Weird Samtools Mpileup Error In Mplp_Func.

Question

[Shellfishgene]

Hi!

What is samtools mpileup trying to tell me with this error (or warning):

[mplp_func] Skipping because 298 is outside of 0 [ref:103424]

There are many like it. I double checked that I have the same reference fasta in the mpileup and the actual bwa mem mapping commands, and I do.

Cheers

Answer 1

[Sam]

Reading the samtools mailing list using google found this:

http://sourceforge.net/mailarchive/forum.php?thread_name=20111223191408.GA21080%40rzlab.ucr.edu&forum_name=samtools-help

Basically, what it is doing is exclude reads outside of the reference sequence From what the error is saying, it seems like your reference id 103424 has length of 0, therefore all the reads were excluded (I am guessing that the id is not exactly what you will see in your reference. It might be the internal index id of your reference, but I am not sure about that...)

Answer 2

[ketilm417]

I got this when mapping reads with bwa 0.6.2 (0.7.5 would segfault on certain files), and all reference sequences appear to be incorrectly mapped in the mpileup file after this occurs. It seems to be a bug either in bwa (producing an incorrect BAM file), or in samtools mpileup. Probably the latter, since examining the alignments with samtools tview doesn't reveal anything out of the ordinary.

This is such a mess: I can look at each of the BAM files with tview, and they look perfectly all right. But mpileup on the whole files fail, and if I try to look at the troublesome reference sequence (using -p to specify region), it only works for one of the files - the one which comes out incorrectly in the combined mpileup...

Answer 3

[Emad]

Use exactly the same references in the bwa and mpileup.