DiSNPindel: improved intra-individual SNP and InDel detection in direct amplicon sequencing of a diploid

BMC Bioinformatics

Table 1 Comparison of software performance in intra-individual SNP detection

Software^a	Grade or score threshold^b	Found SNPs (%)	Missed SNPs (%)
DiSNPindel	1	60 (90.9 %)	6 (9.1 %)
	2	61 (92.4 %)	5 (7.6 %)
	3	61 (92.4 %)	5 (7.6 %)
	4	63 (95.4 %)	3 (4.6 %)
	5	63 (95.4 %)	3 (4.6 %)
	6	63 (95.4 %)	3 (4.6 %)
novoSNP	18	0 (0.0 %)	66 (100.0 %)
	13	11 (16.7 %)	55 (83.3 %)
	9	16 (24.4 %)	50 (75.6 %)
	6	19 (28.8 %)	47 (71.2 %)
	3	25 (37.9 %)	41 (62.1 %)
	1	25 (37.9 %)	41 (62.1 %)
Mutation surveyor	Medium sensitivity	20 (30.3 %)	46 (69.7 %)
	High sensitivity	20 (30.3 %)	46 (69.7 %)

The numbers of found and missed SNPs were experimentally verified with 66 CAPS-associated intra-individual SNPs originating from 62 single traces [22]
^aPolyPhred did not detect any SNPs at ranks 1–6 and was thus excluded from the comparison
^bGrade 2, score 13 and medium sensitivity could be a high threshold for reliable SNP detection in DiSNPindel, novoSNP and Mutation Surveyor, respectively, where Grade 4, score 6 and high sensitivity could be a low threshold

ISSN: 1471-2105