ReliefSeq: A Gene-Wise Adaptive-K Nearest-Neighbor Feature Selection Tool for Finding Gene-Gene Interactions and Main Effects in mRNA-Seq Gene Expression Data
Each scatter plot is a replicate of the same XOR simulated model. Cases and controls are labeled in the 2-dimensional space of the two interacting transcripts (labeled INTERACT1 and INTERACT2). The simulated data includes approximately 16,000 null genes, but only the two interacting genes are shown. The XOR model is designed so that neither interacting gene by itself has a statistically significant differential expression between cases and controls (no main effect).