twenty six February: “–recode vcf” header changed for compatibility with VCFtoolspact one thousand Genomes stage step one studies data released into the resources page.
Single-accuracy binary matrix production (age
2 February: –bcf now handles BCFv2.dos nonzero shed genotype and you can prevent-of-vector beliefs (emitted by the age.g. bcftools step one.1) properly. –mendel-multigen is no longer blocked by demand-line parser when a household-oriented connection sample is work with. –lost, –freq, –sturdy, and you can –het today help gzipped production (are the ‘gz’ modifier). –R set in innovation build.
15 January: –keep/–eliminate now performs safely to your newly-upgraded IDs whenever –update-ids is within the exact same run. –mac/–max-mac added to advancement make.
twelve January: –dosage + –covar + –gender today really works securely, and you will an incorrect warning towards PLINK step one.07 execution might have been got rid of. –serving logistic regression not any longer accounts poor p-values to possess very small examples.
(beta 3): –epistasis implemented, with heavy optimization for quantitative traits. “–recode vcf<,-fid,-iid>” now flags reference alleles as “possibly not based on a real reference genome” unless –real-ref-alleles is also specified, and sets ALT alleles to ‘.’ when they are not present in the immediate dataset. –vcf-min-gq and –vcf-min-gp no longer error out when a genotype entry has fewer fields than expected, since the VCF specification explicitly states that this is an acceptable way to represent trailing missing values. –make-bed no longer segfaults when https://datingranking.net/es/fechado/ resorting a file too large to fit in memory.
: –dummy, –show-tags, –neighbour, –mh/–bd/–mh2/–homog and you will –clump-field bugfixes. –q-score-diversity accompanied for amount studies. –split-x/–merge-x ‘no-fail’ modifier added to assistance investigation sales texts.
Digital –length returns bugfix
thirteen December: Case/handle –hardy/–hwe no longer at random excludes a lot of controls away from control-just phase from shot. (It was mainly innocuous, while the all of the-samples take to still has worked, additionally the bug was just attending exist when –hardy/–hwe was in a similar manage given that most other strain such –exclude/–extract. But if you used the UNAFF rows about –robust report getting one thing, you should rerun –sturdy towards the most recent make. ) set-try multiple-review modification today matters nonempty establishes with zero significant variants.
seven December: –to change now applies to the brand new shared sample statistic instead of the additive feeling if former is actually calculated, and no prolonged account linearly genomic-managed p-viewpoints when they would-be incorrect. Particular numeric balance developments for short p-values. –logistic permutation test and –assoc/–model lay-try bugfixes. –linear/–logistic set-ensure that you –amount ‘sex’ modifier used. –amount + –sex mistake message today teaches you you to definitely PLINK 1.07 failed to deal with it flag combination safely.
twenty-five November: Fixed –exclude/–extract recollections administration bug in 2 November create. –get now supporting amount studies. –write-amount no more produces wrong IDs when sample strain is actually used. –dose ‘noheader’ can be taken with ‘list’ when for each and every batch only has one document. –assoc/–design lay-shot followed. –set-missing-var-ids longer allowing addition away from allele brands, and –set-missing-snp-ids/–set-missing-nonsnp-ids retired. UNADJ line viewpoints are now actually proper during the “–to switch gc” profile.
2 November: Enhanced variation ID search speed. g. –make-grm-bin) grew to become predicated on twice-precision interior data, while the ‘bin’ + ‘single-prec’ modifier integration might have been replaced with ‘bin4’.
15 Oct: Repaired an insect in the –vcf’s handling of alternatives that have 10 or even more alternate alleles. –serving Zout don’t segfaults at the end. Merger no longer scrambles centimorgan coordinates.
20 September: Fixed low-rigid –biallelic-only bug when dealing with multiallelic variants. For those who have made use of –biallelic-merely in the place of ‘strict’ towards VCF files which have triallelic versions, i suggest rerunning the fresh new operation on the most recent build. Repaired quantitative trait –assoc bug you to definitely caused it to type for each production range double, and –linear/–logistic mishandling of some datasets with heterozygous haploid calls. –vcf-min-gp set in innovation build.
18 Sep: –no-fid bugfix. Visibility from nonnumeric phenotype chain (elizabeth.grams. ‘NA’) no more push the fresh phenotype as managed once the decimal. –output-missing-phenotype now welcomes nonnumeric strings. “–amount number” now works safely having numerous batches. –oxford-single-chr flag put in allow packing away from unmarried-chromosome .gen data files which have ignorable SNP ID industry viewpoints. –genome no more does not declaration some parent-children relationship. –update-alleles was still forgotten several suits of missing genotypes; it’s been repaired.