SEQanswers - Unable to intersect BAM file with bedtools

SEQanswers (http://seqanswers.com/forums/index.php)

- Bioinformatics (http://seqanswers.com/forums/forumdisplay.php?f=18)

- - Unable to intersect BAM file with bedtools (http://seqanswers.com/forums/showthread.php?t=82228)

Unable to intersect BAM file with bedtools

Hi everyone, I'm trying to use bedtools intersect to check the number of mapped reads in target regions (in a .bed file) originated by targeted bisulfite sequencing experiment (EpiSeq Roche).

I used the following command.

Code:

./bedtools intersect -bed -abam sample2.bam -b 

 ~/Data/MethylSeq/dataset/Agesmoke_dataset/AgeSmkSop_all_primary_targets.bed

The program terminate with the following message and no result at all.

Code:

* WARNING: File sample2.bam has inconsistent naming convention for record:

NC_000016.9  24163386  24163537  M03971:33:000000000-BN5NL:1:2114:12003:16132/1  255  +



* WARNING: File sample2.bam has inconsistent naming convention for record:

NC_000016.9  24163386  24163537  M03971:33:000000000-BN5NL:1:2114:12003:16132/1  255  +

I tried to modify the original SAM file removing the read that cause the problem (that was the first read in the SAM file) and the problem persists with the second read. I tried also the option -nonamecheck with no results.

Can someone help us? Thank you.
Nicola

Check your chromosome names.
Are they "chr" style in both bed and bam?

Quote:

Originally Posted by Richard Finney (Post 217082)

Check your chromosome names.
Are they "chr" style in both bed and bam?

Hi Richard, thanks for the reply.
About your question, I think not.

In the bed file I have record like this:

Code:

chr1    11123000        11123242        chr1:11123018-11123218

chr1    16696418        16696674        chr1:16696447-16696647

while in the SAM file (and consequently in the BAM file) I have record like that:

Code:

M03971:33:000000000-BN5NL:1:2114:12003:16132    99      NC_000016.9     24163387        255     151M    =       24163431        194     TGATCGGTGGTGA

TGGGTTAGGTAGAGTGTATTAGTTCGTTTTTATGTTGTTGATAAAGATATATTCGAGATTGTGTAATTTATGAAAAAGAGGTTTAATGGATTTGGGGAGGTTTTAATTATGGTGGAAGGTTAAAGTTATGTTTTATAT BCCCCCCBBC

ABGGGGGGFGGGGHHHHFGGHHHHHHHHGGHGGGHHHHHHHHHHHHHHHHHHHHHHHHHHHGHHHHHHHHHHHHHFHHHHGGHHHHHHHHHHHHHHHHGGFGGEHHGHHHHHHHHHHGHHHGHHHHHHHHHHHHHHHHHHH NM:i:1

  ZS:Z:++

Ps. the SAM file is the result of an alignment with BSMAP.

Is this the problem? How can I resolve it?
Nicola

NC_000016 is a name used for "chr16".
This the "official" name used at NCBI : https://www.ncbi.nlm.nih.gov/nuccore/NC_000016.10/

You have to convert the "NCBI name" to "chr" names (or vice versa).

There are many ways to rename fields. You can always brute force it using a custom simple program or script using your favorite programming language : bash, python, perl, C, etc.

Any easy way would be to reheader the bam file. Please see samtools documentation for this.

Quote:

Originally Posted by Richard Finney (Post 217084)

It works, thank you so much!!
I'm sorry for the triviality of the problem, but I'm not very practical with this stuff and the bedtools message wasn't very helpful.
Again, thank you!

Best regards
Nicola