mulitple/split/chimeric reads from bwa bwasw long read aligner and SNP calling

[Maulik23]

I have some queries about split /chimeric reads: 1) Though I have read several defintions of split reads as given in this thread: http://seqanswers.com/forums/showthread.php?t=2267. I am stil unable to grasp the whole idea of split reads. can anyone please elaborate in a simple way? 2). our task is to perform SNP calling from NGS data. I used bwa-sw for mapping the roche 454 data and the mapping yielded multiple alignment for some reads. So should I remove those reads for SNP calling ? If possible, please enlighten me on above issues.

[lh3]

Have a look at the SAM spec about multiple mappings vs. chimeric alignments (also the footnote). Split alignment typically refers to an alignment with the N cigar operator.

For Illumina, 454, Sanger reads and BACs, contigs, bwa-sw has been effectively deprecated by bwa-mem.