FRIDAY, AUGUST 17

8:45 Chairperson’s Remarks

8:50 Assessing the Accuracy of Copy Number Variation Data from Different Microarray Platforms
Dalila Pinto, Ph.D., Postdoc Research Fellow, Genetics and Genomic Biology, The Hospital for Sick Children, Toronto 
Detection of copy number variants (CNVs) on a genome-wide scale in a single experiment is now possible using different approaches, such as array comparative genomic hybridization (CGH), or comparative intensity analysis using single nucleotide polymorphism (SNP) genotyping platforms. A comprehensive comparison of the performance of available microarray platforms and analysis methods depends on the availability of a gold standard – comprised of regions with known CNVs as well as unaffected/invariant regions – that can be used for benchmarking, allowing for an assessment of sensitivity and specificity of each method. Here, we assessed currently available CNV calling algorithms, using a set of reference samples and spike-in controls as gold standards. We will report on the overlap between variants called by different platforms and analysis tools, and discuss the strengths and weaknesses of the various approaches, as well as issues with the interpretation of CNV data. We have also established the Database of Genomic Variants (http://projects.tcag.ca/variation/), to provide a curated comprehensive repository of previously identified CNV regions and to facilitate study design and data interpretation.

9:20 Microarray-Based Genomic Hybridization and DNA Copy Number Alterations in Leukemic Biomarker Surveys
Joëlle Tchinda, Ph.D., Senior Scientist, Department of Pathology, Brigham and Women's Hospital 
Microarray-based comparative genomic hybridization (array CGH) allows the detection of DNA copy number alterations in a genome-wide fashion. Using high resolution array CGH, more copy number changes are often detected in leukemias compared to solid tumors, which are more frequently associated with specific chromosomal translocations and inversions. We have screened 20 DNA samples from established T-cell acute lymphoblastic leukemia (ALL) cell lines using the Agilent 44k oligonucleotide platform ­ which contains 42,920 probes with an average spatial resolution of ~35 kb. A recurrent gain of NRAS was detected in 70% cell lines. Mutations of NRAS have previously been known to be associated with colorectal cancer and thyroid carcinoma, but recurrent gain of NRAS has not been described in leukemias. A parallel study on chronic lymphocytic leukemia (CLL) showed no alteration of NRAS suggesting that this genomic imbalance may be more specific for T-cell ALL. In addition to the recurrent gain of NRAS, array CGH studies on the same T-ALL cohort also identified recurrent gain of the genes KIAA0963 and STK11 in 50% of the samples. While the function of KIAA0963 is unknown, germline mutations and deletions of STK11 (a putative tumor suppressor gene) lead to Peutz-Jeghers syndrome (PJS), which is associated with an increased incidence of malignant cancers of different organ systems. The above mentioned examples are only a subset of recurrent copy number changes detected in our studies using array CGH. Indeed, the increasing resolution of commercial array CGH platforms provides an increasing amount of data that demand better analysis software and accurate discrimination from germline alteration that are now becoming more widely identified in healthy individuals automatically and begs the query of germline copy number variations that are associated with differential disease susceptibility.

9:50 Algorithmic Considerations in Relating Copy Number Variation to Gene Expression
Beth Wilmot, M.S., Graduate Research Assistant, Molecular and Medical Genetics, Oregon Health & Science University
Gene copy number is known to affect the expression of single genes as well as those located within genomic regions altered in tumor samples. However, the impact of copy number variation (CNV) on genome-wide expression profiling studies of non-cancer tissues has not been evaluated. Affymetrix 100K Gene Mapping chips were used to examine the influence of CNV on gene expression of cognitive decline. Several CN algorithms were assessed for their ability to identify genome-wide copy number changes in normal individuals. Critical criteria for CN estimation were algorithm selection, choice of reference data set and genotyping call rate as a proxy for DNA quality. We found a minimum estimate of 22.9% differentially expressed genes possibly affected by CNV in our population.

10:20 Coffee Break, Poster and Exhibit Viewing

11:00 Expression, SNP and CGH Profiling Identifies Critical Loci in Cellular Immortalization and Senescence
Michael Tainsky, Ph.D., Barbara & Fred Erb Professor of Cancer Genetics, Department of Pathology; Director Program in Molecular Biology and Genetics, Karmanos Cancer Institute, Wayne State University School of Medicine 
Li-Fraumeni syndrome patients develop multiple types of cancers and generally at very early ages relative to the general population. Li-Fraumeni syndrome patients' cells from patients, with germline-derived mutations in p53, spontaneously form immortal cultures, which never happens to cells from normal subjects. We will report on the integration of expression profiling, SNP analysis and comparative genome hybridization data to identify critical genes, pathways, and chromosomal loci involved this precancerous process.

11:30 Panel Discussion with Morning Speakers

12:00 Lunch on Your Own or Luncheon Technology Workshop (Sponsorships Available)

1:30 Chairperson’s Remarks

1:35 To a Better Understanding of Immune Responses to HIV and HCV Infection: Combined Analysis of EXON, SNP, and 3-Prime Expression Data to Identify Significant Clinical, Genetic and Functional Annotation Relationships
Richard Lempicki, Ph.D., Senior Scientist, Head, Clinical Services Program, SAIC-Frederick, NIAID/NIH
The association of biological processes and clinical measures with patient expression, alternative splicing and genetic profiles can provide important insight into the immunopathogenesis of HIV and HCV infection. Array results derived from five independent studies of PBMC expression or SNP data from various HIV-1 and/or HCV patient cohorts (n=263 in total; HuFL: 2 studies, U133A: 2 studies, Exon: 1 study, 500K SNP: same samples as on EXON Array, 50K SNP: 1 study) will be presented. Consistent findings between the different gene expression arrays were found including: 1) significant correlations between the level of interferon stimulated genes (IFSGs) and HIV viremia (n=107; p<0.001) and between IFSG levels and failure of PEG-IFN?2b/ribavirin therapy to control HCV replication (n=56; p<0.001). Results were confirmed with real-time PCR and with a new high-throughput, multiplex alternative to real-time PCR, QuantiGene Plex (QGP) bDNA/xMAP bead assay. The QGP assay was validated by showing a strong correlation between QGP and real-time PCR data (n=170, r=0.95, p<0.001). Significant associations between expression, alternative splicing, genetic (SNP array), clinical and functional annotation data (derived from the DAVID Knowledgebase: http://david.niaid.nih.gov) will be discussed.

2:05 Global Analysis of Alternative Splicing during T-Cell Activation
Brendan J. Frey, Ph.D., Associate Professor, Electronical and Computer Engineer, Universiy of Toronto (invited)

2:35 Determining the Disease Significance of Alternative Ion Channel Splicing in Mesial Temporal Lobe Epilepsy and Alzheimers Disease 
Erin Heinzen, Ph.D., Institute for Genome Sciences and Policy, Center for Population Genomics and Pharmacogenetics, Duke University
New technology permitting the screening of alternative splice variants in microarray format has permitted the identification of widespread alterations in ion channel splicing in mesial temporal lobe epilepsy and Alzheimer's disease. The challenge now is to decipher which of these changes impact disease pathophysiology. Elucidating the genetic control mediating these alternative splicing events may provide insight into how these changes impact the disease states. 

3:05 Refreshment Break, Poster and Exhibit Viewing (Last Chance for Viewing)

3:30 Identifying the Isoforms Specific to Individual Human Tissues and How They Got There: Strategies for Alternative Splicing Microarray Profiling and Data Interpretation
John Castle, Ph.D., Research Fellow, Moleculare Informatics, Rosetta Inpharmatics, LLC, part of Merck & Co.
Like gene expression, regulation of alternative splicing, and the resultant isoforms, acts to define tissues. We compare different strategies for monitoring alternative splicing, including array design and analysis, and including trade-offs and different applications. We show results of applying these profiling strategies to treatment versus control sample pairs and to atlas datasets. Finally, we interpret these results in terms of alternative splicing regulation and the functional implications.

4:00 Putting IT Together: Correlation Between Copy Number Alterations, Loss of Heterozygosity, mRNA Expression and Epigenetic Modulation in Gliomas
Jean-Claude Zenklusen, Ph.D., Staff Scientist, Neuro-Oncology Branch, National Institutes of Health / National Cancer Institutes 
Here we present an integrative method for analysis of genomic data by combining the Copy Number/LOH data with mRNA expression obtained from the same samples. To correlate RNA expression with DNA copy number, each HG-U133_Plus_2 probe set was mapped against all the SNPs located in a window of 1 Mbp around the center of that probe set. This approach yielded a substantially reduced list of candidates than CNA/LOH or mRNA expression studies would have produced alone. However, this analysis approach did, by design, exclude genes that may be epigenetically regulated, thus lacking correlation with the genomic changes detected. To solve this shortcoming, we then focused on all probesets mapping to areas where at least 10% of samples showed LOH, and selected those for which a substantial fraction of LOH samples had an mRNA expression lower than the median expression shown by our non-tumor reference samples. In this way, we identified about 400 genes (some of them known TSGs), most of which had a clear pattern of bimodal mRNA expression distribution, while some samples having little or no expression, while others had an expression level consistent with their copy number status. Methylation and mutation analysis of these gene/sample pairs confirmed the epigenetic nature of their mRNA modulation, suggesting that the genes may indeed be novel TSGs. The wealth of genomic data produced may allow for the development of a more rational molecular classification of gliomas and serve as an important starting point in the search for new molecular therapeutic targets.

4:30 Panel Discussion with Afternoon Speakers

5:00 Close of Conference