Calculating polygenic score using plink

I am trying to calculate PRS in test sample after getting best SNPs from PRSice 2 with highest R2 My target data is in the form of bed,bim, fam for 22 chromosomes, I want to use plink for specific p-value threshold in this scenario,

plink2 \
    --bfile T2test3 \                            ## For one chromosome only
    --score T2baseNoBMI.uniq.txt 2 4 6 header \
    --q-score-range range_list SNP.pvalue \
    --extract EUR.valid.snp \
    --out EUR

This is working for just a single chromosome, I want to perform my analysis considering all autosomes and need a final final file with IIDs and their corresponding PRS based on all chromosome SNP information,

Please help

Thanks

prsice2 ukbb plink

• 4,050 views • link

updated 3.7 years ago by Ram

• written 4.7 years ago by m.shoaib •

1 answer

You will need to combine the genotype file. Not sure if --score works with --bmerge. You can have a try.

Alternatively, as you are already using PRSice, do

--bar-levels <p-value threshold> --no-full --fastscore and you should in theory get the PRS at the specific p-value threshold

• 0 views • link

written 4.7 years ago by Sam

Ok. PRSice seems more convenient for me at this stage. So, In order to calculate the PRS in my test data, can I simply provide, GWAS base data, target data with these options, -bar-levels <p-value threshold> --no-full --fastscore? I cannot provide phenotype file because my independent sample comprised of 3 different phenotypic categories and I simply need to calculate the PRS at the specific threshold for specific individuals. I donot need bar plot and models with R2. Thanks

• 0 views • link

written 4.7 years ago by m.shoaib •

if you don't need the plots or R2, then add in --no-regress

• 0 views • link

written 4.7 years ago by Sam

Thank you very much for your response!

• 0 views • link

written 4.7 years ago by m.shoaib •