Fast and SNP-aware short read alignment with SALT

BMC Bioinformatics

Table 2 Statistics on real human datasets

Aligner	Arg	Sen (%)	Acc (%)	Uniq (%)	Unmapped	T1	T4
dataset 1: real-100
SALT.snp	-r 21	99.15	99.76	90.43	33,981	9 m 37 s	2 m 37 s
	-r 10	99.24	99.76	90.19	30,341	11 m 41 s	3 m 8 s
	-r 5	99.27	99.76	90.09	29,380	13 m 24 s	3 m 38 s
SALT.linear	-r 21	99.13	99.76	90.51		9 m 30 s	2 m 26 s
	-r 10	99.23	99.76	90.29	30,619	10 m 34 s	2 m 52 s
	-r 5	99.26	99.76	90.21	29,478	11 m 37 s	3 m 16 s
BWA-MEM		99.99	97.59	91.60	374	18 m 59 s	5 m 42 s
dataset 2: real-148
SALT.snp	-r 21	97.39	99.95	89.35	104,347	18 m 13 s	4 m 35 s
	-r 10	97.44	99.94	89.30	102,474	19 m 45 s	6 m 23 s
	-r 5	97.47	99.94	89.21	101,457	26 m 34 s	7 m 25 s
SALT.linear	-r 21	97.38	99.96	89.49	104,839	14 m 14 s	3 m 47 s
	-r 10	97.43	99.95	89.47	102,905	19 m 38 s	5 m 12 s
	-r 5	97.46	99.95	89.39	101,795	25 m 20 s	6 m 40 s
BWA-MEM		99.51	97.28	93.55	19,450	23 m 18 s	6 m 29 s

SALT.linear and SALT.snp seed alignments with a linear reference and an SNP-augmented reference, respectively
SALT selects seeds in reads every X nt with the setting “-r X”
T1 and T4 are escaped time with 1 thread and 4 threads, respectively
Best results are shown in bold

ISSN: 1471-2105