Day 2

Friday, October 7

7:30 Breakfast Technology Workshop 
(sponsorship available)

9:30 Chair's Remarks
Dr. Sridar Chittur, Director, Microarray and Flow Cytometry Core, Center for Functional Genomics, Suny Albany

9:35 Monoclonal Antibodies Binding onto Microarray Printed Slides and Affymetrix Chips Reveal Imperfections in the Synthesis of DNA 
Ms. Lise Barley-Maloney, Project Manager, Biochemistry, VeriQ: Project Manager, Biochemistry Department, North Carolina State University
Anti Benzoyl (BZ), Anti 4, 4-Dimethoxytrityl (DMT) and Anti Isobutyryl (IBU) monoclonal antibodies are used in a Chemiluminescent assay to detect remaining protective groups on synthetic DNA. These novel reagents and methods developed can be used to screen defective Oligonucleotides before printing or isolate areas on chips which may be defective. Microarray printed slides and Affymetrix Chips probed with these monoclonal antibodies detect incomplete deprotection of Oligonucleotides on the spots and on the chip. Incompletely deprotected Oligos interrupt hybridization of probe to the DNA and report false results. Utilizing the Chemiluminescent assay would help researchers avoid false results by avoiding protected Oligos.

10:05 Connecting Expression Profiles and Quantitative PCR Measurements
Mr. Herbert Auer, Director of the Affymetrix Core, Columbus Children's Research Institute 
Microarray analysis of genome wide expression profiles faces multiple challenges. 1. Internal consistency of multiple measurements within the same array would be favorable: we show that quantitative accuracy can be improved dramatically without generating false positive measurements, if the most sensitive probes within an array are used. Probe sensitivity not only is a sequence-inherent characteristic, but depends on the chemical nature of the hybridization material, namely DNA or RNA. 2. Reproducibility of array results generated at other facilities would be advantageous: robust algorithms provide reproducibility of results. 3. Confirmability by alternative techniques like qPCR should be granted; qPCR results are strictly dependent on normalization relative to housekeeping genes, and microarray data identify adequate non-regulated transcripts. Data supporting all three concepts will be provided.

10:35 Morning Coffee, Poster and Exhibit Viewing

11:15 Benchmark Program to Identify Variance in Microarray Data
Dr. Laura Reid, Director, Research & Development, Expression Analysis
Comparing data between laboratories and over time is one method of validating microarray data. We organized a testing program where the same six RNA samples were processed in 15 different Affymetrix laboratories at four intervals in a year. The results demonstrate the effects of different lab protocols and analysis algorithms, the comparability of results between labs and the consistency of results generated in the same lab at different times. They also identify the relative contributions of different parameters on variation in microarray results. This benchmarking program can be used to monitor lab quality and provide confidence in microarray results.

11:45 Panel Discussion with Morning Speakers

12:15 Lunch On Your Own
(Technology Workshops Available)

1:30 Chair's Remarks
Dr. Eric R. Fedyk, Senior Scientist II, Discovery Technologies, Millenium Pharmaceuticals Inc. 

1:35 Global Gene Expression Analysis in Drug Discovery and Development
Dr. Hans Winkler, Senior Director, Functional Genomics, Johnson & Johnson Pharmaceutical Research & Development 
The analysis of global gene expression provides valuable insights into the dynamic reactions of biological systems to external stimuli. A primary goal in this regard is to be able to deduce from a change in global gene expression which signaling pathways were influenced by a certain stimulus. Such a capability can be applied from Target Identification to late Clinical Development.

2:05 Incorporating Data from Multiple Labs to Develop Prognostic Predictors
Dr. Kevin Dobbin, Mathematical Statistician, BRB and CDP, National Cancer Institute
Numerous studies have discovered apparent associations between gene expression signatures and disease outcome. In order to move these promising findings towards the clinic, they need to be confirmed on larger datasets that allow their robustness and reliability to be better evaluated. Such confirmation studies require large sample sizes, and lead to multi-institutional collaborations, in which microarray data generated in different labs must be combined together. But are data from different labs comparable? We present the results of the largest study of lab-to-lab variation of oligonucleotide microarrays to date, including a breakdown of the variation attributable to tissue cutting, RNA extraction, and fragmentation-labeling-hybridization-scanning. In addition, we present preliminary findings from a not-yet-published "signature confirmation" study of ~600 patients designed to confirm the existence of prognostic signatures in lung carcinoma.

2:35 Breaking the Code of Diagnostic Gene Expression Signatures in Cancer
Dr. Zoltan Szallasi, Assistant Professor, Children's Hospital Informatics Program, Harvard Medical School 
Microarray-based gene expression patterns of limited consistency have been proposed to serve as diagnostic markers in cancer for metastatic ability, patient survival, or resistance to chemotherapy. However, in most cases these gene expression patterns cannot be associated with or explained by current biological knowledge. In other words, it is not clear what the "signature" pattern actually reflects. We will present evidence that diagnostic gene expression patterns in cancer often reflect significant levels of cross-hybridization and/or chromosome level aberrations. Therefore, although the diagnostic gene expression patterns may truly reflect a certain state of the cell, they have limited association with the specific genes involved. Most importantly, however, analytical strategies based on the above assumptions show remarkably robust cancer classification results.

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

3:30 Microarray Validation in the Context of Large-Scale Epidemiologic Studies for Biomarker Discovery
Dr. Mangalath Rajeevan, Research Microbiologist, National Center for Infectious Diseases/Division of Viral and Rickettsial Diseases, Centers for Disease Control and Prevention (CDC)
Microarrays enable large-scale gene expression measurements but the reliability of microarray results is uncertain. Methods to validate microarray results remain a challenge. A CDC Molecular epidemiology program led the early efforts to validate microarray results by quantitative RT-PCR as part of discovery and validation of biomarkers for chronic fatigue syndrome and cervical cancer. This presentation will provide an overview of challenges facing microarray validation and gene expression measurements by quantitative RT-PCR, housekeeping genes and normalization, and RNA amplification in the context of large-scale epidemiologic studies.

4:00 Role of Inflammation in Heightened Seizure Susceptibility After Early-Life Seizures: Quantitative and Biological Validation of Microarray Data
Dr. Sookyong Koh, Assistant Professor, Department of Pediatrics, Northwestern University, Feingberg School of Medicine 
Accumulating evidence suggests a functional role for inflammation as a cause of neuronal excitability and degeneration. We have developed a "two-hit" rat seizure model demonstrating that an early life seizure, without causing overt cellular injury, increases susceptibility to seizures and to seizure-induced neuronal injury in adulthood. To elucidate the mechanism linking seizures in the developing brain to later epilepsy, we used high-density oligonucleotide gene arrays and characterized the molecular cascades occurring after early life seizures. Kainic acid seizures were associated with a rapid and transient increase of genes coding for classical inflammatory mediators, complements and cytokines. Real-time quantitative reverse transcriptase-polymerase chain reactions (qRT-PCR) were performed to confirm microarray data, and therapeutic intervention as well as inhibitor of microglia activation were used to validate the significance of microarray results.

4:30 Global Gene Expression Profiling During Acute Viral Infections
Dr. Irene Bosch, Assistant Professor of Medicine, Center for Infectious Disease and Vaccine Research, University of Massachusetts Medical School
Analysis of gene expression in infected circulating blood cells using high density DNA arrays (Affymetrix) and RT-PCR high throughput microfluidic cards (Applied Biosystems) for in vitro and in vivo conditions have been used to discover key factors involved in acute disease of hemorrhagic fevers. In vitro dengue virus infection of human cells was used to apply tools to discover disease markers and disease pathology mechanisms. Tested in clinical samples by qRT-PCR and using protein detection systems during acute phase of disease, we have confirmed and validated these new infection markers. Overall, the techniques described here can be applied outside the infectious diseases field to study other inflammatory diseases including diabetes type II, atherosclerosis, rheumatoid arthritis and asthma among others. 

5:00 Panel Discussion with Afternoon Speakers

5:30 Close of Conference