PLINK - Sample(s) present with no genotype data

Question

[liorglic]

Hello,
I am running PLINK v1.90b6.12 , trying to perform a PCA analysis on a vcf file.
I run the following command:

plink --vcf all.maf005.rehead.vcf --pca --allow-extra-chr

and getting the ouput:

PLINK v1.90b6.12 64-bit (28 Oct 2019)          www.cog-genomics.org/plink/1.9/
(C) 2005-2019 Shaun Purcell, Christopher Chang   GNU General Public License v3
Logging to plink.log.
Options in effect:
  --allow-extra-chr
  --pca
  --vcf all.maf005.rehead.vcf

128770 MB RAM detected; reserving 64385 MB for main workspace.
--vcf: plink-temporary.bed + plink-temporary.bim + plink-temporary.fam written.
17 variants loaded from .bim file.
3820 people (0 males, 0 females, 3820 ambiguous) loaded from .fam.
Ambiguous sex IDs written to plink.nosex .
Using up to 15 threads (change this with --threads).
Before main variant filters, 3820 founders and 0 nonfounders present.
Calculating allele frequencies... done.
Total genotyping rate is 0.198984.
17 variants and 3820 people pass filters and QC.
Note: No phenotypes present.
Relationship matrix calculation complete.
Error: Sample(s) present with no genotype data.  Use --mind to filter them out.

However, there is no missing data in my vcf - all genotypes are "./." or "1/1".
If I run:

plink --vcf all.maf005.rehead.vcf --pca --allow-extra-chr --mind

the ouput is:

PLINK v1.90b6.12 64-bit (28 Oct 2019)          www.cog-genomics.org/plink/1.9/
(C) 2005-2019 Shaun Purcell, Christopher Chang   GNU General Public License v3
Logging to plink.log.
Options in effect:
  --allow-extra-chr
  --mind
  --pca
  --vcf all.maf005.rehead.vcf

128770 MB RAM detected; reserving 64385 MB for main workspace.
--vcf: plink-temporary.bed + plink-temporary.bim + plink-temporary.fam written.
17 variants loaded from .bim file.
3820 people (0 males, 0 females, 3820 ambiguous) loaded from .fam.
Ambiguous sex IDs written to plink.nosex .
Error: All people removed due to missing genotype data (--mind).
IDs written to plink.irem .

Can anyone help figure out this issue?
Thanks!

Answer 1

[chrchang523]

./. encodes a missing genotype.

[liorglic]

Stupid me - I thought ./. means ref. Now I see it should be 0/0. The vcf was created with paftools call - apparently it puts and unknown genotype whenever it doesn't find a variant. Thanks anyway.