This S4 class includes methods to infer posterior association networks and enriched modules of functional gene interactions from rich phenotyping screens.
The data set we use here comes from quantitative morphological screening for 249 gene-overexpression or RNAi knock-down experiments. For each individual cell, 145 different geometric features were computed by imaging analysis, and are subsequently scored with NNs trained to discriminate seven reference TCs with distinctive morphologies. For each TC, NN z-scores were computed from all scored cells in this TC (more details in Bakal 2007).
Powered by function write.graph in package igraph, this function writes the inferred PAN or module graphs to files in a variety of formats that are supported by igraph.