and Dept. of Computer Science);
Short Abstract: We developed an accurate supervised learning approach to predict bacterial effectors by integrating protein sequence, structure and genomic information. We train three SVM classifiers, one that detects the signal in N-terminal, another in C-terminal, and the third one in entire sequence. We classify effector and non-effector proteins with 95-96% accuracy.
Long Abstract:Click Here

Poster A37
A new disease-specific machine learning approach for the prediction of cancer-related SNPs
Emidio Capriotti- Stanford University
Russ B. Altman (Stanford University, Departments of Bioengineering and Genetics);
Short Abstract: Non-synonymous Single Nucleotide Polymorphisms (nsSNPs) occurring in coding regions may affect protein function and lead to a diseased state. We developed a disease-specific machine learning approach to predict if a nsSNP can be associated to cancer. Our predictor resulted in 75% accuracy, 0.50 correlation coefficient, and AUC of 0.82.
Long Abstract:Click Here

Poster A38
The implications of protein subcellular localization for disease profiling
Solip Park- POSTECH
Jae-seong Yang (POSTECH, school of interdisplinary bioscience and bioengineering); Jihye Hwang (POSTECH, Life science); Young-Eun Shin (POSTECH, Life science); Yeonjoo Yoo (POSTECH, school of interdisplinary bioscience and bioengineering); Sung Key Jang (POSTECH, school of interdisplinary bioscience and bioengineering); Sanguk Kim (POSTECH, school of interdisplinary bioscience and bioengineering);
Short Abstract: We investigated the relationship of protein localizations and disease associations. We discovered that the spatial constraint from localization significantly improved the protein-disease associations. We found that two diseases displayed high cormorbidity if the disease-causing proteins were found in same localization. We newly identified 9,185 pairs of diseases in the context
Long Abstract:Click Here

Poster A39
The Pearls of Swine Flu: The structural and sequence evolution of the H1N1 proteome
Samantha Warren- University of Missouri
Gavin Conant (University of Missouri, Division of Animal Sciences and Informatics Institute); Dmitry Korkin (University of Missouri, Department of Computer Science and Informatics Institute);
Short Abstract: We integrated structural and genomic data from 75 strains of the influenza H1N1 virus to study the evolution of its proteome. We find (1) a single coherent pattern of faster evolution on the protein surfaces and (2) large clusters of surface residues that show 100% sequence conservation across all strains.
Long Abstract:Click Here

Poster A40
statistical analysis of peripheral tolerance models
Reuma Magori-Cohen- Bar Ilan University
Yoram Louzoun (Bar Ilan University, Mathematics);
Short Abstract: Self-tolerance occurs through the deletion of self reacting lymphocytes. However, lymphocyte-receptors are naturally highly cross-reactive and some error rate must be tolerated. We propose a statistical formalism to compute the accuracy of different tolerance models. Our conclusions can be extended from self/non-self discrimination to signal processing in cross-reactive biological systems.
Long Abstract:Click Here

Poster A41
Modeling Trypanosomal TOR Kinase Domains: Implications for the Design of Anti-Parasitic Drugs
Zhouxi Wang- Northeastern University
Mary Jo Ondrechen (Northeastern University, Chemistry and Chemical Biology); Michael P. Pollastri (Northeastern University, Chemistry and Chemical Biology);
Short Abstract: The kinase domains of trypanosomal Target of Rapamycin proteins (TrypTORs) have been shown to be promising targets for the control of trypanosomal diseases. Comparative models of the kinase domains for TrypTORs, predictions of binding residues, and docking of derivatives of "hit" compounds provide computational guidance for TrypTOR inhibitor design.
Long Abstract:Click Here

Poster A42
The influence of sex on prognostic markers for non-small cell lung cancer
Paul Boutros- Ontario Institute for Cancer Research
No additional authors
Short Abstract: Sex plays a major role in non-small cell lung cancer etiology and response to therapy. The combination of a novel meta-analysis technique and sex-specific survival analysis is used to demonstrate that it also affects the efficacy of transcriptomic prognostic markers for this disease.
Long Abstract:Click Here

Poster A43
Cancer-specific High-throughput Annotation of Somatic Mutations
Hannah Carter- Johns Hopkins
No additional authors
Short Abstract: Large-scale sequencing of cancer genomes has uncovered thousands of DNA alterations of ambiguous functional relevance to tumorigenesis. To assist experimentalists in the search for causative agents, we have developed CHASM, a sequence-based machine learning method to prioritize somatic missense mutations likely to cause functional changes that enhance tumor cell proliferation.
Long Abstract:Click Here

Poster A44
Gene variant databases need structured general disease information to facilitate automatic variant pathogenicity assessment
Peter Taschner- LUMC
Johan den Dunnen (LUMC, Human Genetics);
Short Abstract: Automated pathogenic effect prediction of sequence variants is necessary to analyze genome-wide re-sequencing data. Locus-Specific DataBases (LSDBs) store information about genetic variants involved in hereditary disease, but not much about disease mechanisms. LSDBs need to provide this to support variant pathogenicity assessment and correct individual risk prediction by computational means.
Long Abstract:Click Here

Poster A45
Fixed-parameter haplotyping algorithms for general pedigrees with small number of sites
Duong Doan- University of New Brunswick
No additional authors
Short Abstract: Computing the minimum number of recombination events for general pedigrees with a small number of sites is NP-hard. We reduce it to solving Graph Bipartization by Edge Removal with additional parity constraints, and present a O(2^{k}2^{m^2}n^{2}m^{3}) algorithm to solve the problem for n members, m sites, and k recombination events.
Long Abstract:Click Here

Poster A46
A SURVEY ON EPITOPE PREDICTION METHODS FOR PARASITES GENOMES
Daniela Resende- Universidade Federal de Ouro Preto
Alexandre Reis (Universidade Federal de Ouro Preto, Laboratório de Imunopatologia); Nesley Oliveira (Instituto René Rachou, Laboratório de Parasitologia Celular e Molecular); Rommel Ramos (Universidade Federal do Pará, Laboratório de Polimorfismo de DNA); Louise Cerdeira (Universidade Federal do Pará, Laboratório de Polimorfismo de DNA); Raul Torrieri (Instituto René Rachou, Laboratório de Parasitologia Celular e Molecular); Patrícia Ruy (Instituto René Rachou, Laboratório de Parasitologia Celular e Molecular); Izabella Batista (Instituto René Rachou, Laboratório de Imunologia Celular e Molecular); Rodrigo Corrêa-Oliveira (Instituto René Rachou, Laboratório de Imunologia Celular e Molecular); Jerônimo Ruiz (Instituto René Rachou, Laboratório de Parasitologia Celular e Molecular);
Short Abstract: Immunoinformatics uses genome sequences as material for antigen identification. In this work, we developed a database approach to evaluate epitope prediction performances and the best group of tools capable of predicting B/T-cell epitopes in parasites. Results show that NetCTL, NetMHCII and AAP12 predict 96.63%, 93.59% and 98.45% of tested epitopes.
Long Abstract:Click Here

Poster A47
Phenotype-driven genetic analysis of Autism spectrum disorders
Benjamin Georgi- University of Pennsylvania
No additional authors
Short Abstract: The genetic basis of Autism spectrum disorders (ASD) is complex and the
heterogeneity of ASD genotype data makes association studies extremely
challenging. We stratify genotypes in a phenotype-driven approach to obtain more
homogeneous samples. In a first analysis we find significant, subgroup-specific variation on both the phenotype and genotype level.
Long Abstract:Click Here

Poster A48
IDENTIFICATION OF ALTERNATIVE POLYADENYLATION VARIANTS DIFFERENTIALLY EXPRESSED IN HUMAN ASTROCYTOMAS
Suzana Ezquina- Ludwig Institute for cancer research
Suely Marie (FMUSP- Faculdade de Medicina da Universidade de São Paulo, Neurology-LIM15); Sueli Oba-Shinio (FMUSP- Faculdade de Medicina da Universidade de São Paulo, Neurology-LIM15); Sandro Souza (Ludwig Institute for cancer research, Computational Biology);
Short Abstract: Alternative polyadenylation is an important source of variability in human transcripts, affecting several cellular mechanisms. Microarray data of non-neoplastic brain tissues and 3 grades of astrocytomas, were used to estimate gene expression of affected transcripts.
Alternative transcripts are differentially expressed in glioblastoma showing a possible expression shift in tumors.
Long Abstract:Click Here

Poster A49
Transcriptional Profiling of Peripheral Neuropathy in Type 2 Diabetes using db/db mouse model
Manjusha Pande- University of Michigan
Junguk Hur (University of Michigan, Bioinformatics); Kelli Sullivan (University of Michigan, Neurology); Matthias Kretzler (University of Michigan, Internal Medicine); Eva Feldman (University of Michigan, Neurology);
Short Abstract: We studied gene expression in sciatic nerve of non-diabetic (db/+) and diabetic (db/db) mice using microarrays to identify the role of hyperlipidemia in the development of diabetic neuropathy (DN). Analysis of differentially expressed genes suggests increased lipid metabolism may lead to elevated mitochondrial oxidative stress thus contributing to DN progression.
Long Abstract:Click Here

Poster A50
Clustering analysis for an integrated genomic data from ovarian cancer tissues
Chang Sik Kim- Korean Bioinformation Center/Korea Research Institute of Biotechnology and Bioscience
Namjin Koo (Korean Bioinformation Center/Korea Research Institute of Biotechnology and Bioscience, Biomedical Informatics); Chae Hwa Seo (Korean Bioinformation Center/Korea Research Institute of Biotechnology and Bioscience, Biomedical Informatics); JiHong Kim (The Catholic University of Korea, Integrated Research Center for Genome Polymorphism); Sanghyuk Lee (Korean Bioinformation Center/Korea Research Institute of Biotechnology and Bioscience, Systems Bioinformatics); In-Sun Chu (Korean Bioinformation Center/Korea Research Institute of Biotechnology and Bioscience, Biomedical Informatics);
Short Abstract: It is a great challenge in systems biology community to develop novel methods for integrating and analyzing multiple omics data into cohesive way. This study summarized the process in integrating and analyzing three types of omics datasets with large number of ovarian cancer samples from The Cancer Genome Atlas project.
Long Abstract:Click Here

Poster A51
is-rSNP: A novel technique for in silico regulatory SNP detection
Geoff Macintyre- University of Melbourne
James Bailey (University of Melbourne, Department of Computer Science and Software Engineering); Izhak Haviv (Baker IDI heart and diabetes Institute, Bioinformatics and systems integration); Adam Kowalczyk (NICTA VRL, Diagnostic Genomics);
Short Abstract: Experimental screening for regulatory SNPs is expensive and labour intensive. We provide an in silico alternative that accurately predicts SNPs that disrupt certain transcription factor binding sites. Our novel convolution methods, that determine the complete distributions of position weight matrix scores and ratios between allele scores, facilitate statistically robust predictions.
Long Abstract:Click Here

Poster A52
Evolution model of host pathogen interaction applied to Mycobacterium Tuberculosis
Ashwani Kumar- Vellore Institue of Technology University
Sumanta Mukherjee (Indian Institute of Sciences,Bangalore, Mathematics); I Arnold Emerson (VIT UNiversity, Bioinformatics);
Short Abstract: Here we address host pathogen interactions and evolution of their survival strategies in a single mathematical model. This model is generic to all bacteria and mammalian host interactions and incorporates both microscopic and mesoscopic evolution in a single mathematical framework. We analyze the same for Mycobacterium Tuberculosis infection.
Long Abstract:Click Here

Poster A53
Comparing Two Computational Methods for Common Insertion Site Detection in Transposon Insertional Mutagenesis Data
Jelle ten Hoeve- The Netherlands Cancer Institute
Alistair G. Rust (2The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, Experimental Cancer Genetics); Jeroen de Ridder (Faculty of Electrical Engineering, Mathematics, and Computer Science, Delft University of Technology, Delft, Delft Bioinformatics Lab); David J. Adams (The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, Experimental Cancer Genetics); Lodewyk Wessels (The Netherlands Cancer Institute, Amsterdam, Division of Molecular Biology);
Short Abstract: In the realm of transposon insertional mutagenesis in mice, CIMPLR and a complementary Monte Carlo (MC) simulation approach have been employed to detect Common Insertion Sites in various cancer screens. We compare both methods on simulated and real datasets, and interpreted them in the context of the simulation experiments.
Long Abstract:Click Here

Poster A54
Robust Pharmacogenomic Predictors Identified from Breast Cancer Cell Lines by Meta-analysis Improve the Prediction Accuracy of Breast Cancer Patient Chemotherapy Responses
Kui Shen- Precision Therapeutics Inc.
Nan Song (Precision Therapeutics Inc., Informatics); Shara Rice (Precision Therapeutics Inc., Research and development); Dave Gingrich (Precision Therapeutics Inc., Informatics); Chunqiao Tian (Precision Therapeutics Inc., Informatics); Dakun Wang (Precision Therapeutics Inc., Research and development); Zhenyu Ding (Precision Therapeutics Inc., Informatics); Stacey Brower (Precision Therapeutics Inc., Research and development); Paul Ervin (Precision Therapeutics Inc., Research and development); Mike Gabrin (Precision Therapeutics Inc., Informatics); Shuguang Huang (Precision Therapeutics Inc., Informatics);
Short Abstract: Pharmacogenomic predictors developed from cancer cell lines have been used to predict patient clinical outcomes. By applying meta-analysis to independent microarray studies coupled with in vitro chemoresponse assays, we demonstrated the feasibility and reproducibility of using breast cancer cell lines to identify robust predictors of breast cancer patients' chemotherapy response.
Long Abstract:Click Here

Poster A55
Efficient Inference: Perturbagen Effect on Protein Subcellular Location Pattern
Armaghan Naik- Carnegie Mellon University
Joshua Kangas (Carnegie Mellon University, Lane Center for Computational Biology); Christopher J. Langmead (Carnegie Mellon University, Lane Center for Computational Biology , Computer Science Department); Robert Murphy (Carnegie Mellon University, Lane Center for Computational Biology );
Short Abstract: Learning a predictive model of the effect of perturbagens on protein subcellular location from limited, sparse data is an important alternative to exhaustive measurement. Several competing models may be inferred from such data; we compare a nonparametric Bayesian model with parametric and tensorial regression methods in scale and accuracy.
Long Abstract:Click Here

Poster A56
Stem Cell Discovery Engine
Kimberly Begley- Harvard School of Public Health
Dorothy Reilly (Harvard School of Public Health, Biostatistics); Oliver Hofmann (Harvard School of Public Health, Biostatistics); Ray McGovern (Harvard School of Public Health, Biostatistics); Gabriel Altshuler (Harvard School of Public Health, Biostatistics); Jieun Jeong (Harvard School of Public Health, Biostatistics); Ramakrishna Sompallae (Harvard School of Public Health, Biostatistics); Ramesh Shivdasani (Dana-Farber Cancer Institute, Genetics); Scott Armstrong (Dana-Farber Cancer Institute, Pediatrics); Winston Hide (Harvard School of Public Health, Biostatistics);
Short Abstract: SCDE is a collection of computational tools with which to integrate multiple forms of genome-wide data and allow stem cell properties to be systematically compared across species and tissues. Experimental data is collected, curated, processed for quality control, and stored where it can be queried and integrated with additional data.
Long Abstract:Click Here

Poster A57
Identification of novel chromosomal rearrangements via massively parallel sequencing and primer-directed in silico assembly
Joseph Fass- UC Davis
Vince Buffalo (UC Davis, Genome Center Core Facilities - Bioinformatics Core); Shyh-Jen Shih (UC Davis, Radiation Oncology); Andrew Vaughan (UC Davis, Radiation Oncology); Dawei Lin (UC Davis, Genome Center Core Facilities - Bioinformatics Core);
Short Abstract: Chromosomal rearrangements involving human MLL have previously been characterized using inverse PCR (IPCR), followed by time-consuming and expensive lab work and Sanger sequencing. We have developed primer-initiated assembly algorithms which allow discovery of known or novel rearrangements solely by analysis of single-ended Illumina sequencing of mixed IPCR products.
Long Abstract:Click Here

Poster A58
Integrative Biomarker Discovery from Gene Expression and Protein-protein Interaction Data Using Error-tolerant Pattern Mining
Rohit Gupta- University of Minnesota
Ze Tian (University of Minnesota, Computer Science); Rui Kuang (University of Minnesota, Computer Science); Vipin Kumar (University of Minnesota, Computer Science);
Short Abstract: A novel error-tolerant pattern mining-based technique is proposed for integrative biomarker discovery using gene-expression and protein-protein interaction data. This approach directly addresses the issue of noise in the data and systematically and efficiently discovers all biomarkers that are more reproducible and predictive of disease phenotype as compared to single-gene markers.
Long Abstract:Click Here

Poster A59
Brief Exposure
Emmett Sprecher- Yale University
David Tuck (Yale University, Pathology Informatics / Computational Biology & Bioinformatics Program); Lyndsay Harris (Yale University School of Medicine, Breast Cancer Program, Yale Cancer Center); Zongzhi Liu (Yale University, Pathology Informatics); Ao Li (Yale University, Pathology Informatics); Ian Krop (Dana-Farber Cancer Institute and Harvard Medical School, Medical Oncology); Kimberly Lezon-Geyda (Yale University School of Medicine, Breast Cancer Program, Yale Cancer Center); Sudipa Sarkar (Yale University School of Medicine, Breast Cancer Program, Yale Cancer Center);
Short Abstract: Biomarkers of trastuzumab-resistance in HER2+ breast cancer have proved elusive. Employing a "brief exposure" paradigm, we discovered 61 genes whose expression changed in responders across a single dose of trastuzumab (q-val < 0.05), yet none in resistant tumors. Many patients also displayed extensive copy number differences over treatment.
Long Abstract:Click Here

Poster A61
Comparative pathway analysis of aging associated genes in humans and model organisms
Ari Berman- Buck Institute for Age Research
Tobias Wittkop (Buck Institute for Age Research, Bioinformatics Core); Uday Evani (Buck Institute for Age Research, Bioinformatics Core); Sean Mooney (Buck Institute for Age Research, Bioinformatics Core);
Short Abstract: We compared two gene sets associated with aging, GenAge (human genes), and AnAge (animal model genes), to determine if the gene sets from different organisms carried any functional similarities. Comparisons were performed using gene ontology and ingenuity pathway analysis. The results show common elements between humans and animal models.
Long Abstract:Click Here

Poster A62
An RNA-Seq analysis of mis-regulated alternative splicing in Alzheimer’s Disease
Xinchen Wang- University of Toronto
Sandy Pan (University of Toronto , Banting and Best Department of Medical Research); Angela Hodges (King's College London, MRC Centre for Neurodegeneration Research); Simon Lovestone (King's College London, MRC Centre for Neurodegeneration Research); Benjamin J. Blencowe (University of Toronto, Banting and Best Department of Medical Research);
Short Abstract: We used high-throughput RNA sequencing (RNA-seq) to perform a genome-wide analysis of disease associated changes in the transcriptome of Alzheimer's Disease (AD) patients. Distinct sets of functionally related genes were identified with disease associated alternative splicing and mRNA expression changes.
Long Abstract:Click Here

Poster A63
Predictive Modeling of Macrophage Transcriptional Response to Nanoparticle Exposure
Michelle Costa- PNNL
Jason McDermott (PNNL, Computational Biology & Bioinformatics); Katrina Waters (PNNL, Computational Biology and Bioinformatis); Harish Shankaran (PNNL, Computational Biology & Bioinformatics); Brian Thrall (PNNL, Biological Science Divison);
Short Abstract: None On File
Long Abstract:Click Here

Poster A64
ISOLATE: A model for in silico purification of tumor gene expression profiles
Gerald Quon- University of Toronto
Paul Boutros (Ontario Institute for Cancer Research, ); Quaid Morris (Terrence Donnelly Centre for Cellular and Biomolecular Research, Banting and Best Department of Medical Research);
Short Abstract: Tumour gene expression profiling has proved valuable in cancer diagnosis and treatment. However, contamination of signal by non-tumor cells is still a central obstacle to their effective use. We have developed ISOLATE, a model to perform in silico purification of tumor expression profiles.
Long Abstract:Click Here

Poster A65
Revealing the role of conserved disordered regions in the human proteome using comparative proteomics
Taehyung Kim- University of Toronto
Philip Kim (University of Toronto, Banting and Best department of Medical Research);
Short Abstract: Intrinsically disordered regions are protein segments without a fixed conformation. Recently, studies have shown the positive correlation between IDRs and human diseases. More recently, the concept of conserved disorder was introduced which showed their evolutionary importance. We investigate the role of conserved disordered regions in human proteome in evolution perspective.
Long Abstract:Click Here

Poster A66
Personalized Medicine: Optimizing HIV Treatment to Patient and Virus
Betty Cheng- Carnegie Mellon University
Betty Cheng (Carnegie Mellon University, Language Technologies Institute, School of Computer Science);
Short Abstract: Drug resistance is a major obstacle in the treatment of many infectious diseases, particularly with rapidly mutating retroviruses like HIV. Combination drug therapy is often used to delay the emergence of resistance. A patient develops resistance not only to drugs which s/he has taken but also to similar drugs; hence treatment optimization at each step is crucial. Current treatment outcome prediction systems focus on susceptibility to a single drug given the virus genotype, ignoring the selective pressure exerted by drugs working in combination and the individual differences between patients. Individual differences may impact treatment outcome directly or indirectly via treatment adherence. Here, we investigated the importance of patient information versus virus genotype in predicting short-term and long-term treatment outcome. Patient information included demographic data (e.g. age, race, financial stability, substance abuse), HIV treatment history, and other infections and medications. Defining the problem as a 3-class classification task, we employed chi-square to automatically select informative features and applied four different classifiers on the selected features. Combination of patient information and virus genotype consistently yielded higher accuracy than either information source alone, suggesting the two to be complimentary in treatment outcome prediction. The four classifiers yielded on average 58.4% accuracy compared to baseline majority 33.8%. Further analysis showed much of the model's errors lie close to class boundaries and may coincide with noise in viral load measurements. Our next step is to cluster features to model non-linear feature interactions and learn whether some genotype mutations affect treatment outcome in specific patient groups.
Long Abstract:Click Here

Poster A67
Novel statistics reveal cancer universal microRNA activity
Roy Navon- Agilent Laboratories
Hui Wang (Agilent Laboratories, Agilent Laboratories); Israel Steinfeld (Technion, CS); Anya Tsalenko (Agilent Laboratories, -); Amir Ben-Dor (Agilent Laboratories, -); Zohar Yakhini (Agilent Laboratories, -);
Short Abstract: microRNAs (miRNAs) regulate genes and play important roles in cancer pathogenesis and development. Variation amongst individuals is a significant confounding factor in miRNA (or other) expression studies. The true character of biologically or clinically meaningful differential expression can be obscured by inter-patient variation. We will present data from microarray profiling of more than 700 miRNAs in 28 matched (same patient) tumor/normal samples from 8 different tumor types (breast, colon, liver, lung, lymphoma, ovary, prostate and testis) – a design that minimizes tissue type and patient related variability. We will then describe novel statistical methods used in analyzing this data. The analysis revealed several miRNA that are consistently differentially expressed over multiple tumor types. These differentially expressed miRNAs include known oncomiRs as well as miRNAs that were not previously universally associated with cancer, such as miR-133b and miR-486-5p, both consistently down regulated in cancer, in the context of our cohort.
Long Abstract:Click Here

Accepted Posters


View Posters By Category
Search Posters:
Poster Number Matches
Last Name
Co-Authors Contains
Title
Abstract Contains






↑ TOP