varScan output, all “N” in the REF column

[xgong]

Hi,

I'm beginner of varScan. I'm using Varscan.v2.3.2 to do SNPs calling. Here is what I did according to the on line manual.(I’m using samtool 0.1.18)

samtools mpileup -d8000 -f ./hg19.fa ./mybam.bam | java -jar ./VarScan.v2.3.2.jar mpileup2snp --output-vcf 1 >my.mpileup2snp.vcf

In the result vcf file, there are all “N” in the column 4 (REF column).. Here is the example part of output by “less -S my.mpileup2snp.vcf” command

#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT
1 131249 . N G . PASS ADP=8;WT=0;HET=
1 131250 . N G . PASS ADP=8;WT=0;HET=
1 131251 . N C . PASS ADP=8;WT=0;HET=
1 131252 . N C . PASS ADP=9;WT=0;HET=
1 131253 . N C . PASS ADP=9;WT=0;HET=
1 131254 . N A . PASS ADP=8;WT=0;HET=
1 131255 . N G . PASS ADP=9;WT=0;HET=
1 131256 . N C . PASS ADP=8;WT=0;HET=
1 131257 . N A . PASS ADP=9;WT=0;HET=

Anything I did was wrong? Would you please help?
Thank you,
Xin

[swbarnes2]

Usually, the all N's in the pileup output means that something went wrong with that step. So double-check that you are using the same ref file that you used to align, and make sure the chromosome names are the same between your .bam and that reference (like that you don't have weird spaces or characters that might be truncated in the .bam.

If that all checks out, the most likely problem is that the reference fasta index didn't get made correctly. mpileup will try to make this file if it sees there is no such file, but if it can't make it, mpileup will carry on without it, and you'll get all N's, and it won't necessarily warn you that there's a problem.

So run samtools faidx on the reference fasta, see if there are any errors. Then check the fai file itself to see if it looks right.

[xgong]

Hi swbarnes2,

Thank you so much for your great help. It is very helpful. I did use the different ref files to align and to mpileup, and the chromosome names are different between the two ref files.

The ref file used to align did not work for mpileup. I got the error message of "Floating point exception". I found one post in this forum indicated the something wrong with the ref file header. I could not figure out exactly, so I used the different ref file for mpileup.

Here is the information of the header from the ref file used to align.
>1 dna:chromosome chromosome:GRCh37:1:1:249250621:1
>2 dna:chromosome chromosome:GRCh37:2:1:243199373:1
>3 dna:chromosome chromosome:GRCh37:3:1:198022430:1
..................................
>X dna:chromosome chromosome:GRCh37:X:1:155270560:1
>Y dna:chromosome chromosome:GRCh37:Y:2649521:59034049:1
>MT gi|251831106|ref|NC_012920.1| Homo sapiens mitochondrion, complete genome
>GL000207.1 dna:supercontig supercontig::GL000207.1:1:4262:1
>GL000226.1 dna:supercontig supercontig::GL000226.1:1:15008:1
.................................................

Do you have any idea how to fix the "Floating point exception" problem for the mpileup?.

Thank you again for your kind help.

Xin

[swbarnes2]

I don't know if this will fix your problem, but I'd get rid of those spaces in the names, for starters.

[dkoboldt]

Agreed, make the reference chromosome names as simple as possible. Make sure the reference that you aligned to is the same you provide for mpileup. If you still encounter the floating point exception, send an e-mail to the SAMtools help mailing list.

[xgong]

Thank you very much swbarnes2 and dkoboldt. It worked.

I'm sorry for the late reply.

Xin