Module Anchored Network Inference: A Sequential Module-Based Approach to Novel Gene Network Construction from Genomic Expression Data on Human Disease Mechanism

Figure 4

Window #1 network inference. (a) Time-series gene expression data of 7 genes within the adipogenesis network collected at 0, 6, 12, 24, 48, 72, 96, and 672 hours during adipocyte differentiation [9]. GEO accession number of gene expression data is GSE6795. Gene expression values were normalized to a maximum value of 1. Expanded view of gene expression data between 0 and 100 hours is shown in Supplementary Figure S1. (b) Time-series expression profiles of the three genes selected from the pool of 7 genes for window #1. (c) The parameters of different Regulatory Relationships (RRs) were fitted to the time-series data of genes shown in (b). Since KLF4 was determined to be the gene at the top of the hierarchy, the two remaining genes were fitted in parallel or serial regulatory relationships. Parameters describing the different regulatory relationships are as described in Figure 3. BIC scores of regulatory relationships were estimated as . Parallel regulatory relationship (RR #1) was chosen as optimal to describe gene relationships in window #1 because of its smallest BIC score. The parameters estimated for the selected regulatory relationship are listed in Table 1.