search

omerwe / polyfun

PolyFun (POLYgenic FUNctionally-informed fine-mapping)
MIT License
89 stars 22 forks source link

plink reference files #14

Closed bschilder closed 4 years ago

bschilder commented 4 years ago

Hi Omer,

So I finally got my Conda issues sorted out! Since then things have been running pretty smoothly, until I got to this step. I'm realizing it's because I haven't provided any plink reference files. In the documentation it says to use a closely related reference database (e.g. 1000 Genomes), but since I'm using UKBB for every other step, is there somewhere I can find the UKBB ref files? (either on your server or UKBB's).

Thanks in advance! Brian

 (polyfun_venv) [schilb03@lh03c03 Fine_Mapping]$ python ./echolocatoR/tools/polyfun/polyfun.py --compute-ldscores --output-prefix ./Data/GWAS/Nalls23andMe_2019/_genome_wide/PolyFun/output/PD_GWAS --bfile-chr /sc/orga/projects/pd-omics/tools/polyfun/annotations/baselineLF2.2.UKB/reference.  --allow-missing
*********************************************************************
* PolyFun (POLYgenic FUNctionally-informed fine-mapping)
* Version 1.0.0
* (C) 2019 Omer Weissbrod
*********************************************************************

[WARNING]  no ld-wind argument specified.  PolyFun will use --ld-cm 1.0
Traceback (most recent call last):
  File "./echolocatoR/tools/polyfun/polyfun.py", line 898, in <module>
    check_files(args)
  File "./echolocatoR/tools/polyfun/polyfun.py", line 130, in check_files
    get_file_name(args, 'bim', chr_num, verify_exists=True)
  File "./echolocatoR/tools/polyfun/polyfun.py", line 238, in get_file_name
    raise IOError('%s file not found: %s'%(file_type, file_name))
OSError: bim file not found: /sc/orga/projects/pd-omics/tools/polyfun/annotations/baselineLF2.2.UKB/reference.1.bim
bschilder commented 4 years ago

Actually, I'm just now realizing why sharing UKBB might be problematic. Don't those plink files contain individual level genotype data? In which case I imagine I'd have to apply to access to them directly from UKBB.

omerwe commented 4 years ago

Yes, exactly...

On Mon, Nov 25, 2019 at 3:39 PM Brian M. Schilder notifications@github.com wrote:

Actually, I'm just now realizing why sharing UKBB might be problematic. Don't those plink files contain individual level genotype data? In which case I imagine I'd have to apply to access to them directly from UKBB.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/omerwe/polyfun/issues/14?email_source=notifications&email_token=ACNCB47U443PTB6FXTK5ABDQVQZWNA5CNFSM4JROIUCKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEFDXCWY#issuecomment-558330203, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACNCB44VCRFMZHHZ6Y6KGU3QVQZWNANCNFSM4JROIUCA .

bschilder commented 4 years ago

Got it, thanks! I'll go ahead and use 1KG