plink2 segmentation fault (core dump) for --make-bed from --haps files
21 views
Skip to first unread message
E C
Jun 9, 2025, 6:26:21 PM6/9/25
to plink2-users
Hello,
I am trying to convert haps files to bed files, and I keep running into this same error:
I also tried this with a subsample of my dataset (included attached to this), same error:
The number of variants does not seem to matter, nor the number of samples. I've thoroughly check the quality and format of my files and I see nothing amiss. This is my log:
That's all there is. The only thing I can think of is that possibly plink2 doesn't like the way some of my variants are named? I am using this on some atypical haps files that include knockoff variants denoted with a '.k' suffix. A previous workaround I used was to keep the knockoffs and real variants haps files separate, convert to bed files with plink2, then rename the knockoff variants with the proper suffix and merge them together again. This fix is no longer working, as you can see from the first screenshot working on the non-merged files (it was working last week...). Hopefully someone can help or has seen this before! I can share more of my workflow if need be, but essentially the goal is to turn these haps files back to bed files.
I added my merged subset files for reproducibility. Thanks!
I also tried this with a subsample of my dataset (included attached to this), same error:
The number of variants does not seem to matter, nor the number of samples. I've thoroughly check the quality and format of my files and I see nothing amiss. This is my log:
That's all there is. The only thing I can think of is that possibly plink2 doesn't like the way some of my variants are named? I am using this on some atypical haps files that include knockoff variants denoted with a '.k' suffix. A previous workaround I used was to keep the knockoffs and real variants haps files separate, convert to bed files with plink2, then rename the knockoff variants with the proper suffix and merge them together again. This fix is no longer working, as you can see from the first screenshot working on the non-merged files (it was working last week...). Hopefully someone can help or has seen this before! I can share more of my workflow if need be, but essentially the goal is to turn these haps files back to bed files.
I added my merged subset files for reproducibility. Thanks!
Chris Chang
Jun 9, 2025, 7:07:47 PM6/9/25
to E C, plink2-users
Thanks for reporting this. I've replicated the crash, and will try to post a bugfix tonight.
--
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/5555f5e0-0873-4b9b-92e4-d37596da82dcn%40googlegroups.com.
Chris Chang
Jun 10, 2025, 12:49:12 AM6/10/25
to E C, plink2-users
Bugfix has been posted; let me know if you have any problems with --haps.
On Mon, Jun 9, 2025 at 3:26 PM E C <ellen...@gmail.com> wrote:
Reply all
Reply to author
Forward
0 new messages