- Update to 1.10.2
* This release fixes crashes reported on files including integer
INFO tags with values outside the range officially supported
by VCF. It also fixes a bug where invalid BCF files would be
created if such values were present.
- Update to 1.10.0
+ Numerous bug fixes, usability improvements and sanity checks were added to prevent common user errors.
+ The -r, --regions (and -R, --regions-file) option should never create unsorted VCFs or duplicates records again. This also fixes rare cases where a spanning deletion makes a subsequent record invisible to bcftools isec and other commands.
+ Additions to filtering and formatting expressions
* support for the spanning deletion alternate allele (ALT=*)
* new ILEN filtering expression to be able to filter by indel length
* new MEAN, MEDIAN, MODE, STDEV, phred filtering functions
* new formatting expression %PBINOM (phred-scaled binomial probability), %INFO (the whole INFO column), %FORMAT (the whole FORMAT column), %END (end position of the REF allele), %END0 (0-based end position of the REF allele), %MASK (with multiple files indicates the presence of the site in other files)
+ New plugins
* +gvcfz: compress gVCF file by resizing gVCF blocks according to specified criteria
* +indel-stats: collect various indel-specific statistics
* +parental-origin: determine parental origin of a CNV region
* +remove-overlaps: remove overlapping variants.
* +split-vep: query structured annotations such INFO/CSQ created by bcftools/csq or VEP
* +trio-dnm: screen variants for possible de-novo mutations in trios
+ annotate
* new -l, --merge-logic option for combining multiple overlapping regions
+ call
* new bcftools call -G, --group-samples option which allows grouping samples into populations and applying the HWE assumption within but not across the groups.
+ csq
* significant reduction of memory usage in the local -l mode for VCFs with thousands of samples and 20% reduction in the non-local haplotype-aware mode.
* fixes a small memory leak and formatting issue in FORMAT/BCSQ at sites with many consequences
* do not print protein sequence of start_lost events
* support for "start_retained" consequence
* support for symbolic insertions (ALT="<INS...>"), "feature_elongation" consequence
* new -b, --brief-predictions option to output abbreviated protein predictions.
+ concat
* the --naive command now checks header compatibility when concatenating multiple files.
+ consensus
* add a new -H, --haplotype 1pIu/2pIu feature to output first/second allele for phased genotypes and the IUPAC code for unphased genotypes
* new -p, --prefix option to add a prefix to sequence names on output
+ +contrast
* added support for Fisher's test probability and other annotations
+ +fill-from-fasta
* new -N, --replace-non-ACGTN option
+ +dosage
* fix some serious bugs in dosage calculation
+ +fill-tags
* extended to perform simple on-the-fly calculations such as calculating INFO/DP from FORMAT/DP.
+ merge
* add support for merging FORMAT strings
* bug fixed in gVCF merging
+ mpileup
* a new optional SCR annotation for the number of soft-clipped reads
+ reheader
* new -f, --fai option for updating contig lines in the VCF header
+ +trio-stats
* extend output to include DNM homs and recurrent DNMs
+ VariantKey support
OBS-URL: https://build.opensuse.org/request/show/770397
OBS-URL: https://build.opensuse.org/package/show/science/bcftools?expand=0&rev=7
- Update to 1.9
* `annotate`
- REF and ALT columns can be now transferred from the annotation
file.
- fixed bug when setting vector_end values.
* `consensus`
- new -M option to control output at missing genotypes
- variants immediately following insersions should not be skipped.
Note however, that the current fix requires normalized VCF and may
still falsely skip variants adjacent to multiallelic indels.
- bug fixed in -H selection handling
* `convert`
- the --tsv2vcf option now makes the missing genotypes diploid,
"./." instead of "."
- the behavior of -i/-e with --gvcf2vcf changed. Previously only
sites with FILTER set to "PASS" or "." were expanded and the -i/-e
options dropped sites completely. The new behavior is to let the -i/-e
options control which records will be expanded. In order to drop
records completely, one can stream through "bcftools view" first.
* `csq`
- since the real consequence of start/splice events are not known,
the aminoacid positions at subsequent variants should stay unchanged
- add `--force` option to skip malformatted transcripts in GFFs
with out-of-phase CDS exons.
* `+dosage`: output all alleles and all their dosages at multiallelic
sites
* `+fixref`: fix serious bug in -m top conversion
* `-i/-e` filtering expressions:
- add two-tailed binomial test
- add functions N_PASS() and F_PASS()
- add support for lists of samples in filtering expressions, with
many samples it was impractical to list them all on the command line.
Samples can be now in a file as, e.g., GT[@samples.txt]="het"
- allow multiple perl functions in the expressions and some bug
fixes
- fix a parsing problem, '@' was not removed from '@filename'
expressions
* `mpileup`: fixed bug where, if samples were renamed using the `-G`
(`--read-groups`) option, some samples could be omitted from the
output file.
* `norm`: update INFO/END when normalizing indels
* `+split`: new -S option to subset samples and to use custom file
names instead of the defaults
* `+smpl-stats`: new plugin
* `+trio-stats`: new plugin
* Fixed build problems with non-functional configure script produced
on some platforms
OBS-URL: https://build.opensuse.org/request/show/633582
OBS-URL: https://build.opensuse.org/package/show/science/bcftools?expand=0&rev=6
