Mentoring Vision Research in Oklahoma
COBRE
Department of Ophthalmology
University of Oklahoma Health Sciences Center
HOME | PI | PJIs | CORES | MENTORS | IAC | EAC | SPOTLIGHT | SEMINARS | CALENDAR | AFFILIATES

Proteomics/Bioinformatics

Image Acquisition and Production

Animal Core

Microinjection

Microarray

Molecular Biology
MICROARRAY CORE

A stated goal of the Human Genome Project is cataloging all human cDNA sequences. Coordinate with the completion of that project has been the evolution of a science, termed "functional genomics", whose purpose is to bring a biologic understanding to this catalogue and, to determine the totality of gene and gene product function in both physiologic and pathophysiologic states. The first effective, rapid, and nearly universally applicable tool of functional genomics to evolve that can comprehensively define molecular mechanisms is the DNA microarray. The advent of DNA microarray technology has been heralded as the beginning of a new epoch in biomedical research principally because this technology is easily integrated both operationally and, more importantly, conceptually into ongoing research programs. Any molecular or cellular lab can now define essentially most of the genes that are likely to be relevant to the processes they study using standard DNA hybridization techniques on a standard microscope slide coated with thousands to tens of thousands of individual gene-specific probes. This technology, therefore, brings the study of gene expression to a state where it can be routinely employed. The OMRF microarray research facility was founded to produce high-density human and mouse microarrays and provide the bioinformatics expertise necessary to derive biomedically-relevant data from array experiments for core users. Scientists who work at the facility and those who utilize the services of the facility forge close intellectual collaborations such that the most useful approaches for conducting microarray experiments and deriving biomedically-relevant data from these experiments are carried forth. By achieving these goals the facility has grown to be a functional genomics research hub for academic institutions and biotechnology companies on a state-wide basis.

Microarray Production Technology

The rapid pace of discovery within the human and mouse genome projects fuels technological development of microarrays. Continual capital investment by the OMRF is maintained such that a state-of-the art arrays can be produced within this facility. Initial production focused on cDNA-based microarrays. We are the first academic array production lab to develop gene array analytical software specifically for quality control assessment of array production and validation of data quality (manuscript in preparation). Utilizing this software we have demonstrated that a majority of the clones used to produce cDNA-based arrays in academic labs do not provide reproducible measurements of expression levels. It has been shown that a majority of these cDNAs contain significant areas capable of intra- and intermolecular cross-hybridization and therefore cannot be used as microarray probes. By spotting long oligomeric probes (50-70-mers) derived from relatively unique regions within genes we can avoid many of the problems with cross-hybridization occurring with the current generation of cDNA-based arrays and greatly enhance the quality of academically-produced arrays. Based on these considerations, our facility has been developing next-generation oligo-based microarray production and hybridization technology. Our core facility scientists have been collaborating with the leading commercial providers of oligos (Operon-Qiagen) and labeling technology (Clontech/Stratagene/PE-NEN/Genesphere) to enhance the dynamic range and signal-to-noise ratios of oligo-based arrays. We currently focus the bulk of our production on such oligo-based genome-scale glass microarrays for mouse and human.

High-density Mouse and Human arrays

Mouse microarrays arrays are produced using a commercially available library of 70 base pair long DNA oligos (70-mers, Qiagen/Operon Technologies). The oligos were derived from the functionally defined genes present in the UniGene database (http://www.ncbi.nlm.nih.gov/UniGene/). Each oligo corresponds to a unique region within or proximal to the 3'end of a given gene. Length and sequence specificity have therefore been optimized, reducing or eliminating the cross-hybridization problems encountered with cDNA-based arrays. The mouse oligo set contains 70-mer oligos representing 13,443 mouse genes. The library utilized for array production is therefore among the most comprehensive commercially available set of array-ready functionally defined mouse genes with approximately 1/2 of the estimated number of genes in the genome, and all known human mRNAs present within the Unigene database represented. This set will therefore allow us to comprehensively identify key differences in experimental and controls samples. A complete listing of the genes can be seen at the following web address: http://www.operon.com/arrays/humangenome.prn. Similarly, human arrays are produced using a commercially available library of 70 base pair long DNA oligos (70-mers, Qiagen/Operon Technologies). The oligos were derived from the functionally defined genes present in the UniGene database (http://www.ncbi.nlm.nih.gov/UniGene/). The human genomic oligo set contains 70-mer oligos representing 16,659 human genes. The library utilized for array production is therefore also among the most comprehensive commercially available set of array-ready functionally defined human genes. A complete listing of the 16,659 genes can be seen at the following web address: http://www.operon.com/arrays/humangenome.prn.

Custom Arrays

To satisfy the specific needs of individual investigators, we construct custom arrays using DNA oligos, PCR products, cDNA clones, or proteins and peptides selected for specific research goals. In addition, we have collaborated on tissue-specific gene discovery projects using suppressive subtractive hybridization and arrayed the DNA PCR products derived from this method for further characterization.

Bioinformatics

Data normalization is performed in the bioinformatics section of the OMRF Microarray Core Facility using biostatistics software developed by the OMRF microarray core facility bioinformatician Dr. Igor Dozmorov. The approach, denoted the "Associative method" associates variation in experimental gene expression to a common standard derived from a family of low variability genes derived from control experiments. The associative method enhances the sensitivity of analysis greater than previous modifications of the T-test and increases the number of differentially expressed genes identified without significantly increasing the misidentification of false positives.

The method begins by adjusting for technical variation among nominally replicate biological samples utilizing the normal distribution of data from nonexpressed, or background, gene expression levels. Interarray signal variation derived from technical sources is further corrected, without significantly impacting on true biologic variation, using linear robust regression analysis in a manner that is essentially undistorted by expressed genes. To identify and classify differentially expressed genes, results from standard paired statistical analyses of normalized data are compared with those from a novel statistical test, denoted an associative T-test. This method supplements the standard procedure of multiple paired comparisons by associating the expression level of each gene in the experimental group with a family of similarly and stably expressed genes in the control group. Once identified, differentially expressed genes are clustered according to expression levels and function using the Jaguar Software Suite (GMS) such that pathways relevant to the processes understudy can be delineated.

Personnel

The Oklahoma Medical Research Foundation has made a major financial commitment to establish a functional genomics resource center at the OMRF. Dr. Michael Centola from the Genetics Section of the National Institute of Arthritis Musculoskeletal and Skin Diseases, NIH intramural program was recruited to direct this effort. Dr. Centola's previous studies have focused on inflammatory disease genomics and he has participated in the cloning of the human disease genes causing familial Mediterranean fever (FMF), TNF receptor-associated periodic fever syndrome (TRAPS), and Non-type I cystinuria. Additionally, Dr. Centola's work on gene expression profiling in atopic responses was the first published in the field. The OMRF has also been fortunate in the successful recruitment of Dr. Igor Dozmorov as the coordinator of bioinformatics and software development at the facility. Dr. Dozmorov, who completed doctorate degrees in both physics and immunology, is a leader in the field of microarray biostatistics, and has extensive experience with both biorobotics hardware and software engineering. In addition, two senior-level scientists, Drs. Bart Frank and Craig Cadwell, provide the molecular biology expertise required to successfully maintain cutting-edge technology development in this fast-paced arena. The OMRF microarray research facility also employs 2 technicians, and one doctoral and one masters student.

Equipment

The facility is equipped with biorobotic liquid handing equipment built around the Beckman Biomek 2000 platform. The production stream consists of two large capacity Beckman Stacker Carousels and a four head 364-well MJ Research Tetrad PCR machine. Barcodes are used for tracking clone libraries in the production stream and allow for automated picking and transfer of clones to generate array-ready sublibraries. This production stream maximizes efficiency and minimizes potential human error. All steps in the stream are controlled using in-house produced programs written in Beckman "Bioscript" programming language. Custom arrays are printed using an Affymetrix 417 microarrayer and high-density arrays are printed with a GeneMachines Omnigrid arrayer. Scanning is performed with an Affymetrix 428 scanner. This biorobotics production platform is the direct result of the combined efforts of Dr. Centola and facility personnel, to establish a high-throughput production infrastructure to serve the needs of the Oklahoma academic community state-wide.