Predicting Cellular Growth from Gene Expression Signatures

摘要: Maintaining balanced growth in a changing environment is fundamental systems-level challenge for cellular physiology, particularly microorganisms. While the complete set of regulatory and functional pathways supporting proliferation are not yet known, portions them well understood. In particular, governed by mechanisms that highly conserved from unicellular to multicellular organisms, disruption these processes metazoans major factor development cancer. this paper, we develop statistical methodology identify quantitative aspects underlying Saccharomyces cerevisiae. We find expression levels small genes can be exploited predict instantaneous rate any culture with high accuracy. The predictions obtained fashion robust biological conditions, experimental methods, technological platforms. proposed model also effective predicting rates related yeast bayanus diverged Schizosaccharomyces pombe, suggesting signature across wide range evolution. investigate significance gene based upon multiple perspectives: perturbing network through Ras/PKA pathway, observing strong upregulation even absence appropriate nutrients, discovering putative transcription binding sites, enrichment growth-correlated genes. More broadly, enables insights about at an time scale, inaccessible direct methods. Data tools enabling others apply our methods available http://function.princeton.edu/growthrate.