Get SNP information with association test with PLINK

Question

[landscape95]

Hi everyone,

Does anyone know how to get the SNP information containing CHR ID A1 REF etc with association test on quantitative trait from plink (output file: .qassoc). I done the association test and I want to obtain these information also. In --linear with plink2, I have it, but with plink1.9 association with quantitative trait, I don't have that information (--assoc in plink1.9 runs a lot faster than --linear in plink2 without covariates while --linear with in plink2 gives me memory error:

Error: Out of memory. The --memory flag may be helpful).

I understand that I can use the .bim file, but having integrated information would be very helpful.

Your help is really appreciated!

Answer 1

[chrchang523]

Can you rerun the analysis with covariates with the latest plink2 build, and post the .log file if it still fails?

[landscape95]

Thanks for your prompt reply,

Here is the code and output of plink1.9:

plink --bfile test --assoc --adjust --out output --allow-no-sex --threads 100

PLINK v1.90b4.1 64-bit (30 Mar 2017)
Options in effect:
  --adjust
  --allow-no-sex
  --assoc
  --bfile test
  --out output
  --threads 100

Start time: Wed Feb 20 11:27:04 2019

Random number seed: 1550622424
257682 MB RAM detected; reserving 128841 MB for main workspace.
1133273 variants loaded from .bim file.
230337 people (0 males, 0 females, 230337 ambiguous) loaded from .fam.
Ambiguous sex IDs written to output.nosex .
230337 phenotype values loaded from .fam.
Using 1 thread (no multithreaded calculations invoked).
Before main variant filters, 230337 founders and 0 nonfounders present.
Calculating allele frequencies... done.
Total genotyping rate is 0.992698.
1133273 variants and 230337 people pass filters and QC.
Phenotype data is quantitative.
Writing QT --assoc report to output.qassoc
... done.
--adjust: Genomic inflation est. lambda (based on median chisq) = 1.79545.
--adjust values (1133273 variants) written to
output.qassoc.adjusted .

End time: Wed Feb 20 11:46:25 2019

Here is the code and output with plink2:

plink2 --bfile test --linear --adjust --out output --threads 100

PLINK v2.00a2LM 64-bit Intel (2 Feb 2019)
Options in effect:
  --adjust
  --bfile test
  --glm
  --out output
  --threads 100

Start time: Wed Feb 20 11:35:19 2019

Random number seed: 1550622919
257682 MiB RAM detected; reserving 128841 MiB for main workspace.
Using up to 100 threads (change this with --threads).
230337 samples (0 females, 0 males, 230337 ambiguous; 230337 founders) loaded
from test.fam.
1133273 variants loaded from test.bim.
1 quantitative phenotype loaded (230337 values).
Calculating allele frequencies... done.

Error: Out of memory.  The --memory flag may be helpful.

[chrchang523]

Hmm, that's a bug, it isn't actually running out of memory, instead some other error is occurring that's getting misreported as "out of memory". Trying to replicate this now.

Update: Replicated the crash, will post a bugfix tonight.

[landscape95]

Thank you so much, Chang.