Is it a bug in BWA?

[SD2010Bioinfo]

After finishing BWA alignment, the result displays a strange phenomenon. That is a read whose flag showed it is unmapped read however ending with specific aligned information(map quality, map length etc.).


Is it a bug in BWA? How can I figure it out without any error information when processing BWA? Anybody encountered the same case?

[alex_kur]

It's not a bug: http://bio-bwa.sourceforge.net/:

-I see a read stands out the end of a chromosome and is flagged as unmapped (flag 0x4). What is happening here?

-Internally BWA concatenates all reference sequences into one long sequence. A read may be mapped to the junction of two adjacent reference sequences. In this case, BWA will flag the read as unmapped, but you will see position, CIGAR and all the tags. A better solution would be to choose an alternative position or trim the alignment out of the end, but this is quite complicated in programming and is not implemented at the moment.


So, before using your alignment, you have to filter unmapped reads with samtools view with flag -F 4

[SD2010Bioinfo]

It's not a bug: http://bio-bwa.sourceforge.net/:

-I see a read stands out the end of a chromosome and is flagged as unmapped (flag 0x4). What is happening here?

-Internally BWA concatenates all reference sequences into one long sequence. A read may be mapped to the junction of two adjacent reference sequences. In this case, BWA will flag the read as unmapped, but you will see position, CIGAR and all the tags. A better solution would be to choose an alternative position or trim the alignment out of the end, but this is quite complicated in programming and is not implemented at the moment.


So, before using your alignment, you have to filter unmapped reads with samtools view with flag -F 4

Thank you very much! I understand

If I do not filter these reads, will it disable the following process like picard markduplication?

[alex_kur]

I think picard filters reads by flags itself, but I'm not sure.

[SD2010Bioinfo]

I have done a test about the question I mentioned, picard will print out error information and be disabled. Anyway it is better to do filter before processing BWA bam files.