search

rgcgithub / regenie

regenie is a C++ program for whole genome regression modelling of large genome-wide association studies.
https://rgcgithub.github.io/regenie
Other
187 stars 55 forks source link

minor issue: --sex-specific #316

Closed dvh13 closed 2 years ago

dvh13 commented 2 years ago

Minor issue.

v3.1.2

--sex-specific works if you specify it just in step 1 and not step 2.

if you specifiy --sex-specific for both step 1 and step 2, doesnt work.

Suggest either mention this in the documentation, or force its use just for step 1.

regards, david

joellembatchou commented 2 years ago

Hi David,

Tested with the toy data sets in the example/ folder and it runs fine specifying --sex-specific in step 1 & 2. Could you give more information as to the error you are getting?

Thanks, Joelle

dvh13 commented 2 years ago

this is the error, step 1 runs fine, then step 2 fails i will just re-run it without the --sex-specific male for step 2 thanks, david

@.*** male-only]$ cat 2022_07_04_male-only_fit_binary_out.log Start time: Mon Jul 4 14:37:50 2022

          |=============================|
          |      REGENIE v3.1.2.gz      |
          |=============================|

Copyright (c) 2020-2022 Joelle Mbatchou, Andrey Ziyatdinov and Jonathan Marchini. Distributed under the MIT License. Compiled with Boost Iostream library. Using Intel MKL with Eigen.

Log of output saved in file : 2022_07_04_male-only_fit_binary_out.log

Options in effect: --step 1 \ --bed /data/Blizard-VanHeelLab/GenesandHealth/GSAv3EAMD/Jul2021_44k_TOPMED-r2_Imputation_b38/bfile_forSAIGEgrm_44396_chipgenotypes_indep-pairwise_500_50_0.2_LDpruned_NotChrY \ --sex-specific male \ --covarFile ../GNH.44190.noEthnicOutliers.covariates.20PCs.tab \ --covarExcludeList S1QST_Gender \ --phenoFile ../2022_06_version005_3digitICD10_1to1_42029withbothICD10andGSAJul2021.txt \ --phenoExcludeList PseudoNHS_2022_02_08 \ --bsize 1000 \ --lowmem \ --lowmem-prefix tmp_rg \ --minCaseCount 100 \ --bt \ --gz \ --verbose \ --out 2022_07_04_male-only_fit_binary_out

Fitting null model

etc etc

0.5 : Rsq = 0.0215823, MSE = 0.00845346, -logLik/N = 0.0444422 0.75 : Rsq = 0.0275038, MSE = 0.00843391, -logLik/N = 0.0439637 0.99 : Rsq = 0.0499191, MSE = 0.00832941, -logLik/N = 0.0425981<- min value

List of blup files written to: [2022_07_04_male-only_fit_binary_out_pred.list]

Elapsed time : 2362.72s End time: Mon Jul 4 15:17:13 2022

@.*** male-only]$ cat regenie_step2_binarytraits_QSUB_ARRAY.txt.o2460185.62 Start time: Mon Jul 4 20:43:09 2022

          |=============================|
          |      REGENIE v3.1.2.gz      |
          |=============================|

Copyright (c) 2020-2022 Joelle Mbatchou, Andrey Ziyatdinov and Jonathan Marchini. Distributed under the MIT License. Compiled with Boost Iostream library. Using Intel MKL with Eigen.

Log of output saved in file : 2022_07_04_male-only_GNH_binarytrait_GWAS_firth_ICD10__B02.log

Options in effect: --step 2 \ --pgen /data/scratch/hmw208/regenietemp/chrALLincX.dose.merged_INFO0.3_MAF0.00001_F_MISSING0.2.8bit.sorted \ --sex-specific male \ --covarFile ../GNH.44190.noEthnicOutliers.covariates.20PCs.tab \ --covarExcludeList S1QST_Gender \ --phenoFile ../2022_06_version005_3digitICD10_1to1_42029withbothICD10andGSAJul2021.txt \ --phenoCol ICD10__B02 \ --bsize 1000 \ --minMAC 20 \ --minINFO 0.5 \ --bt \ --firth \ --approx \ --pThresh 0.01 \ --pred 2022_07_04_male-only_fit_binary_out_pred.list \ --gz \ --verbose \ --threads 4 \ --out 2022_07_04_male-only_GNH_binarytrait_GWAS_firth_ICD10__B02

Association testing mode with fast multithreading using OpenMP

On 1 Jul 2022, at 18:03, Joelle Mbatchou @.**@.>> wrote:

Hi David,

Tested with the toy data sets in the example/ folder and it runs fine specifying --sex-specific in step 1 & 2. Could you give more information as to the error you are getting?

Thanks, Joelle

— Reply to this email directly, view it on GitHubhttps://github.com/rgcgithub/regenie/issues/316#issuecomment-1172545877, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASHESISRMNJPLKUIUECPF6LVR4QE3ANCNFSM5ZVTDB2Q. You are receiving this because you authored the thread.Message ID: @.***>

joellembatchou commented 2 years ago

Hi David,

It seems the issue stems from reading sex information in the PSAM file /data/scratch/hmw208/regenietemp/chrALLincX.dose.merged_INFO0.3_MAF0.00001_F_MISSING0.2.8bit.sorted.psam. Is sex information included in the file (so male samples with 1 in the SEX column?)

Cheers, Joelle

dvh13 commented 2 years ago

ah, the sex is NA for all subjects in the stage 2 psam file

i had been providing sex as a covariate instead

guess I must have left in sex in the stage 1 hard genotyped chip data plink files...

On 5 Jul 2022, at 17:05, Joelle Mbatchou @.**@.>> wrote:

Hi David,

It seems the issue stems from reading sex information in the PSAM file /data/scratch/hmw208/regenietemp/chrALLincX.dose.merged_INFO0.3_MAF0.00001_F_MISSING0.2.8bit.sorted.psam. Is sex information included in the file (so male samples with 1 in the SEX column?)

Cheers, Joelle

— Reply to this email directly, view it on GitHubhttps://github.com/rgcgithub/regenie/issues/316#issuecomment-1175230740, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASHESIXJPZ5DPRQQ6Y5MR73VSRMLJANCNFSM5ZVTDB2Q. You are receiving this because you authored the thread.Message ID: @.***>