Vcf : remove variant absent in affected sample

Question

[quentin54520]

Hello,

I have a multi sample vcf with affected and unaffected samples. Usually i remove variant absent in affected samples After annotations and conversion of vcf to rdv file, through a python script, but it's a bit useless to annotate all these variants to remove it immediatly After.

So i'm searching a tool to filter directly the vcf before annotations. I look on bcftools or vcftools but i didn't find the option for what i want.

Answer 1

[Pierre Lindenbaum]

using vcffilterjdk http://lindenb.github.io/jvarkit/VcfFilterJdk.html

java -jar dist/vcffilterjdk.jar  -e ' final List<String> controls =  Arrays.asList("S1","S2","S3","S4"); return controls.stream().map(S->variant.getGenotype(S)).noneMatch(G->G.isHet() || G.isHomVar());' in.vcf

[quentin54520]

Thank you for your reply. I'm not sure I understand the command, I have to replace the ("S1","S2","S3","S4") by the list of my unaffected samples?

[Pierre Lindenbaum]

by the list of my unaffected samples?

yes

[quentin54520]

Unfortunately when i try to install it with:

git clone "https://github.com/lindenb/jvarkit.git"

cd jvarkit

./gradlew vcffilterjdk

i get the error:

Exception in thread "main" java.lang.RuntimeException: Could not create parent directory for lock file /shared/home/quentin67100/.gradle/wrapper/dists/gradle-6.7-bin/efvqh8uyq79v2n7rcncuhu9sv/gradle-6.7-bin.zip.lck at org.gradle.wrapper.ExclusiveFileAccessManager.access(ExclusiveFileAccessManager.java:43) at org.gradle.wrapper.Install.createDist(Install.java:48) at org.gradle.wrapper.WrapperExecutor.execute(WrapperExecutor.java:107) at org.gradle.wrapper.GradleWrapperMain.main(GradleWrapperMain.java:63)

[Pierre Lindenbaum]

do you have the right to create directories under /shared/home/quentin67100/.gradle ?

[quentin54520]

You're right, it's an issue with space disk quota on my home.

Answer 2

[sbstevenlee]

It sounds like you are already familiar with Python, so here's one solution using the pyvcf submodule from the fuc package I wrote.

Let's imagine you have 3 controls (C1-C3) and 3 affected samples (A1-A3).

>>> from fuc import pyvcf
>>> data = {
...     'CHROM': ['chr1', 'chr1', 'chr1'],
...     'POS': [100, 101, 102],
...     'ID': ['.', '.', '.'],
...     'REF': ['G', 'T', 'T'],
...     'ALT': ['A', 'C', 'A'],
...     'QUAL': ['.', '.', '.'],
...     'FILTER': ['.', '.', '.'],
...     'INFO': ['.', '.', '.'],
...     'FORMAT': ['GT', 'GT', 'GT'],
...     'C1': ['0/1', '0/1', '0/0'],
...     'C2': ['0/0', '1/1', '0/0'],
...     'C3': ['0/1', '0/1', '0/0'],
...     'A1': ['0/0', '0/1', '1/1'],
...     'A2': ['0/0', '1/1', '0/1'],
...     'A3': ['0/0', '0/0', '0/1'],
... }
>>> vf = pyvcf.VcfFrame.from_dict([], data)
>>> # vf = pyvcf.VcfFrame.from_file('in.vcf')
>>> vf.df
  CHROM  POS ID REF ALT QUAL FILTER INFO FORMAT   C1   C2   C3   A1   A2   A3
0  chr1  100  .   G   A    .      .    .     GT  0/1  0/0  0/1  0/0  0/0  0/0
1  chr1  101  .   T   C    .      .    .     GT  0/1  1/1  0/1  0/1  1/1  0/0
2  chr1  102  .   T   A    .      .    .     GT  0/0  0/0  0/0  1/1  0/1  0/1

You can remove variants that are absent in the affected.

>>> filtered_vf = vf.filter_sampany(['A1', 'A2', 'A3'])
>>> filtered_vf.df
  CHROM  POS ID REF ALT QUAL FILTER INFO FORMAT   C1   C2   C3   A1   A2   A3
0  chr1  101  .   T   C    .      .    .     GT  0/1  1/1  0/1  0/1  1/1  0/0
1  chr1  102  .   T   A    .      .    .     GT  0/0  0/0  0/0  1/1  0/1  0/1

Optionally write the VCF data to an output file.

# filtered_vf.to_file('out.vcf')

Let me know if you have any questions.