HAPDeNovo: a haplotype-based approach for filtering and phasing de novo mutations in linked read sequencing data

Table 1 The performance of FreeBayes, TrioDeNovo, GATK, and DeNovoGear with sequencing depth threshold 15 and their optimal parameters (GL = − 50 for FreeBayes, DQ = 7 for TrioDeNovo, PL = 450 for GATK, and PP = 1E-4 for DeNovoGear)

After further applying HAPDeNovo, the number of false positives decreases significantly for all four inputs. HAPDenovo also calculates the confidence of DNMs. A high proportion of TP (33/44, 32/43) comes from high-confidence DNMs. TP (True Positive): number of de novo mutations in both candidate set and the gold standard. FP (False Positive): number of mutations in the candidate set but not in the gold standard. HC (High Confidence): high-confidence DNMs. LC (Low Confidence): low-confidence DNMs

ISSN: 1471-2164