InferringStableGeneticNetworks fromSteady-StateData ?

摘要: Gene regulatory networks capture the interactions between genes and other cell substances, resulting from fundamental biological process of transcription translation. In some applications, topology network is not known, has to be inferred experimental data. The data consist expression levels genes, which are typically measured as mRNA concentrations in micro-array experiments. a so called genetic perturbation experiment, small perturbations applied equilibrium states changes activity measured. This paper develops novel algorithms that identify sparse stable explains obtained noisy Our identication algorithm based on convex relaxations sparsity stability constraints can also incorporate variety prior knowledge structure. Such either qualitative, specifying positive, negative or no quantitative, range interaction strengths. approach both synthetic data, for SOS pathway Escherichia coli, results show specication only ensures consistency with steady-state assumptions, but signicantly increases performance. Since method optimization, it eciently large scale networks.