Inference of Quantitative Models of Bacterial Promoters from Time-Series Reporter Gene Data

• DOI: 10.1371/journal.pcbi.1004028
• Publication Date: 2015-01-15T20:47:04Z
• Authors: Diana Stefan; Corinne Pinel; Stéphane Pinhal; Eugenio Cinquemani; Johannes Geiselmann; Hidde de Jong
• Keywords: Gene regulatory network; Bacterial transcription

The inference of regulatory interactions and quantitative models gene regulation from time-series transcriptomics data has been extensively studied applied to a range problems in drug discovery, cancer research, biotechnology. application existing methods is commonly based on implicit assumptions the biological processes under study. First, measurements mRNA abundance obtained experiments are taken be representative protein concentrations. Second, observed changes expression assumed solely due transcription factors other specific regulators, while activity machinery global physiological effects neglected. While convenient practice, these often not valid bias reverse engineering process. Here we systematically investigate, using combination experiments, importance this possible corrections. We measure real time vivo genes involved FliA-FlgM module E. coli motility network. From data, estimate concentrations by means kinetic expression. Our results indicate that correcting for commonly-made improves quality inferred data. Moreover, show simulation improvements expected even stronger systems which have longer half-lives varies more strongly across conditions than module. approach proposed study broadly applicable when transcriptome learn about structure dynamics networks. In case module, our demonstrate active FliA FlgM FliA-dependent promoters.