Malformed data files created by particular dated merges no further end up in segfaults

Malformed data files created by particular dated merges no further end up in segfaults

dos Sep: –amount ‘noheader’ will be now work properly whenever specific phenotypes was shed. –vcf + –vcf-filter out today parses semicolon-delimited Filter out areas precisely. –vcf-min-gq banner added to advancement build. –exclude/–extract no longer have dreadful performance whenever the chief dataset additionally the –exclude/–extract variation listing has actually an incredible number of ‘.’ entries.

28 August: –recode HV[-1chr] today always uses ‘0’ given that missing phenotype password, because the Haploview cannot accept -9. –write-covar + –with-phenotype not segfaults towards case/handle phenotypes, and multichar allele .ped loader no more segfaults on nonstandard tri/quadallelic variations. –meta-investigation set in innovation create.

fourteen August: Fixed an excellent typo on the eleven August build and that triggered covariate packing and some most other features to hang.

eleven August: –linear/–logistic not uses a pram Huber-Light practical error estimator when groups try defined. –output-chr now really works securely which have –make-bed if type in .bim is actually unsorted, and you may a long allele code within the an enthusiastic unsorted .bim not grounds –make-bed so you’re able to segfault.

step 1 August: –dose logistic regression bugfix. –make-just-bim and you may –make-just-fam flags added. .bim/.fam files is now able to become processed rather than an associated .sleep significantly less than specific issues. –recode-allele now work safely having A great-transpose function.

18 July: Text filesets that have one another multiple-profile allele rules and you may an unsorted .map file no further result in the autoconverter in order to freeze. –pfilter should now consistently filter ‘NA’ entries. –serving chromosome password production bug fixed. –read-freq today tons A1 allele requirements if they are missing throughout the main dataset, unlike erroring call at that state. –show-tags placed into development make.

Earliest Cochran-Mantel-Haenszel and Breslow-Time evaluating placed into creativity build

cuatro July: Fixed two combine pests and that probably brought about research over the past partners samples of a feedback otherwise blended fileset to get mishandled. 90 alpha/beta stimulates, and most two trials were with it, we recommend redoing the latest blend with a more recent build. (Extremely merges was in fact unchanged, however, best secure than just sorry.)

For those who have merged filesets generated having earlier PLINK step one

1 July: Augment recently brought (sorry about that) –data/–gen/–test order-line parsing insect. –indep not misreports what amount of pruned versions whenever indeed there try detailed multicollinearity. –fst, –homog, and you can –oxford-pheno-label flags added.

twenty-seven June: “–recode oxford” not dumps incorrect IDs when utilized in a similar work with once the a sample filter. –fill-missing-a2 banner added. –recode ‘A-transpose’ and ‘include-alt’ modifiers additional. “–het quick-sample” function additional.

20 June: Repaired –mendel zero chromosome password/segfault bug. –merge-record no more need a reference fileset. –fast-epistasis + –synchronous bugfix. –indep-pairphase and QFAM attempt finished in development create.

10 Summer: –merge-setting step one today correctly merges lost calls that have one nonmissing call. –r/–r2 chromosome edge approaching bugfixes. ‘0X’/’0Y’/’0M’ chromosome rules produced of the Oxford units are actually approved, and now have supported by –output-chr. –vcf-half-telephone call banner put into regulate management of ‘0/.’ VCF GT philosophy, and you can default behavior is starting to become ‘error’ form to force a mindful ent generate.

5 Summer: –ld-snp-record and you can multipass –r/–r2 bugfixes. Nonstandard ‘0/.’ and you may equivalent VCF GT community beliefs are now processed because the once they did not have the fresh behind ‘/.’, in the place of leading to a great segfault. (Management of this situation could be configurable afterwards, tune in.) –linear/–logistic permutation bugfix (permutation achievements count array wasn’t initialized to all the-zero). –amount linear regression added to advancement generate.

3 June: Gzipper no further deletes the new productivity document when getting expected in order to append (this was resulting in larger n-pass calculations such as for example –r2 gz to only secure the production of the past citation). Fixed –genome double-missing-name addressing bug, and you may a great –r/–r2 mixed autosomal/nonautosomal data handling bug. “–serving are present” and you will –write-dosage put into creativity make.

e-related segfaults. “–r2 dprime” missing data handling bugfix. Corrected misnamed –filter-attrib[-indiv] flags to –attrib[-indiv], fixed a positive matching bug, and added support for gzipped attribute files. “–recode tab” don’t produces room regarding the heading range. –stops now has good ‘no-pheno-req’ modifier hence takes away the unnecessary phenotype criteria. –annotate put into innovation generate.


Vaša adresa e-pošte neće biti objavljena.