Trouble with PLINK's logistic regression analysis and covariates

Question

[Diego.Morales]

Hi,

I am new to PLINK and I am having trouble using one of the commands. I am trying to perform logistic regression analysis on several bed, bim and fam files. Here is the command I put in:

plink \
  --bfile chr22.dose \
  --logistic \
  --covar COPD_NHW_Phenotype_PC_Final.txt \
  --covar-name FamilyID,PatientID, FatherID, MotherID, Sex, Affection, Status, Age_Enroll, EV=6.02683, EV=2.8176, EV=1.74892, EV=1.65407, EV=1.55682, EV=1.46005, EV=1.44006, EV=1.39075, EV=1.19516, EV=1.17052, \
  --out chr22.dose.output

I get back the following message and error:

Logging to chr22.dose.output.log.
Options in effect:
  --bfile chr22.dose
  --covar COPD_NHW_Phenotype_PC_Final.txt
  --covar-name FamilyID,PatientID, FatherID, MotherID, Sex, Affection, Status, Age_Enroll, EV=6.02683, EV=2.8176, EV=1.74892, EV=1.65407, EV=1.55682, EV=1.46005, EV=1.44006, EV=1.39075, EV=1.19516, EV=1.17052,
  --logistic
  --out chr22.dose.output

516866 MB RAM detected; reserving 258433 MB for main workspace.
Allocated 145368 MB successfully, after larger attempt(s) failed.
652195 variants loaded from .bim file.
5269 people (0 males, 0 females, 5269 ambiguous) loaded from .fam.
Ambiguous sex IDs written to chr22.dose.output.nosex .
Using 1 thread (no multithreaded calculations invoked.

Error: --covar file doesn't have a header line for --covar-name.

I've searched online and I can't find anything on this error. Can anyone please help me out?

Diego

[Maxime Lamontagne]

Post the first ten lines of COPD_NHW_Phenotype_PC_Final.txt.

[Diego.Morales]

Family ID       Patient ID      Father ID       Mother ID       Sex     Affection Status        Age_Enroll      EV = 6.02683    EV = 2.8176     EV = 1.74892    EV = 1.65407    EV = 1.55682    EV = 1.46005    EV = 1.44006    EV = 1.39075 
COPDGene_A01220 WG0117516-DNAA11_GS005261413RD                  1       0       57.5    -0.007512605    -0.012820598    -0.004879983    0.025435055     0.004947331     -0.007355293    -0.003296038    0.003340138     -0.032716866    -0.01
COPDGene_A04559 WG0117509-DNAG03_GS005291934RD                  1       0       52.1    -0.01147134     0.051274805     -0.007390259    0.005100506     0.005220221     -0.001276495    -0.003518743    -0.001657514    -0.013286697    -0.02
COPDGene_A05032 WG0117520-DNAD01_GS00522201XRD                  0       1       66.9    -0.006408793    0.001497303     -0.012935682    -0.00470666     0.003673016     -0.002200746    -0.005253198    0.013026227     0.004765291     -0.00
COPDGene_A05779 WG0117554-DNAG02_S-001023023                    0       1       57.8    -0.000327937    -0.009131555    0.01642668      0.012317617     -0.013601817    0.003708468     0.009020178     0.016711946     0.000854154     -0.01
COPDGene_A07157 WG0117508-DNAA04_GS005204353RD                  0       1       71.6    -0.00082518     -0.00516709     0.00748709      0.013504962     -0.00020616     0.00611136      0.005039274     0.031565566     0.003620531     0.013
COPDGene_A07385 WG0117552-DNAH04_S-001023165                    1       0       70.9    0.009356437     0.017223722     0.002700628     0.005383449     -0.002617253    -0.011164449    -0.000777176    0.016380215     0.000739365     -0.00C
OPDGene_A07406 WG0117652-DNAB06_S-001036919                    0       1       61.9    -0.005360755    -0.008164267    -0.012285158    0.003808289     0.014703391     0.008186209     -0.013219803    -0.022217068    -0.035664881    -0.00
COPDGene_A07466 WG0117482-DNAE02_GS010769550RD                  0       1       77.1    0.007348406     0.000907873     -0.000553854    0.015803624     0.021634317     0.00952734      0.001227569     0.012463894     -0.004593513    -0.01
COPDGene_A07688 WG0117512-DNAD02_GS005202058RD                  1       1       68.1    -0.004916379    0.025772344     0.011330522     0.00831186      0.006428147     -0.001998264    0.006493551     0.008381912     0.003852375     -0.01

[Maxime Lamontagne]

First column should be FID.

Second column should be IID.

The Father and the Mother ID columns look empty. You can't have empty data. Add -9 or 0 if empty.

[Diego.Morales]

Family ID to EV = 1.39075 are all separate columns

Answer 1

[christopher medway]

You need to name your first two columns 'FID' and 'IID' .

[christopher medway]

I would also lose all the spaces and "=" in the header

[Diego.Morales]

Thanks Christopher,

I've just tried that and got the following error:

Error: Missing --covar-name token in --covar file header line.

I've checked the top line of my covariate file, and don't see anything different there from what I put in my command line.

[christopher medway]

Post up your command line arguments and the header of your covar file again

[Diego.Morales]

The command:

plink \
  --bfile chr22.dose \
  --logistic \
  --covar COPD_NHW_Phenotype_PC_Final.txt \
  --covar-name FID IID Father_ID Mother_ID Sex Affection_Status Age_Enroll EV6.02683 EV2.8176 EV1.74892 EV1.65407 EV1.55682 EV1.46005 EV1.44006 EV1.39075 EV1.19516 EV1.17052 \
  --out chr22.dose.output

The header:

FID     IID     Father_ID       Mother_ID       Sex     Affection_Status        Age_Enroll      EV6.02683       EV2.8176        EV1.74892       EV1.65407       EV1.55682       EV1.46005       EV1.44006       EV1.39075    COPDGene_A01220 WG0117516-DNAA11_GS005261413RD                  1       0       57.5    -0.007512605    -0.012820598    -0.004879983    0.025435055     0.004947331     -0.007355293    -0.003296038    0.003340138     -0.03
COPDGene_A04559 WG0117509-DNAG03_GS005291934RD                  1       0       52.1    -0.01147134     0.051274805     -0.007390259    0.005100506     0.005220221     -0.001276495    -0.003518743    -0.001657514    -0.01COPDGene_A05032 WG0117520-DNAD01_GS00522201XRD                  0       1       66.9    -0.006408793    0.001497303     -0.012935682    -0.00470666     0.003673016     -0.002200746    -0.005253198    0.013026227     0.004
COPDGene_A05779 WG0117554-DNAG02_S-001023023                    0       1       57.8    -0.000327937    -0.009131555    0.01642668      0.012317617     -0.013601817    0.003708468     0.009020178     0.016711946     0.000COPDGene_A07157 WG0117508-DNAA04_GS005204353RD                  0       1       71.6    -0.00082518     -0.00516709     0.00748709      0.013504962     -0.00020616     0.00611136      0.005039274     0.031565566     0.003
COPDGene_A07385 WG0117552-DNAH04_S-001023165                    1       0       70.9    0.009356437     0.017223722     0.002700628     0.005383449     -0.002617253    -0.011164449    -0.000777176    0.016380215     0.000COPDGene_A07406 WG0117652-DNAB06_S-001036919                    0       1       61.9    -0.005360755    -0.008164267    -0.012285158    0.003808289     0.014703391     0.008186209     -0.013219803    -0.022217068    -0.03
COPDGene_A07466 WG0117482-DNAE02_GS010769550RD                  0       1       77.1    0.007348406     0.000907873     -0.000553854    0.015803624     0.021634317     0.00952734      0.001227569     0.012463894     -0.00COPDGene_A07688 WG0117512-DNAD02_GS005202058RD                  1       1       68.1    -0.004916379    0.025772344     0.011330522     0.00831186      0.006428147     -0.001998264    0.006493551     0.008381912     0.003

[christopher medway]

I would lose the "." characters in the header. Make sure all of you columns have a data entry (-9 if missing).

Answer 2

[Maxime Lamontagne]

Add , between each covariate name.

Example from the website: plink --file mydata --covar c.txt --covar-name AGE,BMI-SMOKE,ALC

[Diego.Morales]

Thanks,

Here is the new command I put in:

plink --bfile chr22.dose \
  --logistic \
  --covar COPD_NHW_Phenotype_PC_Final.txt \
  --covar-name FID, IID, Father_ID, Mother_ID, Sex, Affection_Status, Age_Enroll, EV6_02683, EV2_8176, EV1_74892, EV1_65407, EV1_55682, EV1_46005, EV1_44006, EV1_39075, EV1_19516, EV1_17052 \
  --out chr22.dose.output

The heading of the covariate file:

FID     IID     Father_ID       Mother_ID       Sex     Affection_Status        Age_Enroll      EV6_02683       EV2_8176        EV1_74892       EV1_65407       EV1_55682       EV1_46005       EV1_44006       EV1_39075       EV1_19516
COPDGene_A01220 WG0117516-DNAA11_GS005261413RD  -9      -9      1       0       57.5    -0.007512605    -0.012820598    -0.004879983    0.025435055     0.004947331     -0.007355293    -0.003296038    0.003340138     -0.032716866    -0.01
COPDGene_A04559 WG0117509-DNAG03_GS005291934RD  -9      -9      1       0       52.1    -0.01147134     0.051274805     -0.007390259    0.005100506     0.005220221     -0.001276495    -0.003518743    -0.001657514    -0.013286697    -0.02
COPDGene_A05032 WG0117520-DNAD01_GS00522201XRD  -9      -9      0       1       66.9    -0.006408793    0.001497303     -0.012935682    -0.00470666     0.003673016     -0.002200746    -0.005253198    0.013026227     0.004765291     -0.00
COPDGene_A05779 WG0117554-DNAG02_S-001023023    -9      -9      0       1       57.8    -0.000327937    -0.009131555    0.01642668      0.012317617     -0.013601817    0.003708468     0.009020178     0.016711946     0.000854154     -0.01
COPDGene_A07157 WG0117508-DNAA04_GS005204353RD  -9      -9      0       1       71.6    -0.00082518     -0.00516709     0.00748709      0.013504962     -0.00020616     0.00611136      0.005039274     0.031565566     0.003620531     0.013
COPDGene_A07385 WG0117552-DNAH04_S-001023165    -9      -9      1       0       70.9    0.009356437     0.017223722     0.002700628     0.005383449     -0.002617253    -0.011164449    -0.000777176    0.016380215     0.000739365     -0.00
COPDGene_A07406 WG0117652-DNAB06_S-001036919    -9      -9      0       1       61.9    -0.005360755    -0.008164267    -0.012285158    0.003808289     0.014703391     0.008186209     -0.013219803    -0.022217068    -0.035664881    -0.00
COPDGene_A07466 WG0117482-DNAE02_GS010769550RD  -9      -9      0       1       77.1    0.007348406     0.000907873     -0.000553854    0.015803624     0.021634317     0.00952734      0.001227569     0.012463894     -0.004593513    -0.01
COPDGene_A07688 WG0117512-DNAD02_GS005202058RD  -9      -9      1       1       68.1    -0.004916379    0.025772344     0.011330522     0.00831186      0.006428147     -0.001998264    0.006493551     0.008381912     0.003852375     -0.01

and the message and error:

Logging to chr22.dose.output.log.
Options in effect:
  --bfile chr22.dose
  --covar COPD_NHW_Phenotype_PC_Final.txt
  --covar-name FID, IID, Father_ID, Mother_ID, Sex, Affection_Status, Age_Enroll, EV6_02683, EV2_8176, EV1_74892, EV1_65407, EV1_55682, EV1_46005, EV1_44006, EV1_39075, EV1_19516, EV1_17052
  --logistic
  --out chr22.dose.output

516866 MB RAM detected; reserving 258433 MB for main workspace.
Allocated 145368 MB successfully, after larger attempt(s) failed.
652195 variants loaded from .bim file.
5269 people (0 males, 0 females, 5269 ambiguous) loaded from .fam.
Ambiguous sex IDs written to chr22.dose.output.nosex .
Using 1 thread (no multithreaded calculations invoked.
Error: Missing --covar-name token in --covar file header line.

I'm still getting this error.

Diego

[Maxime Lamontagne]

1. Remove the Father_ID and Mother_ID columns if they only have -9 value.
2. Remove all the space in --covar-name FID,IID,Father_ID,Mother_ID,Sex,....
3. Make sure these FID and IID match those in the fam file.
4. Sex column is wrong. It should be 1 and 2 (1=male; 2=female; other=unknown) but you have 0 and 1.
5. Affection_Status: Same thing.

[christopher medway]

Diego,

Why are you including "FID","IID" as covariates. It seems you have an awful lot of covariates, some of which I don't think you really want to include. Try running the same command on just "Age_Enroll" to see what happens. FID and IID are there for the purpose of sample identification only, you don't want to call them into your regression.