Integrating expression-based drug response and SNP-based pharmacogenetic strategies into a single comprehensive pharmacogenomics program

Application of high-resolution capillary array electrophoresis with automated fraction collection for GeneCalling trade mark analysis of the yeast genomic DNA

Capillary array instrument was applied to transcript profiling of the yeast genomic DNA using GeneCalling trade mark chemistry. The instrument integrated a 12-capillary array for DNA separation with a replaceable sieving matrix, laser-induced fluorescence detection and an automated microfraction collector. The DNA fractions, exiting the separation capillaries, were continuously deposited in a 1536-well collection plate made of agarose gel. DNA fragments recovered from selected fractions were cloned and then sequenced. Over 80% of theoretically predicted fragments could be recovered in the collected fractions, cloned and sequenced with an average redundancy of threefold. Excellent correlation of the experimentally obtained sequences with the theoretically predicted gene fragments demonstrated the suitability of capillary array electrophoresis for micropreparative recovery of DNA fragments. This approach, useful especially for rapid DNA expression profiling of unknown genes for nonsequenced organisms, demonstrates the practical capability of the prototype multicapillary fraction collector.

Biomarker identification by feature wrappers

Gene expression studies bridge the gap between DNA information and trait information by dissecting biochemical pathways into intermediate components between genotype and phenotype. These studies open new avenues for identifying complex disease genes and biomarkers for disease diagnosis and for assessing drug efficacy and toxicity. However, the majority of analytical methods applied to gene expression data are not efficient for biomarker identification and disease diagnosis. In this paper, we propose a general framework to incorporate feature (gene) selection into pattern recognition in the process to identify biomarkers. Using this framework, we develop three feature wrappers that search through the space of feature subsets using the classification error as measure of goodness for a particular feature subset being "wrapped around": linear discriminant analysis, logistic regression, and support vector machines. To effectively carry out this computationally intensive search process, we employ sequential forward search and sequential forward floating search algorithms. To evaluate the performance of feature selection for biomarker identification we have applied the proposed methods to three data sets. The preliminary results demonstrate that very high classification accuracy can be attained by identified composite classifiers with several biomarkers.

The porcine endogenous retrovirus long terminal repeat contains a single nucleotide polymorphism that confers distinct differences in estrogen receptor binding affinity between PERV A and PERV B/C subtypes

Porcine endogenous retroviruses (PERV) have been shown to have zoonotic potential, both in vitro and in vivo. Once integrated into the host cell genome activation of the proviral genes is ultimately dependent upon transactivation of the long terminal repeat (LTR). Currently there is no direct evidence of host cell transcription factors interacting with PERV LTRs. Using comparative genomics we discovered a potentially functional single nucleotide polymorphism (SNP) within the U5 region downstream of the TATA box in the PERV LTR that distinguishes PERV A from PERV B and PERV C subtypes. We demonstrated that the SNP occurs within a potential hormone-responsive region where it has a profound effect, not only upon estrogen receptor binding but also upon the binding of other transcription factors at this site. These results suggest that differences in transcriptional regulation between PERV subtypes are subtle and, as for other retroviruses, transcription can be mediated by steroid hormone receptors.