Screening the millions of reads that next-generation sequencing produces presents a major challenge when searching for candidate SNPs. Both the Maq and GSNAP algorithms include SNP screening capabilities.
Maq(1) has a two level screening process which searches initially for differences between the reads and the reference sequence. Next a filtering step takes place which sifts the initial results looking for a minimum number of variants of the same class per column of reads, and variants embedded within high quality regions. The results are presented in a comprehensive report. You can also view your results in a Maqview or Tablet.
With GSNAP(2) the SNP analysis takes a different approach looking at both previously reported SNPs as well as new candidates. The user must supply a list of known SNPs as well as the reads and a reference sequence. GSNAP performs a SNP-tolerant alignment of all major and minor alleles. The algorithm enables the minor alleles to be differentiated from mismatches. The results are presented in a comprehensive report. You can also view your results in Tablet.
(1) Heng Li, Jue Ruan and Richard Durbin
Mapping short DNA sequencing reads and calling variants using mapping quality scores
Genome Research 2008 18:1851-1858
(2) Thomas D. Wu and Serban Nacu
Fast and SNP-tolerant detection of complex variants and splicing in short reads
Bioinformatics 2010 26: 873-881