Fig. 2: Simulation results under different parameter settings.
From: BIT: Bayesian Identification of Transcriptional regulators from epigenomics-based query region sets
a Mean squared errors for estimation of global parameters using 100 replicate datasets in each simulation setting. b The average of MSEs for estimation of importance parameters (averaged over different TRs) over 100 replicates. c Box plots of Spearman’s rho correlation over 100 replicates between ranks from BIT or the baseline method and the true ranks of simulated TRs. Box plots show the median (center line), 25th and 75th percentiles (box edges), and whiskers extending to the largest or smallest value that is within 1.5 times of the interquartile range, and points beyond this range are plotted as outliers. d The distributions of simulated vs. real numbers of ChIP-seq datasets of individual TRs. e Data generated in our sensitivity analysis from t and gamma distributions, to examine whether heavy-tailedness and skewness would significantly affect the performance of BIT. f Box plots of Spearman correlation over 100 replicates under t and gamma distributions, with baseline under normal distribution. As in (c), box plots show the median (center line), 25th and 75th percentiles (box edges), and whiskers extending to the largest or smallest value that is within 1.5 times of the interquartile range, and points beyond this range are plotted as outliers. Source data are provided as a Source Data file.
