HISAT2 / hisat2_extract_snps_haplotypes_VCF.py / --genotype-vcf & VCF_fnames

[kevinrue]

Dear all,

I would like to use the Python script hisat2_extract_snps_haplotypes_VCF.py.

I cannot fully understand some of the command line options from the description printed using the --help option.

I am confused about the role of the positional argument VCF_fnames and the optional argument --genotype-vcf.

Here is a copy paste of those two argument descriptions (let me know if I should paste the complete output of --help, it is not so long):
VCF_fnames A comma-seperated VCF files (plain text or gzipped
file is accepted: GRCh38_dbSNP_no_SVs.vcf or
GRCh38_dbSNP_no_SVs.vcf.gz

--genotype-vcf GENOTYPE_VCF
VCF file name for genotyping (default: empty)


I have access to merged VCF files: one per human chromosome, each file contains genotypes and VEP predictions for the same set of samples.
I thought these files should go as the positional argument VCF_fnames, however, I would like to know:

1. whether my files contain the right information for the positional argument
2. what kind of VCF file should go in the optional argument --genotype-vcf

Many thanks in advance!

Kevin

[GenoMax]

Can you provide a reference for where you got the said script from?

[kevinrue]

https://github.com/infphilo/hisat2/blob/master/hisat2_extract_snps_haplotypes_VCF.py

[kevinrue]

Just to clarify, in my previous message I sent the link to the Python script itself.
However, I installed the software from the source code availabe at https://ccb.jhu.edu/software/hisat2/...html#options-1

Thanks in advance for looking into this!

[xris]

Hi Did you resolve your question? I have the same doubt.

If I run

hisat2_extract_snps_haplotypes_VCF.py -v path/to/.fa path/to/.vcf test

I get only empty output files .snp .haplotype

Thanks

[kevinrue]

Hi xris,

Sorry, no, never sorted that one out. Ended up using using the UCSC snp146Common.txt file.

Regarding you empty .snp and .haplotype issue, does this other post help you in any way?

You say empty, but in my case, I had one entry in the output, because the script relies on the rsid field to uniquely identify variants, while in my case all the variants had "." and therefore all variants after the first one were skipped. However, that was over a year ago, and the script may behave differently now.

Best wishes,
Kevin