From Hype through Frustration to the “Gene-meter” – a Long Path of the Microarray Technology
Oral Health Science, University of Washington
January 22, 2015
Foege N130A, Wallace H. Coulter Seminar Room
The microarray used to be a fashionable technology of the 90s. Personalized medicine, bioterrorism and cancer diagnostics – to name a few – were the applications where microarrays were expected to excel. Our research group was one of the first to question if the bioinformatics was sufficient to design a microarray with expected sensitivity and specificity. In reality, the hybridization of nucleic acids on the surface was a major unpredictable factor confounding the bioinformatics approaches based on sequence matches and mismatches. To the contrary, the belief in bioinformatics was so strong that our results were dismissed by many. Several groups (including ours) were trying to come up with an accurate physicochemical model for microarray response in attempts to make microarrays as accurate as other analytical instruments (e.g., pH meter). Other groups took a path of complicated normalization and more bioinformatics. Recently, all prominent physical chemists of the microarray field came together and concluded that we do not understand what is taking place on the microarray surface… Not only the microarray surface, but the beads of the next gen sequencing pose the same issue! Finally our group took a fundamentally different approach – to calibrate every probe on the microarray using a simple dilution series of a biological sample. A calibration curve is associated with every microarray probe, which allows vetting probes with undesired response. The calibration curve also enables direct calculation of gene expression or copy number variation from the signal intensity. The calibration approach greatly simplifies normalization and supersedes many convoluted filtering algorithms. The talk will provide a case in point demonstration of the utility of the calibrated microarrays for copy number variation in Mus musculus as well as postmortem gene expression analysis of Mus musculus and Danio rerio.
Currently Dr. Pozhitkov is a senior fellow at the Department of Oral Health, UW. In 1998, he graduated with M.Sc. in chemistry from the Moscow State University (Russia) and in 2004 he earned his PhD in genetics from Albertus Magnus Universität zu Köln in Germany. Since 2004, Dr. Pozhitkov’s research focus has been multiplexed detection and quantification of specific nucleic acids (mRNA, 16S rRNA). He worked at the Department of Civil Engineering, UW; Gulf Coast research Laboratory, USM and the Max Planck Institute for Evolutionary Biology, Germany. Dr. Pozhitkov views himself as a multidisciplinary researcher. His projects involve surface chemistry, electrochemistry, electrical engineering, bioinformatics and robotics. Recently Dr. Pozhitkov received funding from the International Team for Implantology (Switzerland) to understand the microbial-assisted corrosion of dental implants. An additional area of his research interests is animal postmortem gene expression dynamics and microbial colonization. Dr. Pozhitkov published 30+ papers in peer-reviewed journals.