search

ole-tange / barmaids

0 stars 0 forks source link

bammds crashes when running on a different bam file #57

Closed sapfo closed 10 years ago

sapfo commented 10 years ago

./bammds testinputfiles/Wollstein.hg19.chr21.chr22.one_allele.txt testinputfiles/STM1_SSLchr21.bam output: Use of uninitialized value $ in scalar chomp at ./bammds line 16086. Use of uninitialized value $out[3] in join or string at ./bammds line 16103, line 8220.

Now there might be a problem between the names of chromosome between the bam and the reference panel...but i think if thats the case the output should be a bit more obvious as to what the problem is.

ole-tange commented 10 years ago

stdout ./bammds.pl --debug testinputfiles/Wollstein.hg19.chr21.chr22.one_allele.txt testinputfiles/STM1_SSL_chr21.bam|head

gives:

samtools mpileup -Bl /tmp/VEIKa0feb_.bed -q 20 -Q 30 testinputfiles/STM1_SSL_chr21.bam /tmp/CopPpmn_Qw.bam |

If the samtools command is run, then there is no output, so the bam-file cannot be pileupped.

ole-tange commented 10 years ago

I can confirm the problem is the name of the chromosomes:

In /tmp/VEIKa0feb.bed they have a number. In the BAM-file they are called chrNN. If I change NN to chrNN in /tmp/VEIKa0feb.bed then the pileup generates output.

Can we somehow detect this? I have made bammds fail if the mpileup gives no output.

Victor: Can we somehow correct this (e.g. tell mpileup that chrNN == NN)?

sapfo commented 10 years ago

yes, this is exactly what i think we need to do. chrNN=NN or CHRNN=NN. sorry about that.... do you get those emails?

On 10-Oct-13 9:48, Ole Tange wrote:

I can confirm the problem is the name of the chromosomes:

In /tmp/VEIKa0feb_.bed they have a number. In the BAM-file they are called chrNN.

Can we somehow detect this?

Can we somehow correct this (e.g. tell mpileup that chrNN == NN)?

— Reply to this email directly or view it on GitHub https://github.com/ole-tange/barmaids/issues/57#issuecomment-26034741.

ole-tange commented 10 years ago

chromNN=NN ChromNN=NN CHROMNN=NN

chrE

Can we do any substring of "chromosome" of at least 3?

samtools view -H testinputfiles/STM1_SSL_chr21.bam |grep SN samtools view -H testinputfiles/Han_70000reads_hg19.bam|grep SN

Our fakebam should include the correct spelling.

ole-tange commented 10 years ago

Fixed in [238].