No VCF after running run_calls.pl

32 views
Skip to first unread message

ymc...@gmail.com

unread,
Sep 10, 2016, 12:08:19 PM9/10/16
to cortex_var
Dear all,

     I am trying to replicate the results of this research:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4703848/

     It claimed that it ran cortex_var with the following command:
perl CORTEX_release_v1.0.5.21/scripts/calling/run_calls.pl --ref Absent --fastaq_index INDEX --first_kmer 31 --auto_cleaning yes --outdir OUTDIR --ploidy 2 --genome_size 2800000 --mem_height 20 --mem_width 100 --qthresh 10 --do_union no --logfile log.txt --workflow joint --vcftools_dir /tank/vcftools/vcftools_0.1.9/

     After I ran it with SRR1182410 and ERR410084, I couldn't find any vcf files in the OUTDIR directory. I looked at the log files in OUTDIR as well as log.txt but I couldn't find any errors either. I am getting warnings like this, however:
===================
Warning message:
In xy.coords(x, y, xlabel, ylabel, log) :
  8876 y values <= 0 omitted from logarithmic plot
Warning message:
In xy.coords(x, y, xlabel, ylabel, log) :
  6415 y values <= 0 omitted from logarithmic plot
====================

INDEX file looks like:
==================
MRSA0001 . mrsa0001_pe1 mrsa0001_pe2
CoNS0001 . cons0001_pe1 cons0001_pe2
====================

The four data files named mrsa0001_pe1, mrsa0001_pe2, cons00001_pe1 and cons0001_pe2 are one liner that are SRR1182410_1.fastq.gz, SRR1182410_2.fastq.gz, ERR410084_1.fastq.gz and ERR410084_2.fastq.gz respectively.

      So what is going wrong here???

Thanks a lot in advance.
Yee Man

Zamin Iqbal

unread,
Sep 10, 2016, 5:50:53 PM9/10/16
to cortex_var
Hi there

>The four data files named mrsa0001_pe1, mrsa0001_pe2, cons00001_pe1 and cons0001_pe2 are one liner that are SRR1182410_1.fastq.gz, SRR1182410_2.fastq.gz, ERR410084_1.fastq.gz and ERR410084_2.fastq.gz respectively.

Do those files have carriage-returns at the end of them? If they have one line, with no carriage-return, then Cortex scripts would run

wc -l  mrsa0001_pe2

and get 0 back, and decide there is no data.

That bug is fixed on the tip of the github repo. Let me know if that fixes it
Reply all
Reply to author
Forward
0 new messages