Getting error at the clumping stage of PRSice 2
1
0
Entering edit mode
3.0 years ago
m.shoaib • 0

Hi,

I using PRSice 2 and getting error at the clumping stage, total variants included by PRSice are 4517408

Rscript PRSice.R\
    --prsice PRSice_linux\
    --dir /home/shoaib/projects/base-files/T2test/PRSice\
    --base T2baseNoBMI.txt\
    --target T2basePRS#\
    --snp rsID\
    --chr CHR\
    --bp BP\
    --A2 A2\
    --A1 A1\
    --stat BETA\
    --binary-target T\
    --pvalue Pvalue\
    --pheno pheno-cas-cont.txt\
    --extract PRSice.valid\
    --no-clump\
    --out PRSice-res\

ERROR

Start performing clumping
std::bad_alloc
Begin plotting
Current Rscript version = 2.3.3
Error in fread(paste0(argv$out, ".summary"), data.table = F) :
  File 'PRSice.summary' does not exist or is non-readable.

when I am using --no-clump, it's working alright

Please suggest how to avoid error

Thanks

PRSice2 UKBB R • 1.3k views
ADD COMMENT
2
Entering edit mode
3.0 years ago
Sam ★ 4.8k

Not enough memory. This is something I am trying to solve at the moment. Using single thread can sometime solve the problem. If not, I will have to generate a copy where it uses a less aggressive algorithm for memory storage.

ADD COMMENT
0
Entering edit mode

I am reserving two cores for the analysis

#SBATCH --ntasks=2
#SBATCH --mem-per-cpu=500M

You think, reserving large memory would probably solve the issue?

ADD REPLY
1
Entering edit mode

Definitely. 500M x 2 is just 1G. That isn't enough even for genotyped data with 500k SNPs. You will need almost 100G for this.

ADD REPLY

Login before adding your answer.

Traffic: 2567 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6