Combinatorial and graph theoretical approach to systems biology and mol. evo.

点击翻译按钮获取中文摘要

My group continued to work on computational methods to study the dynamics of biological networks, impact of genetic variations and structural variation on gene expression, organismal phenotype and complex diseases. In particular we continued to work on methods to delineate genetic interactions underlying complex traits. We focused on epistatic interactions, that is interactions which are characterized by a non-additive/non-independent effect of two loci on a quantitative trait. In our recent publication (1) we developed a new method to predict epistatic interactions and applied it to study the interaction map in Plasmodium discovering epistatic interaction hotspots present in the genome of this organism. We also initiated studies on loci interactions underlying yeast drug resistance phenotype. We continued to work on the question on the impact of copy number variation son gene expression. Copy number variations (CNV) are a frequent type of polymorphisms and often a disease causing genetic aberration especially in cancer. Understanding the effect of copy number on gene expression is prerequisite for systematic study of the effect of such variation on the whole molecular system. It is often assumed that increased gene copy number implies increased expression of a given gene. In collaboration with Brian Oliver group we preformed experimental and computational studies of the impact gene dose on gene expression and the propagation of these effects in the fly interaction network (2). Our studies demonstrated that relation between CNV variations and gene expression is more complex and gene dependent. Another line of our research relates to the DNA and RNA structures. Namely, we continued to study the relation of DNA structure and gene expression (collaboration with David Levens and Rafael Casellas) and the impact of mutations on RNA structure and their relation to disease (collaboration with Michael Gottesman; Chava Kimchi-Sarfaty). Single Nucleotide Polymorphisms (SNPs) are often linked to critical phenotypes such as diseases, or responses to vaccines, medications, and environmental factors. However, the specific molecular mechanisms by which a causal SNP acts is usually not obvious and changes in RNA secondary structure increasingly emerge as a possible explanation. We postulated that to measure such effects one has to consider whole Boltzmann ensemble of RNA conformers and compare Boltzmann enables of the native structure and the mutant. This postulate was the basis in our work on a new powerful method to measure the impact of a SNP/mutation on RNA structure has been selected for oral presentation on RECOMB 2012, which belong to the top conferences in Computational Biology. The manuscript describing this work is in submission. We also added a new aspect to our work on RNA structure. Specifically in collaboration with Zuben Sauna, FDA, we began experimental and computational studies of the properties of Aptamers. In particular we have developed a computational approach to identify sequence/structure motifs of SELEX derived aptamers. This work has been recently published in Bioinformatics (3) and selected for oral presentation at ISMB 2012 which is another top computational biology conference. We have extended our interest in developing computational methods to analyze heterogeneous data from study of complex diseases to analysis of singe cell expression. Within isogenic cell population, the stochastic nature of gene expression promotes cell-to-cell differences in protein level, usually referred to as noise. Several transcription features, including presence of TATA box has been linked to increased expression noise. We have investigated the question of to what extent sequence features known or postulated to accompany translation efficiency can also be associated with noise differential. Strikingly, we found that the impact on noise strength associated with high tRNA adaptation index is comparable to the impact of the presence of a TATA box indicating that the translation originated noise has been greatly underappreciated. We have recently published these results in PloS Computational Biology (4). Finally we continue to apply our computational expertise in collaboration with other group including evolutionary analysis (5,6), disease studies (7), and recombination hotspots (8).