Day 1 | Day 2 | Day 3 | Short Courses
Companion Meeting: Evolution of Next-Generation Sequencing
TUESDAY, SEPTEMBER 28, 2010
|
Sponsored by
7:30 am Breakfast Presentation
Data Intensive Discovery Initiative – Gene/Environment Interaction Analysis on Hybrid FPGA Data Intensive Supercomputers
Murali Ramanathan, Associate Professor of Pharmaceutical Sciences, University at Buffalo, SUNY
In the bio-pharma research domain, rapid advancement in next-gen sequencing instruments, imaging systems, simulations, and discovery-to-market processes are generating vast data sets that need to be analyzed effectively to create new knowledge. Identifying critical interactions between many variables is a key problem in many applications. Massively parallel FPGA database appliances parallelize both data storage and processing. They have been widely accepted for business/Internet search applications. Database appliances analyze massive data sets rapidly, delivering hundred-fold or even thousand-fold speedups over server-based architectures.
 8:15 Successful Sequencing Discussion Groups
Grab a cup of coffee and join a facilitated discussion group focused around specific themes. This unique session allows conference participants to exchange ideas, experiences, and develop future collaborations around a focused topic.
Strategies for NGS Data Retention
Paul Smith, Ph.D., Data Architect, Global Information Services, Illumina
Discussion topics include:
• Can we afford to keep all our data forever?
• How do we handle NGS output growing an order of magnitude faster than storage capacity?
• How do we chose what files to keep and for how long? What are the crown jewels and what can we recreate?
• How do we keep track of all this data and find what we no longer need?
• How do you automate the process of managing your data?
• Can we just keep the source data and the recipe?
Asking Physical Questions Using Sequencing as the Output
Erez Lieberman-Aiden, Ph.D., Fellow, Harvard Society of Fellows, Harvard University
Discussion topics include:
• How can sequencing best be applied to classical questions that may not have been amenable to molecular biological approaches in the past?
• Is there a niche for new types of sequencing technologies that could cater to these new experiments?
Defining and Assessing "Accuracy" and "Errors" in NGS Data
James Lyons-Weiler, Ph.D., Scientific Director, Senior Research Scientist, Genomics and Proteomics Core Laboratories, University of Pittsburgh
Discussion topics include:
• What measures of accuracy and error exist, and what do they really mean?
• Do quality scores correlate with sequencing accuracy?
• Which types of information are critical for assessing confidence in 'detected' variants?
• What biological complexities might reduce assessed accuracy?
• What sequencing technologies help overcome these complexities?
Strategies for Maintaining Large Scale Metagenomic Shotgun Data
Folker Meyer, Ph.D., Computational Biologist, Mathematics and Computer Science, Argonne National Laboratory
Discussion topics include:
• Can we afford redundant computational analysis efforts for metagenome data sets?
• Can we devise a community approach to maintaining computationally analyzed data sets?
• How can we devise the standardization required for this process?
• What are the community needs for a standardized analysis pipeline?
• What data should be included in this process?
• What is the governance model for this? Should it be placed under the auspices of the Genomics Standards Consortium?
Managing Change in the NGS Pipeline
Gregg TeHennepe, Senior Manager, Research Liaison, Information Technology, The Jackson Laboratory
Discussion topics include:
• What are the most frequent causes of problems in the NGS pipeline?
• What NGS environments are most sensitive to the challenges of change in the pipeline?
• What sections of the NGS pipeline see the most change?
Taking NGS to the Clinic
Stephan Sanders, Ph.D., Postdoctoral Associate, Yale University
Discussion topics include:
• What hurdles prevent sequencing results being used in routine clinical practice?
• Which specialties are likely to benefit first?
• How can clinicians be trained to understand the results?
• Which of the 3 million variants should be reported back?
• Where should the results be stored?
• Is the data quality good enough yet?
Computer Hardware, Software and Operating Systems – What do I really need to meet my needs (and those of others who rely on me)?
Stephanie Costello, Director, Sales, DNAStar
Farhan Quraishi, Technical Support Specialist, DNAStar
Discussion topics include:
• Desktop computer, in-house Linux cluster, cloud computing or something in between
• Operating system options – Win, Mac or Linux
• Commercial software, freeware or a combo
How to Make the Most out of Your NGS Data Using Bioinformatics: Service or Software?
Didier G. Perez, COO & CFO, Eureka Genomics
Discussion topics include:
• What are the essential tools that you need?
• Which is the most valuable?
• Criteria for selection of a bioinformatics software and/or service?
Estimating False Positive Rates of Variants
Victor Weigman, Ph.D., Computational Biologist, Expression Analysis
Discussion topics include:
• What effect does heterogeneous tissue (i.e tumor) have on determining alleles?
• What is the concordance between different variant-calling algorithms?
• How do low-depth regions affect calls?
• Are sample-pooling strategies effective?
Selecting the Right Tool for Validation of Next-Generation Sequencing Experiment
Sean Ferree, Ph.D., Director of Product Development, Nanostring Technologies
Discussion topics include:
• How can we accelerate the validation process by leveraging new technologies?
• What are the key parameters of a typical validation study? Where are the bottlenecks?
• What are the most important factors informing the selection of a tool for validation?
Upcoming Technical Challenges Brought About by Next-Generation Sequencing
W. Richard McCombie, Ph.D., Professor, Cold Spring Harbor Laboratory
Discussion topics include:
• How will next-gen sequencing data shape future experiments?
• How can those needed experiments be carried out?
• What problems can't be solved with current technology?
• What new technologies will need to be to maximize utility of next-gen sequencing data?
Challenges and Opportunities in Data Intensive Computing
Murali Ramanathan, Associate Professor of Pharmaceutical Sciences, University at Buffalo, SUNY
Discussion topics include:
• Architectures for data intensive computing
• Algorithms
• Data intensive computing resources
|
9:15 Chairperson’s Remarks
9:20 Challenges in Analysis and Management of Sequencing Data
at Large Scale
Matthew Trunnell, Ph.D., Manager, Research Computing, The Broad Institute of MIT and Harvard
The Broad Institute’s adoption of second-generation sequencing technologies over the past three years has driven a 25-fold growth in the size of its data repositories, placing new pressures on its IT infrastructure, motivating different approaches to data analysis, and demanding a new level of attention to issues of data management. As second and third-generation sequencing technologies continue to drive down the cost of sequencing, IT and informatics expenses comprise an increasingly large fraction of total sequencing costs to the point where consideration of data life cycle management must explicitly weigh financial factors against scientific expectations. This talk will discuss The Broad Institute’s ongoing adjustment to data management at this new multi-petabyte scale and will also briefly highlight some new approaches to data analysis being actively explored.
9:50 Setting Up the Data Infrastructure of NGS Data
Samuel Fulcomer, Assistant Director, Center for Computation and Visualization, Brown University
Using Brown’s system as a case study, I will discuss strategies for designing the low-level system storage and I/O systems for a parallel computing environment. After providing a short survey of the I/O behaviors of some typical informatics applications, I will then present a case for changing the demarcation of the application/system interface, and moving much more of data management into an active repository system better able to optimize storage and caching for application access patterns.
10:20 Improving Sequence Data UtilitySponsored by
Heather Koshinsky, Ph.D., CSO and Co-Founder, Eureka Genomics
What is most useful: longer reads, paired end reads, mate pair reads, or just more reads no matter what they are? The logic behind each type of sequence data will be explored and specific examples of how using the best type of data for a specific project goal can dramatically increase the success of the project.
10:35 Networking Coffee Break, Poster and Exhibit Viewing
11:15 Managing Change in a Next-Generation Sequencing Pipeline
Gregg TeHennepe, Senior Manager, Research Liaison, Information Technology, The Jackson Laboratory
There are multiple stages and dozens of components in next-generation sequencing pipeline. Each component, from chemistries to image analysis to base calling algorithms, has regular updates that improve its quality and/or reliability. Ensuring that these changes are handled in a reliable way is critical to preventing failed runs and tens of thousands of dollars in lost chemicals, samples, and staff time.
11:45 Composite Priority Scores for Next-Generation Sequence Interpretive Analysis: Order from Disorder
James Lyons-Weiler, Ph.D., Scientific Director, Senior Research Scientist, Genomics and Proteomics Core Laboratories, University of Pittsburgh
This presentation will describe the development of novel information scores for establishing priority hits for follow-up validation at the base locus level. We have adopted two next-generation sequencing
platforms; for whole genome sequencing the error rates, while low, result in tens of thousands of potential false positives to screen through. Our novel score, Ambiguity, is an entropy-based score that appears to be super-critical for ruling out false positive leads.
12:15 pm Close of Session
Sponsored by
12:30 Luncheon Presentation
From Deluge to Discovery: Achieving Greater Research Productivity with Powerful Yet Simple Storage Solutions
Chris Blessington, Senior Director of Marketing and Communications, Isilon
Next-generation sequencing is currently experiencing unprecedented growth of critical file-based research data, which traditional storage was simply not designed to manage. Though next-gen sequencing relies heavily on massive data stores, the IT staffs tasked with supporting this research are often small and without dedicated storage technicians, further complicating the problem. However, there is a solution. Scale-out storage accelerates discovery while simplifying management and reducing costs. Through real world examples including multiple other organization presenting at NGSDM, attendees will learn best practices for leveraging scale-out storage in next-gen sequencing environments to generate significantly greater research productivity at much less cost.
I’ve Bought My Sequencer – Now What the Heck Do I Do with the Data?
As never before, the sequencing data deluge has partnered platform manufacturers and software developers with NGS end-users. Join software developers and NGS users in a unique session that partners the user’s data deluge with a company’s software solution, leading to comprehendible biological results.
2:00 Chairperson’s Remarks
2:05 Partnering Company #1: Sponsored by
.jpg?n=1553 "DNAStar")
Deciphering Microbial Diversity in Complex Environments -A Case Study
Scot E. Dowd Ph.D., Director, Research and Testing Laboratory, Director, Pathogenius Diagnostics
Determining the microbial diversity and composition of complex environments has been advanced greatly by next generation pyrosequencing approaches. There are several important tools for estimating microbial diversity. These often involve complex workflows of multiple alignments, distance matrices, and modeling. Here we provide an interesting approach to evaluating the diversity and evenness of environments using DNASTAR's SeqMan NGen assembler as an example of the flexibility of this sequence assembly engine.
2:35 Partnering Company #2: Sponsored by
RNAseq as a Tool to Investigate the Genetic Risk for Alcoholism
Richard A. Radcliffe, Ph.D., Associate Professor of Pharmacology, Department of Pharmaceutical Sciences, Anschutz Medical Campus, University of Colorado Denver
Acute sensitivity to alcohol is thought to be an important genetically-mediated risk factor for the development of alcoholism. Using RNAseq methods, gene expression and coding region differences were examined from the brains of mouse lines that were selectively bred for extreme differences in acute alcohol sensitivity. The results provide insights into the genes, pathways, and networks that contribute to individual differences in risk for developing alcoholism.
Sponsored by
3:05 Sponsored Presentation
A Novel Web 2.0 Solution to Address the Sequencing Data Analysis Bottleneck
Andreas Sundquist, Ph.D., Co-Founder and CEO, DNAnexus
DNA sequencing output has completely outstripped the pace of Moore's law, making a data analysis bottleneck inevitable. At the same time, sequencing costs have plummeted, bringing large-scale sequencing projects within the grasp of more institutions. New, easily accessible informatics solutions are needed to enable researchers to fully harness the advancements in DNA sequencing technology for practical use. DNAnexus will present an integrated, web-based sequence analysis platform powered by cloud computing that removes the data analysis bottleneck from next-generation sequencing.
3:20 Refreshment Break, Poster and Exhibit Viewing
4:15 Mapping Genomes in 3D
Erez Lieberman-Aiden, Ph.D., Fellow, Harvard Society of Fellows, Harvard University
This presentation describes Hi-C, a novel technology for probing the three-dimensional architecture of whole genomes by coupling proximity-based ligation with massively parallel sequencing. We used Hi-C to construct spatial proximity maps of the human genome at a resolution of 1Mb. These maps confirm the presence of chromosome territories and the spatial proximity of small, gene-rich chromosomes. We identified an additional level of genome organization that is characterized by the spatial segregation of open and closed chromatin to form two genome-wide compartments. Our results demonstrate the power of Hi-C to map the dynamic conformations of whole genomes.
4:45 ChIP-seq: Data Analysis and Applications to Epigenetics
Peter Park, Ph.D., Assistant Professor, Pediatrics, Harvard Medical School
ChIP-seq combines chromatin immunoprecipitation (ChIP) with next-generation sequencing to identify protein-DNA interactions and chromatin modifications genome-wide. A number of practical issues in analysis of ChIP-seq data will be discussed, including experimental design, detection of binding sites, and determination of whether a sufficient depth of sequencing has been achieved. Application of ChIP-seq to the study of X-chromosome dosage compensation and stem cell differentiation will be described.
5:15 Technology SpotlightSponsored by
Talk title and speaker to be announced
5:30 Close of Day and Short Course Registration
6:00-9:00 Dinner Short Course*
SC3: Cloud Computing
Click here for short course details
*Separate Registration Required
Day 1 | Day 2 | Day 3 | Short Courses
Companion Meeting: Evolution of Next-Generation Sequencing