Timeline for PLINK clump behavior on missing SNPs?
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 1, 2018 at 4:07 | history | edited | gringer♦ | CC BY-SA 3.0 |
added workflow
|
| Jan 18, 2018 at 18:31 | comment | added | Christopher Chang | Thanks for mentioning this, I'll investigate. I noticed that the recommended pipelines now include a "bcftools sort" step after VCF file generation even though that VCF would be position-sorted, so there may be a poorly documented assumption about how multiple variants sharing the same bp coordinate must be sorted, and/or an assumption about chromosome order. | |
| Jan 17, 2018 at 21:14 | comment | added | gringer♦ | Yes, but the plink-generated bgz-VCF files didn't work properly with the Michigan Imputation Server last time I tried, and it's a real pain testing that out due to the queue length. | |
| Jan 17, 2018 at 19:59 | comment | added | Christopher Chang | fyi, plink can directly generate bgzipped VCF files (--recode vcf bgz), and it'll use multiple threads to do the compression. | |
| Jan 12, 2018 at 1:55 | history | answered | gringer♦ | CC BY-SA 3.0 |