Generating polygenic scores without a target data phenotype

[JessG1981]

Hi

I am trying to generate polygenic scores using PRSice v2 without phenotype data for the target population. i.e. I just want to use the GWAS summary statistics (and specifically the effect sizes and p values for the various SNPs) to use the SNP data from my target population (a very small population, less than 50 participants) to generate polygenic index values for each of my participants. I then want to use these polygenic index values to run various correlations etc.

However, if I try to run PRSice without any phenotype data (by inserting "NA" into the phenotype column of the .fam file for each participant, I get an error message "No phenotype presented
Error:
Execution halted"

I have the following inputs to use with PRSice:

• Base GWAS summary statistics file
• Target SNP data (i.e., the alleles for SNPs within my study population)
• Covariates (ancestry principal components)
• A linkage disequilibrium reference dataset

I guess my question therefore is: Is it possible to run PRSice to generate individual polygenic scores using GWAS data without a target phenotype? Just to create a set of numbers that summate the SNP effect sizes for my sample population, that I can then use as a variable in further analyses?

Thank you in advance for help.

[choishingwan]

Not sure if I have replied this: You need the —no-regress flag. Though in this scenario, it might be better to use pseudo validation from lassosum, auto from ldpred2 or the cv function from megaprs Sam

…

On Thu, Jun 20, 2024 at 10:10 AM JessG1981 ***@***.***> wrote: Hi I am trying to generate polygenic scores using PRSice v2 *without phenotype data for the target population*. i.e. I just want to use the GWAS summary statistics (and specifically the effect sizes and p values for the various SNPs) to use the SNP data from my target population (a very small population, less than 50 participants) to generate polygenic index values for each of my participants. I then want to use these polygenic index values to run various correlations etc. However, if I try to run PRSice without any phenotype data (by inserting "NA" into the phenotype column of the .fam file for each participant, I get an error message "No phenotype presented Error: Execution halted" I have the following inputs to use with PRSice: - Base GWAS summary statistics file - Target SNP data (i.e., the alleles for SNPs within my study population) - Covariates (ancestry principal components) - A linkage disequilibrium reference dataset I guess my question therefore is: Is it possible to run PRSice to generate individual polygenic scores using GWAS data without a target phenotype? Just to create a set of numbers that summate the SNP effect sizes for my sample population, that I can then use as a variable in further analyses? Thank you in advance for help. — Reply to this email directly, view it on GitHub <#358>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAJTRYWK7VMYF724YGT7VZ3ZILPFBAVCNFSM6AAAAABJUBVFD2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGM3DINJWGAZTQOA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>