THURSDAY, AUGUST 16

8:00 am Coffee and Conference Registration

8:45 Chairperson’s Opening Remarks

8:50 Genome-Wide Association Scans: Do not Neglect the Phenotype!
Constantin Polychronakos, M.D., Professor, Pediatrics and Human Genetics, McGill University Health Center 
The important points and lessons learned from our genome-wide scan (GWS) for genetic associations with type 2 diabetes will be outlined. We typed 400 k SNPs in two arrays covering the genome at a combined average density of <10kb with focus on transcribed sequences (Human1) or LD block tags (Hap300). These arrays were used in a first stage in which we typed 700 cases and 700 controls. The best 60 associations were fast-tracked to second stage (2.5 k cases + 2.5 k controls) and resulted in the discovery of four novel loci. A full second stage is in progress. A somewhat different two-stage design for type 1 diabetes is in progress and results should be available by the time of the meeting. The general principles of design and analysis of two-stage GWSs will be outlined with special emphasis on dealing with heterogeneity within the phenotypic trait studied. Detailed phenotyping and the study of epistatic interactions can complement each other to enhance the power of such studies and lead to individualized medicine, arguably the most desirable goal of complex-trait genetics.

9:20 Analysis of Genome-Wide Association Data
George Uhl, M.D., Ph.D., Branch Chief, Molecular Neurobiology, NIDA/NIH 
Genome-wide association (GWA) is increasingly a method of choice for positional cloning of the gene variants that predispose to complex disorders. We have used these methods to identify more and more of the allelic variants that predispose to addiction as available arrays have been able to assess SNP allele frequencies that cover more and more of the genome. Currently, although no single approach to these analyses is uniformly accepted, we have had substantial apparent success in using Monte Carlo simulation approaches that do not require assumptions about underlying distributions with supplemental secondary permutation and false discovery rate methods. We will describe use of these approaches in data from several samples of substance dependent vs controls and >600,000 SNPs.

10:05 Coffee Break, Poster and Exhibit Viewing

10:45 SNP Selection Optimization in Multiple-Ethnic Panel
Hugues Sicotte, Ph.D., Biomedical Informatics, Mayo Foundation
Custom genotyping microarrays can only carry a limited number of SNPs, so designing a custom panel involves minimizing the number of selected SNPs while maintaining the power of the panel. The procedure of reducing the SNP redundancy using SNP tagging is well established. However, these SNP tagging algotithms use the LD patterns found in well-defined populations and thus cannot be applied to multi-ethnic panels. We present here a method and associated code to combine tagging SNPS from multiple ethnic panels to minimize the number of SNPs needed to tag a set of genes by making use of the shared patterns of LD across populations. This method uses a flexible probabilistic scoring system that enables to optimize SNPs selection to create a multi-SNP panel with the best chance of success while maximizing the power of the panel (expected number of SNPs being represented by the tagging snps). The method is also able to include constraints typical of SNP design panels such as avoiding SNPs too close to each other and preferential selection of SNPs in regions of interest. Our software can also be used in single-population mode to select an optimal and constrained set of tagging SNPs from the multiple choices left by SNP tagging algorithms such as ldselect.

11:15 PPIL2 Identified as a Novel Regulator of BACE1 mRNA Levels Using siRNA Screening and Microarray Expression Analysis
Paul Hodor, Ph.D., Research Fellow, Department of Molecular Profiling, Merck & Co., Inc.
In a search for novel disease modifying agents of Alzheimer's Disease, a screen for modulators of ß-amyloid precursor protein (APP) processing was carried out, using a library of siRNAs targeting 532 predicted ubiquitin ligases. Among the positives was PPIL2, whose targeting suppressed ß-site cleavage of APP. Manipulation of cellular PPIL2 mRNA levels lead to changes in BACE1 mRNA levels in the same direction. Microarray profiles produced by gene knockdowns offered supporting evidence that BACE1 expression is downstream of PPIL2.

11:45 Panel Discussion with Morning Speakers

12:15 pm Lunch on Your Own or Luncheon Technology Workshop (Sponsorships Available)

1:30 Chairperson’s Remarks

1:35 High-Throughput Mapping of the Chromatin Structure of Human Promoters
Jun S. Song, Ph.D., Member, The Simons Center for Systems Biology, Institute for Advanced Study
It has been known that chromatin structure and epigenetic factors may play crucial roles in eukaryotic gene regulation, selectively enhancing or silencing transcriptional activities specific to cell-types. Our precise understanding of how chromatin structure exactly influences cellular processes, however, has been limited by a lack of nucleosome-positioning data in human cells. We have recently developed a high-resolution microarray approach to examine the translational positions of nucleosomes in 3692 promoters within seven human cell-types. This study presents a global view of nucleosome positioning in human promoters and provides a powerful tool for analyzing chromatin structure in development and disease.

2:05 Special Molecular Interactions in Microarray Hybridization Experiments
Li Zhang, Ph.D., Assistant Professor, Bioinformatics and Computational Biology, M.D. Anderson Cancer Center
DNA/DNA duplex formation is the basic mechanism that is used in genome tiling arrays and SNP arrays manufactured by Affymetrix. However, detailed knowledge of the physical process is still lacking. In this study, we show a free energy analysis of DNA/DNA duplex formation for these arrays based on the positional-dependent nearest-neighbor (PDNN) model, which was developed previously for describing DNA/RNA duplex formation on expression microarrays. Our results showed that the two ends of a probe contribute less to the stability of the duplexes and that there is a microarray surface effect on binding affinities. We also showed that free energy cost of a single mismatch depends on the bases adjacent to the mismatch site and obtained a comprehensive table of the cost of a single mismatch under all possible combination of adjacent bases. The mismatch costs were found to be correlated with those determined in aqueous solution. We further demonstrate that the DNA copy number estimated from the SNP array correlates negatively with the target length; this is presumably caused by inefficient PCR amplification for long fragments. These results provide important insights into the molecular mechanisms of microarray technology and have implications for microarray design and the interpretation of observed data.

2:35 GAPWM: A Genetic Algorithm Method for Optimizing a Position Weight Matrix
Leping Li, Ph.D., Investigator, Biostatistics Branch, National Institute of Environmental Health Sciences/NIH
Position weight matrices (PMWs) are simple models commonly used in motif finding algorithms to identify cis-regulatory motifs on genes. We propose a novel but simple method to improve a poorly estimated PWM using ChIP data. With increasing availability of ChIP data, our method provides an alternative for obtaining high-quality PWMs for genome-wide identification of transcription factor binding sites.

3:05 Technology Spotlight (Sponsorship Available)

3:20 Refreshment Break, Poster and Exhibit Viewing

4:00 Looking for Pathway-Level Differentiated Genes and Pathway-Level Enrichment Patterns for Underlying Biological Themes with WPS
Ming Yi, Ph.D., Program Analyst IV, Advanced Biomedical Computing Center, SAIC-Frederick, Inc./NCI-Frederick 
Various genetic alterations or external stimuli may confer impacts on a biological process, a function category, or a pathway, by influencing different genes, or even the same genes but at different levels (e.g.,expression, post-translational modification) involved in a biological process or pathway. Conventional methods for analyzing data at the individual gene level may miss the opportunity to discover the underlying biological themes if expression patterns of different genes but not the same genes in a pathway are altered within different but relevant biological systems. We are providing evidence that analyzing data at the level of functional categories including well-defined or customized pathways and GO terms may help understand the underlying biological themes at a higher level. Such a method can explore the commonality or uniqueness at the level of a pathway or biological process from different but relevant biological systems studied. This method may have potential to help integration of different types of "omic" data for the whole genome multiple-level studies including expression profiling, proteomics, and post-translational modifications at systems biology level.

4:30 Genomic Signature Association between NCI-60 Cancer Cell Lines and Clinical Tumors for Predicting Patients' Chemotherapeutic Responses
Jae Lee, Ph.D., Associate Professor, Public Health Sciences, University of Virginia
The U.S. National Cancer Institute has used a panel of 60 diverse human cancer cell lines (the NCI-60) to screen >100,000 chemical compounds for anticancer activity. However, not all important cancer types are included on the panel nor are drug responses on the panel predictive of clinical efficacy in patients. We asked, therefore, whether it would be possible to associate and extrapolate from that rich database to predict activity in cell types not included in the NCI-60 panel or, even further, clinical responses in patients with tumors. We address that challenge by developing and applying a novel algorithm "Co-eXpression ExtrapolatioN" (COXEN). COXEN effectively integrates and associates the information between the two independent expression profiling data sets of NCI-60 and clinical tumors and in vitro NCI-60 drug screening data for predicting drug sensitivity of clinical tumors. We demonstrated our COXEN approach in bladder cancer, which is not included in the NCI-60 panel, and breast cancer patients treated with commonly used chemotherapeutics. We also used this approach for in silico screening of 45,545 compounds and identified a novel agent with superior growth inhibition activity against human bladder cancer.

5:00 Panel Discussion with Afternoon Speakers

5:30 Networking Reception

6:30 Close of Day