lskatz / lyve-SET

:dancer: :palm_tree: LYVE-SET, a method of using hqSNPs to create a phylogeny, especially for outbreak investigations
MIT License
25 stars 18 forks source link

mpileup generate empty file #79

Closed alantsangmb closed 5 years ago

alantsangmb commented 5 years ago

Hi, I have installed lyve-SET-1.1.4f and I have run make env to set the perl and perl5lib environoment and no errors returns with "make check"

However, when I run lyve-SET on my fastq data, it returns following error: ########################################################### launch_set.pl: variantCalls: /home/wgs/Tools/lyve-SET-1.1.4f/scripts/launch_varscan.pl /home/wgs/Mabs_lyve/bam/TH18008938.fastq.gz-FLAC029.gbk.sorted.bam --numcpus 16 --tempdir /home/wgs/Mabs_lyve/tmp --reference FLAC029.gbk.fasta --altfreq 0.75 --coverage 10 > /home/wgs/Mabs_lyve/vcf/TH18008938.fastq.gz-FLAC029.gbk.vcf launch_set.pl: Schedule::SGELK::command: Running /usr/bin/perl /home/wgs/Mabs_lyve/log/SGELK/qsub.434103.pl launch_varscan.pl: Creating a pileup /home/wgs/Mabs_lyve/tmp/TH18008938.fastq.gz-FLAC029.gbk.sorted.bam.mpileup launch_varscan.pl: Running mpileup: echo "CP014956 " | xargs -P 16 -n 1 -I {} sh -c 'echo "MPileup on {}" >&2; samtools mpileup -f FLAC029.gbk.fasta -Q 0 -B --region "{}" /home/wgs/Mabs_lyve/bam/TH18008938.fastq.gz-FLAC029.gbk.sorted.bam > /home/wgs/Mabs_lyve/tmp/TH18008938.fastq.gz-FLAC029.gbk.sorted.bam.$$.mpileup' MPileup on CP014956 [mpileup] 1 samples in 1 input files

Set max per-file depth to 8000 launch_varscan.pl: Sorting mpileup results into a combined file launch_varscan.pl: main::varscan: ERROR: the pileup is a zero-byte file /home/wgs/Mabs_lyve/tmp/TH18008938.fastq.gz-FLAC029.gbk.sorted.bam.mpileup Stopped at /home/wgs/Tools/lyve-SET-1.1.4f/scripts/launch_varscan.pl line 119. QSUB ERROR 65280 launch_set.pl: Schedule::SGELK::command: ERROR with command: Inappropriate ioctl for device /usr/bin/perl /home/wgs/Mabs_lyve/log/SGELK/qsub.434103.pl Stopped at /home/wgs/Tools/lyve-SET-1.1.4f/scripts/../lib/Schedule/SGELK.pm line 661. ########################################################### The mpileup file in the tmp directory was found to be empty. I execute samtools in the same terminal and it is version 0.1.19-96b5f2294a.
lskatz commented 5 years ago

This usually points to a difficulty in the fastq file. Could you manually inspect it? Some example commands are:

ls -lh /home/wgs/Mabs_lyve/reads/TH18008938.fastq.gz
zcat /home/wgs/Mabs_lyve/reads/TH18008938.fastq.gz | head # check that the first 10 lines look good
# IDs should be every four lines
zcat /home/wgs/Mabs_lyve/reads/TH18008938.fastq.gz | grep -n '^@' | head
zcat /home/wgs/Mabs_lyve/reads/TH18008938.fastq.gz | grep -n '^@' | tail

And if those don't work, could you validate the fastq file with my script?

git clone https://github.com/lskatz/ROSS
zcat /home/wgs/Mabs_lyve/reads/TH18008938.fastq.gz | perl ROSS/scripts/friends_ross.pl --pe --min-length 1 --min-quality 1 --verbose

If that doesn't work, then we can look at the bam file

lskatz commented 5 years ago

OH! Here is some older troubleshooting documentation in case it helps too. I think the tips here are still valid. https://github.com/lskatz/lyve-SET/blob/master/docs/TROUBLESHOOTING.md#fastq-integrity

alantsangmb commented 5 years ago

Thank you for providing the troubleshooting page. Problem solved if I cp the reference.fasta directly to the reference directory instead of using the symlink