Fig. 4From: Manipulating base quality scores enables variant calling from bisulfite sequencing alignments using conventional bayesian approachesThe shared fraction of true and false positive variants in real and simulated data for A. thaliana (Cvi-0), following analysis with GATK UnifiedGenotyper. Distinct WGBS datasets were simulated from both the real WGS alignments and the real WGS reads, separately. The panels denote A true positives, before and B after filtering, according to recommended hard-filter thresholds in GATK best-practices, and C false positives, also before and D after filtering. The thresholds chosen for filtering are further described in Supplementary Table S3Back to article page