Using Plink, get allele frequencies for a subset of individuals

Question

[Mr Locuace]

Using Plink, I would like to calculate allele frequencies for a subset of individuals (cases) from a total cohort of 188 individuals (92 cases + 96 controls). I have proper .ped and .map files.

I have tried multiple options but I am not able to subset data. These are two typical ways I have tried:

1)

./plink --file /path/in_data --chr 1-22 --allow-extra-chr --filter  /path/cases.raw 1 --freq --make-bed --out out_data_cases

The .raw file has this format:

CAV-001 CAV-001 1
CAV-002 CAV-002 1
CAV-003 CAV-003 0
CAV-004 CAV-004 1

where the first column is Family ID, second column Individual ID, and in third column 1 are cases and 0 controls. I want to subset cases. All Family ID = Individual ID

2)

./plink --file /path/in_data --chr 1-22 --allow-extra-chr --keep  /path/cases.txt --freq --make-bed --out out_data_cases

The cases.txt file includes columns 1 and 2 from the .raw file.

This is what I get in the .log file (some paths are not shown):

16384 MB RAM detected; reserving 8192 MB for main workspace.
.ped scan complete (for binary autoconversion).
Performing single-pass .bed write (868263 variants, 188 people).
....
868263 variants loaded from .bim file.
188 people (0 males, 0 females, 188 ambiguous) loaded from .fam.
Ambiguous sex IDs written to xxxx.nosex
Using 1 thread (no multithreaded calculations invoked).
Before main variant filters, 188 founders and 0 nonfounders present.
Calculating allele frequencies... done.
--freq: Allele frequencies (founders only) written to xxxx.frq
868263 variants and 188 people pass filters and QC.
Note: No phenotypes present.
--make-bed to ....

My question is similar as Cannot remove subjects from Plink files but I have tried what they suggest there, without positive outcome. Please help !

Answer 1

[Kevin Blighe]

My guess is an encoding issue with your hyphens. PLINK may have converted these to something else when you read your data into PLINK initially - have you checked the sample names as they appear in PLINK?

Also check the formatting of the hyphen in your /path/cases.txt file. There are different encodings of hyphens in ASCII.

[Mr Locuace]

Thank you @Kevin Blighe. Plink should recognize the hyphens form the individuals names of the .ped file, as I am able to get the allele frequencies for the whole cohort but not for the cases subset. I did a trial just doing copy & paste from a few individuals from the .fam file (and also from the .ped file) and I get the same output.

[Kevin Blighe]

Yes, you're correct that it should accept hyphens.

I realise that I have never actually used --keep in PLINK. For sample filtering, I have always conducted it outside of PLINK when I have my data in BCF or VCF format, and I filter with a command like this:

bcftools view -Ob --samples-file KeepCases.list MyData.bcf > MyDataCases.bcf

bcftools index MyDataCases.bcf

I then convert these BCFs to PLINK format (> PLINK 1.9 required)

plink --noweb --bcf MyDataCases.bcf --keep-allele-order --indiv-sort file KeepCases.IDSort.list --vcf-idspace-to _ --const-fid --allow-extra-chr 0 --split-x b37 no-fail --make-bed --out MyDataCases

KeepCases.list is jst a single column file of sample IDs KeepCases.IDSort.list is the standard 2-column format, as per your own file. It instructs PLINK to read in the data and maintain the sample ordering as per KeepCases.IDSort.list

I don't know if this is an option for you or not!

[Mr Locuace]

Thank you @Kevin Blighe, I will try that. I have Plink 1.9

Answer 2

[Maxime Lamontagne]

Create a file with only the FID and IID of the subjects you want to keep (NewFile.txt). Then try this :

./plink --file /path/in_data --chr 1-22 --allow-extra-chr --keep NewFile.txt --pheno /path/cases.txt --freq --make-bed --out out_data_case

[Mr Locuace]

Thank you @Maxime Lamontagne. I got the same output, just a .log file and a .nosex file with the 188 individuals. In addition, I got an error of a supposedly Duplicate ID (I check the files and I don't have duplicates):

Error: Duplicate ID 'CAV-131 CAV-131'.

[Maxime Lamontagne]

Try this :

./plink --file /path/in_data --chr 1-22 --allow-extra-chr --keep NewFile.txt --make-bed --out Temp1 --allow-no-sex

./plink --bfile Temp1 --pheno /path/cases.txt --freq --out out_data_case --allow-no-sex

[Mr Locuace]

Thank you very much @Maxime Lamontagne. Now it partially works. Somehow I get the allele frequencies for 5 individuals, but not for the 92 cases and my NewFile.txt has 92 individuals.

[Maxime Lamontagne]

I gues one of your file is incorrectly formatted. If you want, I could take a quick look at them.

Answer 3

[pfs]

Your -keep file format is wrong. Create one -keep file that only includes the people with a '1' identifier. Then run these two commands.

1. ./plink --file /path/in_data --keep new_file.txt --freq --out control
2. /plink --file /path/in_data --remove new_file.txt --freq --out case

[Mr Locuace]

Thank you @pfs, but I already mentioned that I did what you suggest in 2). Either you use --keep without identifier (.txt file with 2 columns) or you use --filter with identifier (.raw file with 3 columns, the third one has the identifier). Even if I use the --filter with only cases (1s), I get the same output.

[pfs]

Can you double check the path to the --keep file and make sure this file has the expected number of lines (wc -l keep_file). What version of Plink are you running?