loading plink format (generated by vcftools) file into Haploview

Question

[ernesto.lowy]

Hi,

I have generated a plink file from a vcf file using vcftools. This is the command I used:

tabix \
  -h ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/release/20110521/ALL.chr13.phase1_release_v3.20101123.snps_indels_svs.genotypes.vcf.gz \
  13:19000000-19050000 |\
  vcftools --vcf - --indv HG00096 --plink

vcftools generate 2 files (out.ped and out.map) and I want to load them into Haploview.

According to Haploview's documentation, Haploview accepts now plink generated files. And these are the requirements according to the User's guide:

Output files from PLINK can be loaded into Haploview using the PLINK tab on the initial loading screen. PLINK files must contain a header and at least one column header must be titled "SNP" and contain the marker IDs for the results in the file. PLINK loading also requires a standard PLINK map or binary map file corresponding to the markers in the output file. The map file can be either three or four headerless columns (the Morgan distance column is optional)"

I have tried with out.ped and out.map that were generated by vcftools and they do not work in Haploview, and reading the documentation it is not quite clear to me how the .ped file has to be formatted in order to be accepted by Haploview

Could anyone help me with this?

Thanks,
ernesto

[jvijai]

Does anyone have a solution for this?

It is unclear from the Haploview documentation, what kinda PLINK file should be an input. A PED and MAP converted from VCF via PLINK does not work.

I am interested in a solution on this too.

Answer 1

[Christopher Walker]

You need to convert the .ped/.map files to .ped/info files by using plink --recode HV. Also, haploview does not seem to accept indel variants, so you need to use the --snps-only just-acgt flag. For example:

plink --file <input_prefix> --recode HV --snps-only just-acgt --out <output_prefix>

The result is output_prefix.chr-*.ped and output_prefix.chr*.info These can be loaded into haploview. Note that the gPLINK can integrate haploview by indicating the path to the haploview.jar file under Project>configure. This allows you to right click on files and view in halploview.

[nute11a]

Hello! I was trying to execute this command using PLINK v1.07 with this command. My file contains only SNPs.

plink --file <input_prefix> --recode HV --out <output_prefix>

which returned this:

ERROR: Problem parsing the command line arguments.

As far as I can tell from the manual, my usage is correct...am I missing something? Thank you!

UPDATE: Error was fixed by using the most current version of PLINK (v1.9) Whoops!