* All validation experiments were performed on a CentOS (ver. 6.3) server having 64 Intel Xeon dual-core 2.3 Ghz processors and 128 GBs of RAM. Individual options of COGNIZER were executed using 32 CP
Running time for MiMeNet was calculated using the recommended settings from the Github page: 20 random searches for hyper-parameter tuning, 10-iterations of 10-fold cross-validation for prediction eva
GATK3 vs GATK4 Genotype Quality (GQ) scores QD (Quality by Depth) scores. Both scores are on PHRED-like scale, expressing the -10*log10 likelihood of an incorrect call. (XLSX)
