|
Overview
| Day 1
| Day 2
| Day 3
| Download
Brochure | Breakout
Discussions
FRIDAY, FEBRUARY 27
INTEGRATING GENOMIC DATA
8:30 Chairperson’s Remarks
8:35 The Genomic Data Pipeline:
Collecting, Cleaning, Analyzing, Integrating, Sharing
Jeanette Papp, Ph.D., Associate Professor, Human Genetics,
Director of GenoSeq Core Laboratory, University of California
Los Angeles
A tour of the Genomic Information Superhighway - from historical
perspective to current challenges and solutions, with a glimpse
into the future. The presentation will cover the evolution of
genomic data management and integration systems. We will
describe some of the challenges and approaches, with examples
from Mendel Enterprise, our in-house solution for data
collection and management, data merging and integration,
statistical analysis, and data sharing.
9:05 Surviving the Data Deluge: Informatics for Next Generation Sequencing
Toby Bloom, Ph.D, Director of Informatics, Genome Sequencing Platform, The Broad Institute
9:35 Integrating Public Genomics Data into Pharmaceutical R&D
Hans-Martin Will, Ph.D., Senior Director, Genomics R&D, Rosetta Biosoftware
Over the past few years, more and more comprehensive genomics data sets have been generated by academic and publicly-funded consortia and made accessible-notable examples are the Framingham Heart Study and the data released by the Wellcome Trust Case Control Consortium. This abundance of data is creating a new reality for pharmaceutical R&D, whereby a large number of data sets relevant to internal R&D projects are generated from publicly-funded, academic and governmental organizations. With this insight, many organizations are in the process of devising strategies for making best use of these data, and implementing approaches for effectively bringing the data sets in-house and integrating them into the context of their on-going scientific research efforts. In this talk, we will discuss the various types of challenges pharmaceutical R&D organizations are facing when bringing in public data sets. Starting with mundane data formatting problems, unknown data quality and varying taxonomies, we will conclude our discussion with a brief survey of approaches and opportunities for mining these data.
10:05 Search Strategies for Correlating Combined Public and Internal Large-Scale Studies
Ilya Kupershmidt, Cofounder and Vice President Products, NextBio
NextBio specializes in the discovery and testing of hypotheses within the growing body of the world’s high-throughput data. To enable real time search NextBio pre-computes billions of findings within highly heterogeneous data. These findings are based on the global meta-analysis across thousands of genomic, genotyping and other large-scale “omics” studies. |
Sponsored by
 |
10:20 Coffee Break
11:00 Platform for the FDA Genetic
Data Submission and Review Process
Weida Tong, Ph.D., Director, Center for Toxicoinformatics,
National Center for Toxicological Research, U.S. Food and Drug
Administration
Rapidly developing technologies for genetic analysis are driving
the emergence of the new research fields of personalized
medicine and targeted therapeutics. In order to guide effective
development and regulation of pharmacogenomics and medical
devices resulting from this research, the expertise, tools, and
processes for utilizing genetic data are needed in the FDA. To
this end, the FDA’s Critical Path Initiative created the
Voluntary eXploratory Data Submission (VXDS) mechanism to
provide a collaborative environment in which the research
community, sponsors and the FDA can work together on data
management, analysis and interpretation outside normal
regulatory interactions. With the increasing number of
submissions based on genetic data, a parallel informatics
platform has been conceived. This talk will introduce this new
initiative, SNPTrack, being undertaken by the NCTR/FDA in
collaboration with Rosetta Biosoftware, for development of FDA
regulatory capability for analyzing VXDS and formal submissions
of genotyping data. The goal is a system which enables FDA
reviewers to reconstruct sponsor analysis of genetic variation
data, explore alternative analysis methodologies and respond to
the sponsor with the agency’s understanding and
recommendations. It is envisioned that the outcome of this
initiative will drive adoption of industry best practices and
standards for formal data submissions.
11:30 Unifying
Disparate Information Contextually to Orchestrate R&D
Processes
Shree Nath, Ph.D., Director, Product Technology,
PointCross Inc.
Pharmaceuticals and Biotech R&D processes involve
high levels of tacit interactions around knowledge and
insights on the basis of which high risk/high reward
decisions have to be made. There are no mechanisms to
readily contextualize the large information base of in
vitro and in situ study data, and even larger quantities
of "-omics" datasets from biotechnology
research, along with rich metadata from collaboration,
analysis, reporting, and submissions. This talk will cover
the need to contextually unify, classify, and relate
disparate unstructured (emails, meeting notes, decisions,
and documents) and structured information (from trial
data, LIMS, and data warehouses). Information
contextualization allows scientists to readily conduct
meta-analysis; and enables search, orienteering and
semantically guided navigation for hidden nuggets of
insights within layers of metadata that in turn can lead
to new discoveries and concepts well beyond the limits of
simple structured data mining. Contexts become meaningful
representations of every aspect of R&D, regulatory
compliance and collaboration, while providing essential
controls to assure information security and IP protection.
We will present a practical ontology-based platform on
which these capabilities are being delivered not only to
sponsor companies, but also extended to multi-party
R&D partner and CRO environments.
|
Sponsored by
 |
12:00 pm Luncheon Presentation (Sponsorship Available) or Lunch on Your Own
DATA INTEGRATION AND MANAGEMENT FOR EARLY AND LATE CHEMISTRY
1:00 Chairperson’s Remarks
1:05 Lilly’s Transition from Paper to Electronic Lab Notebooks
Jeffrey D. Christoffersen, Product R&D, Eli Lilly & Co.
Lilly began evaluating electronic lab notebooks in 2003. By 2005, the Process Chemistry group had fully implemented a paper-less system.
The planned deployment of a single electronic lab notebook solution across the entire company will be described, as well as various challenges related to quality and legal concerns and end-user uptake.
The benefits to Lilly derived from a shift from paper to electronic lab notebooks will be shared, in addition to our recent efforts to transition our third party partners to electronic lab notebooks.
1:35 Knowledge-based expert systems,
(quantitative) structure activity relationship tools and
modeling approaches in preclinical safety studies
Wolfgang Muster, Ph.D., Head of in silico and in vitro
Screens, F. Hoffmann-La Roche Ltd.
This presentation will illustrate how computation tools,
deployed in the early drug development process, can help
predicting toxicity and thereby optimize and select the best
clinical candidates to move forward. A focus will be on in
silico prediction methods roughly classified into so-called “expert
systems” and “data driven systems”.
2:05 Compound Cytotoxicity Profiling
and Characterization of Toxicity Mechanisms Using Quantitative
High-Throughput Screening
Ruili Huang, Ph.D., Research Scientist, Informatics, NIH
Chemical Genomics Center
A large library of compounds previously tested in traditional
toxicological assays were profiled for cytotoxicity using
quantitative high-throughput screening (qHTS). Combining data
generated from these assays we designed a broader array of in
vitro cell-based assays in order to screen large sets of
compounds. Such assays should also elucidate mechanism of
toxicity, prioritize compounds for further toxicological
evaluation and predict in vivo biological response. As a proof
of principle, we applied an unsupervised clustering method to
this data set to identify mechanisms of action and evaluated the
performance of this method by comparing the results against
literature annotations of compound mechanisms.
2:35 Integration of Chemical
Genomics and Structural Biology Informatics: Novel Insights into
the Kinase Gene Family?
Stephan Schürer, Ph.D., Department of Pharmacology, Miller
School of Medicine & Center for Computational Science,
University of Miami
We developed integrated data analysis pipelines to quantify
similarity relationships among the protein kinase complement of
the human genome (the “kinome”) from different perspectives:
domain sequences, small molecule kinase activity data, and
structure-based physicochemical properties of the ATP binding
sites. While we gain insight into differences and synergies of
chemogenomics- and structural biology-informatics based
approaches to identify and utilize gene-family-wide similarity
relationships we also investigate the differences of active and
inactive kinase conformations in the same context. Integrating
large chemical genomics data sets and high-quality experimental
and modeled structures covering almost the entire Kinome we
developed discovery pipelines allowing receptor-site information
and small molecule activity data from entire target families to
be used in the rational design of compounds with desirable
selectivity profiles.
3:05 Close of Conference
Overview
| Day 1
| Day 2
| Day 3
| Download
Brochure | Breakout
Discussions
For questions or suggestions about the meeting, please contact:
Edel O'Regan
Cambridge Healthtech Institute
250 First Avenue, Suite #300
Needham, MA 02494
Tel: 781-972-5423
Fax: 781-972-5425
email: eoregan@healthtech.com
For sales
information, contact:
Carol Dinerstein
Tel: 781-972-5471
email: Dinerstein@healthtech.com
OR
Jon Stroup
Tel: 781-972-5483
email: jstroup@healthtech.com
|
2009 Final
Brochure Download
|
