samtools mpileup floating point error

[Hkins552]

Hi All,
I am having trouble with mpileup today on samtools.

$ samtools mpileup -uf human_g1k_v37.fasta input.recal.bam > input.variants.raw

[mpileup] 1 samples in 1 input files
<mpileup> Set max per-file depth to 8000
Floating point exception


Any ideas on this floating point exception error?

[DrDolittle]

Hi,

I experienced the same error recently on our linux server with samtools 0.1.16 and 0.1.17 (including the lates build). The bam files are ok (checked with Picard). Interestingly, this command (again different samtools versions) runs smoothly on an iMac machine? Can someone give a hint to a solution? Any help appreciated!

The bam files were created as follows (on the linux/red hat server):
bwa aln -I -n 1 -t 10 g1k.fasta in.fastq > out.sai
bwa samse g1k.fasta out.sai in.fastq -f myfile.sam
samtools view -bhS myfile.sam -o myfile.bam
samtools sort myfile.bam myfile_sorted
samtools index myfile_sorted.bam

bwa-0.5.9-r16
samtools-0.1.17 (0.1.17)

[Xueya Zhou]

I encountered the same problem with samtools version 1.13 to 1.17.

I found it can be solved by replacing the numeric chromosome names 1..22 in GRCh37 with chr1..chr22 within the bam file header.

[bw.]

Also getting this error with samtools 0.1.17 and 0.1.18

Hi,
I'm running samtools on a 64bit linux server after compiling the tool on this server using make (no issues during the compilation).

I'm getting the float point error and it happens even if I don't provide any input .bam files:

samtools mpileup -f Libraries/hg19/hg19.fa
[mpileup] 0 samples in 0 input files
Floating point exception

It does also happen when I provide .bam files - making the tool unusable.

The hg19.fa has chromosome names "chr1", "chr2", etc. so the chromosome name work-around mentioned above didn't work in my case.

However, samtools vs. 0.1.12 does work both with and without input .bam files (it was compiled on this same server).

Any help or work arounds would be much appreciated.

-Ben

Ps. I also tried regenerating the .fai file but it didn't help, and also checked for dependency issues but at first glance things look alright:

ldd ./samtools
libncurses.so.5 => /usr/lib64/libncurses.so.5 (0x00000039a8600000)
libm.so.6 => /lib64/libm.so.6 (0x0000003995800000)
libz.so.1 => /usr/lib64/libz.so.1 (0x0000003996400000)
libc.so.6 => /lib64/libc.so.6 (0x0000003995400000)
libdl.so.2 => /lib64/libdl.so.2 (0x0000003995c00000)
/lib64/ld-linux-x86-64.so.2 (0x0000003995000000)

[wencanh]

Wonderful

I encountered the same problem and I solved it with the help of the thread posted by Xueya Zhou. Thank you very much

I encountered the same problem with samtools version 1.13 to 1.17.

I found it can be solved by replacing the numeric chromosome names 1..22 in GRCh37 with chr1..chr22 within the bam file header.

[yanlinlin82]

I met the same error. I found it may be caused by the incorrect index file (.fai). Use 'samtools index' to re-build index for the reference file solved the problem in my case.

[kingsalex]

mpileup floating point error gatk fai file problem

Yes it seems to be fai problem for me too. ie when i use new fai file it works ok . I think it has something to do with the format of the Fai file GATK builds automatically if you let it.

[genomicist]

I met the same error. I found it may be caused by the incorrect index file (.fai). Use 'samtools index' to re-build index for the reference file solved the problem in my case.

Thanks a lot for this solution. Only I think you mean "samtools faidx" and not "samtools index".

[chung2000]

samtools mpileup floating point error

Thanks a lot for this solution. Only I think you mean "samtools faidx" and not "samtools index".

Thank you, that saved me a lot of consternation when it came time for me to re-index my BAM files.

[drashu_11]

Hi
I am having trouble with vcfutils.pl command

I used the follwing command

samtools view mpileup -uf ref.fa aln.bam | bcftools view -bvcg - > raw.bcf

and I have the raw.bcf

But when I used the following command to build concensus

bcftools view raw.bcf | vcfutils.pl vcf2fq > cons.fq

Then I have the fowwing error message

vcfutils.pl command not found

I need help

[drashu_11]

Hi
I am having trouble with vcfutils.pl command

I used the follwing command

samtools view mpileup -uf ref.fa aln.bam | bcftools view -bvcg - > raw.bcf

and I have the raw.bcf

But when I used the following command to build concensus

bcftools view raw.bcf | vcfutils.pl vcf2fq > cons.fq

Then I have the fowwing error message

vcfutils.pl command not found

I need help

Ignore the quoted message I could manage that problem

[dGho]

I encountered the same problem with samtools version 1.13 to 1.17.

I found it can be solved by replacing the numeric chromosome names 1..22 in GRCh37 with chr1..chr22 within the bam file header.

This may be a silly question, but I don't get how you would change the chromosome names in bam files...they are binary, did you mean you changed the headers in the sam files?

[patrickc01]

Ignore the quoted message I could manage that problem

Hi,
I am new to bioinformatics and have run into the same problem. Could you tell me how you solved it?

Thanks!