Hi all,
I have a question related to SNP detection and Illumina short reads. We have a paired-end 54bp dataset, where we can trim larger adaptor sequence matches without too much difficulty. However, when only one or two base pairs at the end of reads are actually the first two bases of the adaptor sequence, it is impossible to detect it as adaptor, the reads align since it is only 1-2 mismatches and subsequently a candidate SNP is detected.
In the example below, the lowercase 'a' at the end represents the first adaptor base, which is detected as a SNP in the alignment.
ATGCAGATTGGAGATCTTAACACGCTCGa
Any ideas about how to filter such likely false positive SNPs, apart from excluding SNPs that are present ONLY in the last one or two bases of reads...?
Matt
I have a question related to SNP detection and Illumina short reads. We have a paired-end 54bp dataset, where we can trim larger adaptor sequence matches without too much difficulty. However, when only one or two base pairs at the end of reads are actually the first two bases of the adaptor sequence, it is impossible to detect it as adaptor, the reads align since it is only 1-2 mismatches and subsequently a candidate SNP is detected.
In the example below, the lowercase 'a' at the end represents the first adaptor base, which is detected as a SNP in the alignment.
ATGCAGATTGGAGATCTTAACACGCTCGa
Any ideas about how to filter such likely false positive SNPs, apart from excluding SNPs that are present ONLY in the last one or two bases of reads...?
Matt
Comment