PROJECT TITLE :

bLARS: An Algorithm to Infer Gene Regulatory Networks

ABSTRACT:

Inferring gene regulatory networks (GRNs) from high-throughput gene-expression information is an important and difficult problem in systems biology. Several existing algorithms formulate GRN inference as a regression downside. The out there regression based algorithms are based on the idea that each one regulatory interactions are linear. But, nonlinear transcription regulation mechanisms are common in biology. In this work, we propose a replacement regression based mostly methodology named bLARS that permits a selection of regulatory interactions from a predefined but otherwise arbitrary family of functions. On three DREAM benchmark datasets, particularly gene expression data from E. coli, Yeast, and a artificial information set, bLARS outperforms state-of-the-art algorithms in the terms of the score. On the individual networks, bLARS offers the most effective performance among currently available similar algorithms, namely algorithms that don't use perturbation data and are not meta-algorithms. Moreover, the presented approach will also be utilized for general feature choice issues in domains other than biology, provided they're of an analogous structure.