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
Same panels and axes as in Figs. 5 and 6, but Relief-F with optimized k nearest neighbors is compared with Random Jungle, Random Forest (R implementation), and tuned Random Forest with optimized mtry. Average ranks of the 96 main effect genes simulated in each data set together with null genes for a total of 16,920 genes. The 96 genes are split into four panels based on fold change (FC). The horizontal axis of each plot is labeled by the combination of negative binomial parameters (θ:µ) of the main effect gene. Each point is the average rank across n = 1000 replicate simulations.