: "-recode rlist" no longer forgets to generate a. In the development build, -dfam consistently errors out with an appropriate message, instead of possibly segfaulting, when no valid cases are present. meta-analysis-chr-field and -meta-analysis-bp-field flags added. a FILTER key and an INFO key were identical in a BCFv2.2 file, -bcf may have computed the wrong string index for FORMAT:GT, in which case import would fail.Ģ1 Jan: "-extract range" and "-exclude range" no longer error out when their input files contain a chromosome code absent from the current dataset. ld-window-r2 can now be used with -r.ġ9 Feb: -bcf now parses header line IDX fields previously, if e.g. segfault that could occur when the chromosome-set flag prohibited X/Y/XY/MT but that chromosome code was in the dataset anyway an appropriate error message is now printed instead. within.perm output files had mismatched variant IDs), but if you're unsure you should rerun -qfam with the latest build.ġ6 Jun: Fixed -het bug that caused the wrong variants to be skipped when chrM was present but not at the end of the file.Ģ8 Apr: Fixed -allow-extra-chr + -autosome-num/-cow/. The bug was pretty obvious if you were affected by it (the. You can remove this restriction with -native.)Ģ1 Sep: Fixed -logistic bug that could cause the entire analysis to be skipped with an inaccurate "Skipping -linear/-logistic since phenotype is constant" warning.ġ6 Sep: Fixed -qfam bug that occurred when a variant filter (-extract, -maf, etc.) was applied simultaneously, and the remaining variant set was not contiguous (so if your only variant filter was -chr on a single chromosome, this bug did not trigger). (This may make some commands a bit slower. : Linux binaries no longer default to allowing Intel MKL to use processor-specific code paths, since this behavior made some results more difficult to reproduce across different machines. These two cases are distinguished ("in [0.9999995, 1)" vs. (Previously, minor variations were possible when there were p-value ties.)Ģ8 May: Genotyping rate is explicitly logged when equal to, or very close to, 1. : -clump results should now be consistent across operating systems. pedmap is now recognized as an alias for -file. ![]() : -recode recognizes 'Av' and 'tped' as aliases for the 'A-transpose' and 'transpose' formats, respectively. This can be controlled with the -missing-genotype2 flag.Ģ Apr: -make-set and similar flags no longer error out when -allow-extra-chr is specified and a contig is present in its input file but absent from the main dataset. bim files generated by PLINK 2.0 that didn't have all allele codes filled in. ![]() bim reader now defaults to treating '.' as a missing allele code without this, merge could fail with a spurious "variants with 3+ alleles present" error when given. : -ld-window-cm should work properly now.ġ8 Oct: Fixed -clump bug that, when multiple input files were provided, could cause it to fail to open the first file. Until then, BEAGLE 3.3 should be more accurate than PLINK for case/control haplotype association.Ħ: Free database software handles these operations in a more flexible and powerful manner.ħ: Just use gzip/ gunzip for this. Future PLINK versions will be able to import phase and dosage information emitted by other programs the haplotype association commands will be reintroduced when that functionality is in place. We recommend migrating away from PLINK 1 here, but if that's not practical, keep a permanent copy of the PLINK 1.07 binary around and modify the relevant lines of your scripts to explicitly refer to it.ģ: We believe this now has almost no practical value, since the file format it expects is too different from VCF.Ĥ: This was not fully developed in PLINK 1.07, and has been superseded by other IBD analysis packages.ĥ: PLINK's haplotype phasing and imputation algorithms are obsolete. red flags will not be supported by the final 1.90 release. ![]() You're more than welcome to try the new implementations just make sure to check some of your results against PLINK 1.07 if you do so.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |