GC overhead limit exceeded with jvarkit vcfrenamechr

[browngcm]

Subject of the issue

I'm trying to use vcfrenamechr to rename the the Trinity headers in my VCF file to a naming format that is accepted by snpEff. My goal is to predict the effect of SNPs in my vcf file with snpEff.

Your environment

• version of jvarkit - latest version
• version of java - 1.8
• which OS - CentOS 6

Steps to reproduce

/scratch/$USER/jvarkit/dist/vcfrenamechr /
-f /scratch/$USER/jvarkit/src/main/resources/chromnames/hg19_to_g1kv37.tsv /
-o SNPs_vcfrenamechr.vcf /scratch/$USER/interspecific_SNPs.vcf

Expected behaviour

Rename Trinity headers to chromosomes

Actual behaviour

[SEVERE][Launcher]GC overhead limit exceeded [INFO][Launcher]vcfrenamechr Exited with failure (-1)

[lindenb]

Hum... I havent used this tool for years.... can you please try with a simple java -jar cmd:

java -jar /scratch/$USER/jvarkit/dist/vcfrenamechr.jar \
-f /scratch/$USER/jvarkit/src/main/resources/chromnames/hg19_to_g1kv37.tsv  \
    SNPs_vcfrenamechr.vcf > /scratch/$USER/interspecific_SNPs.vcf

or you could use vcfsetdict: http://lindenb.github.io/jvarkit/VcfSetSequenceDictionary.html

[lindenb]

Is that to be expected?

of course not.

Can you please show the exact cmd line your have used ?

[lindenb]

opps I'm afraid I've done a typo in my previous comment....

[lindenb]

it should have been

java -jar /scratch/$USER/jvarkit/dist/vcfrenamechr.jar \
-f /scratch/$USER/jvarkit/src/main/resources/chromnames/hg19_to_g1kv37.tsv  \
    input.vcf > output.vcf

sorry, you might have deleted your original VCF file.

[lindenb]

the current version works on my machine:

$ echo -e "ref\tCHANGEDref\nref2\tCHANGEDref2" > tmp.txt && java -jar dist/vcfrenamechr.jar   -m tmp.txt test/resources/toy.vcf.gz | cut -f 1 | grep CHANGED
##contig=<ID=CHANGEDref,length=45>
##contig=<ID=CHANGEDref2,length=40>
CHANGEDref
CHANGEDref
CHANGEDref2
CHANGEDref2
CHANGEDref2

is there any error message on your side ?

[lindenb]

can you please attach to this a few variant of your vcf (+ VCF header) as well as your mapping file.

[lindenb]

grep TRINITY_DN192238_c0_g3_i5 your_mapping.txt > minimapping.txt
 awk '/#/ {print;next;} {print; exit}' > mini.vcf

[lindenb]

attach files minimapping.txt and mini.vcf by selecting them in your next comment area

[lindenb]

what is that bam ? vcfrenamechr doesn't use a bam! It's a simple tab delimited file. http://lindenb.github.io/jvarkit/ConvertVcfChromosomes.html

  * -f, --mapping, -m
      load a custom name mapping. Format (chrom-source\tchrom-dest\n)+