Faculty

Molecular life science has been rocked yet again by a major paradigm shift. Veterans of the last revolution in biology that are able to transition to the new quantitative, integrative, and systematic research methods will be able to ride the new wave of excitement and energy sweeping through research labs today.

Spurred by the development of high-throughput sequencing and robotic drug screening methods, high-throughput parallel technologies are key to producing the large quantitative data sets serving as a major input to the systematic analysis of living systems dubbed systems biology. Analysis of these large data sets, often integrated with genome sequences and annotations, requires continued development of new computational algorithms and concepts. Application of current algorithms and development of new computational approaches is best managed by individuals with deep understanding of both the molecular life sciences and computational sciences. My research program has recently transitioned to this new scientific paradigm from reductionist approaches focused on signal transduction in cancer. The research projects currently in my lab integrate the rather different fields of biology and computational science and produce an ideal environment for individuals seeking training in both areas.

DNA microarray technology is the best example of a biotechnology that provides large data sets for systematic analysis. Unfortunately, it has become clear that without better understanding of the molecular events that regulate gene expression, the full value of microarray expression data cannot be realized. Regulation of gene expression is complex, but it is clear that its major component is the regulation of transcription by proteins that bind to DNA. The complex interplay between these proteins is extremely poorly understood even though those in the largest group of proteins, transcription factors (TFs), bind to DNA in a sequence-specific manner. The research projects in my lab concentrate on developing new technologies, both "wet bench" and computational, for furthering our understanding of transcriptional regulation by transcription factors. This research also has some potential to be applied beyond this focal area to other aspects of gene expression regulation, including chromatin remodeling, RNAi, and alternative splicing.

High throughput DNA sequencing used in genome sequencing remains too expensive and complex for large-scale quantitative analysis of complex mixtures (libraries) generated from certain analytical procedures, such as CAST and ChIP. Although end-tags from these can be oligomerized for serial sequencing as in SAGE, these methods are difficult and not suitable for systematic experimentation. A promising alternative is to sequence many DNAs in parallel using DNA polony technology recently developed by Mitra in Church's lab at Harvard. Polonies are small regions of DNA amplified from a single DNA by PCR. Polony DNA remains covalently localized within regions of a thin polyacrylamide gel on a microscope slide. Sequencing polony DNA has not yet been developed into a practical methodology. This project involves developing a completely new way to sequence polonies that should be very easy as well as inexpensive. This project uses advanced oligoDNA biochemistry and molecular biology methods to accomplish the practical sequencing. Also, optical and electronic technologies are being applied to develop an inexpensive and rapid detection system for the fluorescent dyes used. New data analysis programs are needed for image analysis. Further improvements in this method may be possible through new nucleotide and oligonucleotide chemistry, which represent opportunities for collaboration with interested chemists.

CAST is a PCR-based method developed from SELEX for creating a population of DNAs that are enriched for binding to a sequence-specific DNA binding protein (transcription factor, TF). Starting from a completely random synthetic oligoDNA library, the enriched population is sampled by traditional sequencing of some of the oligoDNAs and the aligned sequences are used to produce a "consensus binding" sequence or a position-specific scoring matrix (PSSM). PSSMs can be used to scan genome sequences for potential binding sites. Naively applied, this finds too many candidate sites. This is due in part to the low accuracy of the PSSM from inadequate sampling of the sequences enriched by CAST, a problem that would be solved by the polony sequencing technology being developed in the first project. It is clear that the concentration of TFs in the cell nucleus is limiting for binding, so the biologically relevant sites are those that are bound with the highest affinity, which varies with the exact sequence. Furthermore TF binding needs to be conceptualized not as a Boolean quantity (on or off), but as continuously variable quantity. New mathematical ways of more accurately modeling binding of TFs to sequences need to be devised for using CAST data to annotate the genome, but this is only the first step of this project. The ultimate goal is to use a novel CAST-inspired method to analyze all the active TFs found in the cell nucleus in a single experiment. Success of this will depend on an inexpensive and rapid DNA sequencing method such as polony sequencing though DNA microarrays may eventually be a feasible alternative.

The third project focuses on a single recently discovered bHLH transcription factor, Mist1, that is essential for normal exocrine pancreatic development. In collaboration with Dr. Konieczny, Purdue Univ. Dept. Biol. Sciences, this project involves basic bioinformatics as well as developing new computational algorithms to model the interactions among known TFs that produce transcriptional "regulatory modules". The current strategy uses model-based clustering methods. In addition, we hope to apply the methods developed in the first two projects as well as new algorithms for identifying biologically significant Mist1 binding sites in the mouse and human genomes. This project also provides an opportunity to extend polony-sequencing technology to DNA obtained by chromosome immunoprecipitation (ChIP) as a systematic way of identifying meaningful Mist1 binding sites.

K. P. Kesavan, C. C. Isaacson, C. L. Ashendel, R. L. Geahlen, and M. L. Harrison "Characterization of the in Vivo Sites of Serine Phosphorylation on Lck Identifying Serine 59 as a Site of Mitotic Phosphorylation," J. Biol. Chem. 277, 14666-14673, (2002).
 
C. Tarn, S. Lee, Y. Hu, C. Ashendel, and O.M. Andrisani "Hepatitis B virus X protein differentially activates RAS-RAF-MAPK and JNK pathways in X-transforming versus non-transforming AML12 hepatocytes" J Biol Chem 276: 34671-80, (2001).
 
W.-C. Xu, Q. Zhou, C.L. Ashendel, C.-t. Chang, and C.-j. Chang, "Novel protein kinase C inhibitors: Synthesis and PKC inhibition of beta-substituted polythiophene derivatives," Bioorg. Med. Chem. Lett. 9, 2279-2282 (1999).
 
M.B. Ramocki, S.E. Johnson, M.A. White, C.L. Ashendel, S.F. Konieczny, and E.J. Taparowsky, "Signaling through MAP kinase and Rac/Rho does not duplicate the effects of activated Ras on skeletal myogenesis," Mol. Cell. Biol. 17, 3547-3555 (1997).