BWA Multiple Hits

[EthoStrain]

I have been using BWA to much success, except one small issue. When I do an alignment I only get back one result per read, but I know that I should be getting more than just a single result.

I have tried using -N and -R but my output has not changed. Could anyone give me some guidance?

Thanks!

[EthoStrain]

A bit clearer

I have narrowed it down so that I know BWA is returning only one hit per read per fasta file. I still am unsure why -N is not working as it claims to return all hits.
I know that bowtie has an 'all' option so I may just go with bowtie instead.

Thanks!

[epigen]

Do you mean the -N from aln? That would not make much sense:
-N Disable iterative search. All hits with no more than maxDiff differences will be found. This mode is much slower than the default.

or that from sampe?
-N INT Maximum number of alignments to output in the XA tag for disconcordant read pairs (excluding singletons). If a read has more than INT hits, the XA tag will not be written. [10]

I guess you'll want to use -n from samse/sampe. This will still return one "main" hit but report the others in the XA tag:
-n INT Maximum number of alignments to output in the XA tag for reads paired properly. If a read has more than INT hits, the XA tag will not be written. [3]

looks like that for a (single end) read with 3 possible locations:
3_32_497 16 chr2 180106465 0 48M * 0 0 CGGTGAAACCCTGTCTCTACTAAAAATACAAAAAATTAGTCAGGCATG .... XA:Z:chr13,-27690265,48M,1;chr16,-85778252,48M,1;

Just set -n <bignumber> to get as many locations as you like and extract them with a script.

[EthoStrain]

Works!

Yes, that was exactly what I was missing. Thank you very much!

[cfrias]

Hi,

I am working with 454 data.
I try to obtain alternative Hits with the bwasw algorithm, Can somebody help, me please?