Sequencing a genome is only the beginning. Several layers of analysis are necessary to convert raw sequence data into an understanding of functional biology. First, error sources in the original raw data from multiple platforms and diverse applications must be accounted for. Then, as computational methods for assembly, alignment and variation detection continue to advance, a broad range of genetic analysis applications including comparative genomics, high-throughput polymorphism detection, analysis of coding and non-coding RNAs and identifying mutant genes in disease pathways can be addressed. CHI’s Sequencing Data Analysis and Interpretation conference combines unique perspectives from a variety of researchers, engineers, biostatisticians and software developers involved in NGS data analysis.
Day 1 | Day 2
Wednesday, August 21
7:30 am Breakfast Technology Workshop (Sponsorship Opportunity Available)
8:45 Chairperson's Remarks
Chris Dwan, Acting Senior Vice President, IT, New York Genome Center
» Featured Presentation
8:50 Building the Next Generation of Next-Gen Pipelines
Toby Bloom, Ph.D., Deputy Scientific Director, Informatics, New York Genome Center Biography
In the early days of next-generation sequencing, the focus of informatics pipelines was on storing all the data, coping with network and compute bandwidth problems and reliably processing huge numbers of short reads. Seven years after that first Solexa instrument arrived, we now face a new set of daunting challenges: we must analyze larger and larger numbers of samples concurrently, combine many types of data in one analysis, identify and locate the needed data and integrate ever-larger amounts of clinical data. We discuss these challenges from both the perspective of the informatics infrastructure to support these analyses and the impact on the design on new methods and algorithms.
9:20 Can We Maintain Sanity as NGS Pipelines Change?
Stuart M. Brown, Ph.D., Professor, Center for Health Informatics and Bioinformatics, New York University School of Medicine Biography
NGS technology and software are evolving very quickly. It is difficult to maintain a consistent analysis pipeline across an experiment or group of related experiments when samples are processed months or years apart. Sample preparation, sequencing technology and software can all influence the results, when the investigator is really only interested in the biology. Does the software alone have a substantial impact on the results reported for an RNA-seq experiment? We re-analyzed some FASTQ data files from experiments previously analyzed by the Illumina CASAVA RNA-seq pipeline and compared the results with alignments by TopHat and STAR and gene expression measured by Cufflinks.
9:50 Selected Oral Poster Presentation: Single-Cell RNA-Seq to Define Cell-Specific Gene Expression In the Developing Lung
Yan Xu, Ph.D., Associate Professor, Pediatrics, Division of Pulmonary Biology and Biomedical Informatics, Cincinnati Children's Hospital Medical Center, University of Cincinnati School of Medicine
Pulmonary research has been challenged by remarkable complexity and diversity of lung cell types, which have limited our ability to find answers to fundamental questions regarding lung structure, function and disease. Single-cell mRNA-Seq is a powerful tool for understanding complex biological systems such as lungs. Studying lung transcriptomes at the single-cell level will significantly improve the sensitivity and resolution of mRNA expression analysis and enable the construction of TRNs that directly reflect the regulation of expression in a cell autonomous manner. Here, we will discuss a pilot study using a single-cell RNA-Seq technique to define gene expression in the developing lung.
10:05 Coffee Break in the Exhibit Hall with Poster Viewing
11:00 Building a Genomic Experiment Tracking and Analysis System
Craig Pohl, Co-Director, Bioinformatics, The Genome Institute, Washington University Biography
As sequencing technology has become a commodity, The Genome Institute at Washington University has expanded from a large-scale sequencing center to a large-scale experimental genomics center. Our system currently tracks thousands of active projects, generating and analyzing their data automatically, where each project's experimental design is customized according to the hypothesis in question. Therefore, our system seeks to address the fundamental problem of translating experimental inputs, such as sequence data and samples with rich metadata, into a highly efficient, cost effective, modular and automated genomic data analysis modeling system that provides customized results for each project. Insights gained from the process of developing this system will be shared.
11:30 Panel Discussion with Morning Speakers
Moderator: Chris Dwan, Acting Senior Vice President, IT, New York Genome Center
High volumes of sequencing data and an incredible range of research needs must be managed efficiently to ensure accurate data analysis and customized results that contribute meaningfully to biology. Considerations of informatics infrastructure, computational power and storage possibilities are necessary to meet this challenge. This panel discussion offers bioinformatics experiences and strategies from NGS data analysis experts.
Panelists:
Toby Bloom, Ph.D., Deputy Scientific Director, Informatics, New York Genome Center
Stuart M. Brown, Ph.D., Professor, Center for Health Informatics and Bioinformatics, New York University School of Medicine
Yan Xu, Ph.D., Associate Professor, Pediatrics, Division of Pulmonary Biology and Biomedical Informatics, Cincinnati Children's Hospital Medical Center, University of Cincinnati School of Medicine
Craig Pohl, Co-Director, Bioinformatics, The Genome Institute, Washington University
12:00 pm Close of Session
12:15 Luncheon Technology Workshop (Sponsorship Opportunity Available)
1:30 Chairperson's Remarks
Mary Ann Brown, Executive Director, Conferences, Cambridge Healthtech Institute
1:35 Network Models: How to Make Sense of Your NGS Data
Ernest Fraenkel, Ph.D., Associate Professor, Biological Engineering, Massachusetts Institute of Technology Biography
How are we to make sense of the torrents of data from next-generation sequencing? In this talk, we will explain how network models provide biological insight into the data and generate testable hypotheses that are the key to understanding disease.
2:05 MBD-Seq and Genotype-Methylation Association Study Using Next-Generation Sequencing Technologies
Andrey Shabalin, Ph.D., Research Scientist, Edwin J.C.G. van den Oord Laboratory, Center for Biomarker Research and Personalized Medicine, Pharmacotherapy and Outcomes Science, Virginia Commonwealth University Biography
Variation in DNA methylation has been repeatedly associated with a wide variety of human diseases. In this study, we aim to comprehensively characterize methylation quantitative trait loci (meQTLs) by associating genome-wide SNP genotypes with methylation measures across the methylome. The statistical and biological analyses of this study involve tests for association between each of over 5,000,000 SNPs and each of 4,954,484 CpG blocks. Our results indicate different mechanisms of local versus long-range genetic control of the methylome.
2:35 ChIP-Seq Analytics: Combining Multiple ChIP-Seq Peak Detection Systems
Christina Schweikert, Ph.D., Division of Computer Science, Mathematics and Science, St. John's University Biography
The abundance of sequencing data being generated enables us to analyze genome-wide protein-DNA interactions, as well as evaluate and enhance computational and statistical techniques for locating protein binding sites. We define methods to merge and rescore the regions of two peak detection systems and analyze the performance based on average precision and coverage of transcription start sites. The results indicate that ChIP-seq peak detection can be improved by applying score or rank combination. System combination and fusion analysis would provide a means to assess available technologies and assist researchers in choosing an appropriate system, and offer another approach to improve the performance of the ChIP-seq peak identification process.
3:05 Refreshment Break in the Exhibit Hall, Last Chance for Poster Viewing
3:35 You Need More Power: Designing Cost-Effective Experiments for Measuring Differential Gene Expression Using RNA-Seq
Michele Busby, Ph.D., Computational Biologist, Broad Institute; former Research Scientist, Biology, Gabor T. Marth Laboratory, Boston College
RNA-Seq is a powerful tool for detecting differential gene expression, but only realizes its full potential when experiments are optimally designed. We will demonstrate how our computational tool Scotty can be used to design an experiment that contains an adequate number of samples sequenced to a sufficient depth to achieve experimental goals. We will further discuss how the performance of different RNA-Seq protocols can dramatically affect the power of an experiment and demonstrate computational techniques for assessing the performance of a protocol.
RNA-Seq Experimental Design and Bioinformatics with Michele Busby
4:05 Deconvolution of Heterogeneous Tissue Samples Based on RNA-Seq Data
Ting Gong, Ph.D., Assistant Professor, Molecular Carcinogenesis, University of Texas MD Anderson Cancer Center Biography
The promising biomedical applications of NGS have spurred the development of new statistical methods to capitalize on the wealth of information contained in RNA-Seq datasets. However, for heterogeneous tissues, measurements of gene expression through RNA-Seq data can be confounded by the presence of multiple cell types present in each sample. Here, we present a statistical pipeline for deconvolution of heterogeneous tissues based on RNA-Seq data.
4:35 Panel Discussion with Afternoon Speakers
Moderator: Jan Vijg, Ph.D., Professor and Chair, Genetics, Albert Einstein College of Medicine
To be experimentally useful, next-generation sequencing data must be properly interpreted for and understood by biologists who have already optimized their experiments for NGS technologies. This expert panel shares network models, computational methods and statistical techniques that can help translate this information for laboratory and clinical settings.
Panelists:
Christine Vogel, Ph.D., Assistant Professor, Center for Genomics and Systems Biology, New York University
Alec Chapman, Research Scientist, X. Sunney Xie Laboratory, Chemistry and Chemical Biology, Harvard University
Michele Busby, Ph.D., Computational Biologist, Broad Institute; former Research Scientist, Biology, Gabor T. Marth Laboratory, Boston College
Ting Gong, Ph.D., Assistant Professor, Molecular Carcinogenesis, University of Texas MD Anderson Cancer Center
Andrey Shabalin, Ph.D., Research Scientist, Edwin J.C.G. van den Oord Laboratory, Center for Biomarker Research and Personalized Medicine, Pharmacotherapy and Outcomes Science, Virginia Commonwealth University
Christina Schweikert, Ph.D., Division of Computer Science, Mathematics and Science, St. John's University
5:00 Close of Sequencing Data Analysis and Interpretation Conference
Day 1 | Day 2