|
Post by scottfunk on Jan 6, 2020 21:43:48 GMT
When using --make-grm-part to compute the GRM for a sizable dataset (N~430,000), I have used 50 parts and merged them according to GCTA's documentation, i.e.:
cat data.part_50_*.grm.id > data.grm.id cat data.part_50_*.grm.bin > data.grm.bin cat data.part_50_*.grm.N.bin > data.grm.N.bin
I have checked and all 50 parts completed successfully, and the merging completed successfully. I want to do relatedness pruning using --cutoff 0.05 but I when I use:
gcta64 --grm data --grm-cutoff 0.05 --make-grm --out data_filtered
I get the message "Error: The GRM id and GRM binary is not matching". Curiously, I only get this error with my largest dataset (data.grm.bin is 328Gb). Other datasets with a smaller sample size work just fine with this pipeline.
I'm not sure where to begin troubleshooting. Any tips or information about this error message would be greatly appreciated.
|
|
|
Post by melkamu on Mar 26, 2024 0:25:40 GMT
I have the same problem with this. any tips please.
the error message is here:
Error: The IDs in GRM and the IDs in the GRM binary file do not match [GRM_GCTA_UKB_0.6_All_45Ksamples]
An error occurs, please check the options or data
|
|