--pmergelist error 'Conflicting REF alleles'

8 views
Skip to first unread message

Carson Gehl

unread,
Oct 28, 2025, 1:04:37 PM (13 days ago) Oct 28
to plink2-users
Hello, I have seen others post on here getting this same error while trying to merge pgen files for multiple chromosomes. The job I am submitting: 

plink2 --pmerge-list merge_list.txt --make-pgen --out merged_all_chrs

The error I get:

Start time: Tue Oct 28 11:54:38 2025
385475 MiB RAM detected, ~370187 available; reserving 192737 MiB for main
workspace.
Using up to 4 compute threads.
--pmerge-list: 22 filesets specified.
--pmerge-list: 4503 samples present.
--pmerge-list: Merged .psam written to merged_all_chrs-merge.psam .
--pmerge-list: 22 .pvar files scanned, headers merged.
Concatenation job detected.
Concatenating... 0/135683641 variants complete.
Concatenating... 10000/135683641 variants complete.
Concatenating... 20000/135683641 variants complete.
Concatenating... 30000/135683641 variants complete.
Concatenating... 40000/135683641 variants complete.
Concatenating... 50000/135683641 variants complete.
Concatenating... 60000/135683641 variants complete.
Concatenating... 70000/135683641 variants complete.
Concatenating... 80000/135683641 variants complete.
Concatenating... 90000/135683641 variants complete.
Concatenating... 100000/135683641 variants complete.
Concatenating... 110000/135683641 variants complete.
Error: Conflicting REF alleles for variant 'rs58504087' at 1:3381400.

Based on replies to other threads, it sounds as if this is an issue with the plink build. However, I just updated to the latest plink build and that has not fixed the issue. Any ideas? Thank you. 

Chris Chang

unread,
Oct 28, 2025, 1:30:57 PM (13 days ago) Oct 28
to Carson Gehl, plink2-users
Have you looked at REF/ALT for all variants named rs58504087 on chr1?  This looks like an irregularity in your dataset, rather than “an issue with the plink build”.  If so, —set-all-var-ids may be useful for addressing the irregularity.

(Meanwhile, if you want no-questions-asked concatenation, you can always export to VCF/BCF and use “bcftools concat”.)

--
You received this message because you are subscribed to the Google Groups "plink2-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to plink2-users...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/plink2-users/a67e6535-bde5-4e7e-a980-ccc2b826f446n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages