Mercurial > repos > matthias > stacks2_procrad
diff news.txt @ 0:a4e62d5c5101 draft
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/stacks2 commit 98327d2948ae1ccb5aef5db9ab88605fd74a0de7-dirty
| author | matthias |
|---|---|
| date | Thu, 29 Nov 2018 11:56:45 -0500 |
| parents | |
| children |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/news.txt Thu Nov 29 11:56:45 2018 -0500 @@ -0,0 +1,181 @@ +Stacks 2.2 - Aug 22, 2018 +-------------------------- + Feature: Added the --bestrad flag to process_radtags. When used it will look for reads that need transposed + before they are processed. + Feature: gstacks: New option --max-debrujin-reads to control the construction of the de Bruijn graph; replaces + --min-kmer-freq which is now deprecated + Bugfix: Added run number to output FASTQ headers in process_radtags to make sure read IDs are always unique. + +Stacks 2.1 - June 25, 2018 +-------------------------- + Bugfix: Fixed a performance regression in sstacks. Recent changes in sstacks made it more likely to + invoke the gapped algorithm to match to the catalog. In some cases, matches to the catalog + would be marked as ambiguous alignments and dropped from the next stages of analysis due to + differences in CIGAR strings from the gapped alignment. + Feature: ustacks Changed --high_cov_thres default value from 2.0 to 3.0. + Feature: gstacks: Changed --min-kmer-freq default value from 0.05 to 0.01. + Feature: Added further checks on zlib calls. + +Stacks 2.0 - Apr 23, 2018 +------------------------- + Feature: modified cstacks and ustacks gapped alignment algorithms to always align the two stacks/alleles + with the most k-mers in common, removed the previous minimum k-mer limit. + Feature: modified cstacks so that when a sample locus matches two or more catalog loci, those catalog + loci are combined, or rolled-up, reducing undermerged loci that generate excess homozygote calls. + Feature: modified tsv2bam so that when two loci from the same sample match the same catalog locus, those + loci are combined. + Bugfix: corrected BbvCI restriction enzyme to add the missing negative strand sequence. + Bugfix: corrected the catalog writing routines for unzipped output files to include missing column. + Bugfix: populations: Fixed the filtering of monomorphic loci. + Bugfix: populations: Now preserving sample ordering in all outputs. + Bugfix: Fixed ICPC compilation. + + 2.0b - May 1, 2018 + Feature: gstacks: Removed the assertion that the first basepair of each locus should + be part of a cutsite (now a warning). + Feature: gstacks: The reported effective coverage is now a more realistic weighted mean. + Bugfix: populations: Fixed STRUCTURE output being corrupted for some unordereds population maps. + +Stacks 2.0 Beta 10 - Apr 10, 2018 +--------------------------------- + Feature: Improved gapped alignment for secondary reads in ustacks. + Feature: Improved populations performance. + Feature: Added enzymes Cac8I, MslI. + Feature: Made population maps more tolerant to spurious extra spaces and lines. + Feature: populations: VCF output: changed the format of the catalog locus field and made the column 1-based. + Feature: gstacks: increased haplotyping rates by adding a filtering of spurious SNPs step. + Bugfix: Fixed populations dramatric slow-down on datasets with more than several hundred samples. + Bugfix: Restored the NS/locus distribution in populations's distributions log. + Bugfix: Fixed the populations --radpainter export. + Bugfix: stacks-dist-extract: Fixed OSX compatibility. + Bugfix: Fixed breaking bug in populations --in-vcf mode filtering statistics. + +Stacks 2.0 Beta 9 - Mar 12, 2018 +-------------------------------- + Feature: Cleaned up tags/snps/alleles/matches files. We removed the batch ID from ustacks and cstacks + output, and the deprecated log likelihood fields from ustacks and cstacks. We also removed + the chromosome/bp/strand fields as they are no longer used in these files. + Feature: Renamed gstacks output files that represent the new components of the catalog: + gstacks.fa.gz => catalog.fa.gz; gstacks.vcf.gz => catalog.calls + Feature: Removed read length restrictions from ustacks/cstacks/sstacks core, reads/loci can vary in + length throughout the pipeline. + Feature: Reimplemented PLINK export format for the populations program. + Bugfix: Updated to HTSLib 1.7; changed to a custom build system that will work with the Stacks build + system. + Bugfix: Made gapped alignments mandatory in ustacks, cstacks, and sstacks. Added check for frameshift + at 3' end of the read -- if found, a match is deferred to the gapped aligner. + +Stacks 2.0 Beta 8 - Feb 03, 2018 +-------------------------------- + Feature: populations: Now calculated deviation from Hardy-Weinberg equilibrium at the SNP level + (using an exact test), and at the haplotype level (using Guo+Thompson's MCMC algorithm). + Feature: populations: Added an export type for FineRADStructure. + Feature: populations: Added the GQ/GL fields in the VCF SNPs output. + Feature: gstacks: Made the default behavior regarding paired-end reads more logical (in + reference-based mode --paired has been replaced with --unpaired). + Feature: gstacks: Added details about samples and coverages to the log outputs. + Feautre: Added enzymes NspI, BbvCI, fixed BfuCI. + Bugfix: corrected a major performance bottleneck in populations when smoothing population statistics + across the genome. + Bugfix: populations: The VCF output now preserves the input sample order. + Bugfix: gstacks: Fixed the handling of a rare special case in the PCR duplicates code. + Bugfix: gstacks: Fixed 100% being added to all per-thread timings. + +Stacks 2.0 Beta 7 - Dec 29, 2017 +-------------------------------- + Feature: gstacks: Added an option to remove PCR duplicates based on insert + size (--rm-pcr-duplicates, plus the related --rm-unpaired-reads). + Feature: populations: Added a haplotype Genepop export. + Feature: populations: improved the help; changed the output names for SNP + files to 'populations.snps.EXT'; added option --no_hap_exports. + Feature: gstacks and populations: Clarified the logs; moved distributions + to a separate '.xlog' file and added script stacks-xlog-extract. + Feature: gstacks: Tweaked the help/interface; especially, replaced --spacer + with --suffix (for BAM directory input). + Feature: Added enzymes BfuI and HinP1. + Feature: Added option --inline_null to clone_filter. + Bugfix: gstacks: Fixed a typo preventing the paired reads from being merged. + Bugfix: populations: Fixed a segfault that occurred with some large datasets. + Bugfix: Made VCF outputs more standard compliant. + Bugfix: populations: Repaired --fasta_samples and --fasta_samples_raw. + Bugfix: populations: Fixed population aborting at the end of the run + when an export option was specified multiple times. + Bugfix: gstacks: Adjusted progression report for catalog asymmetry. + Bugfix: Fixed installation of stacks-integrate-alignments on MacOS. + +Stacks 2.0 Beta 6 - Dec 02, 2017 +-------------------------------- + Feature: Implmented the VCF haplotypes output. + Bugfix: Corrected asset failure in populations when exporting data for genepop or structure output. + +Stacks 2.0 Beta 5 - Nov 27, 2017 +-------------------------------- + Feature: Reimplemented structure, phylip, and phylip_var exports. + Bugfix: Tightened up the overlap algorithm to require 80% of overlapping sequence to be + aligned and of the aligned sequence, 80% must be identities. + Bugfix: Fixed segfault in gstacks when compiled with CLANG on OS X. + Bugfix: gstacks: Fixed how misphasings are reported. + +Stacks 2.0 Beta 4 - Nov 07, 2017 +-------------------------------- + Bugfix: Continued improving overlap algorithm to join SE and PE contigs. + Bugfix: Improved build system to handle new timing functions in gstacks. + +Stacks 2.0 Beta 3 - Nov 01, 2017 +-------------------------------- + Feature: Added output to populations describing mean PE contig size and mean number of + genotyped sites per locus, which reflects the current filtering paramters. + Feature: Improved the output of gstacks and populations. + Feature: Added script `stacks-integrate-alignments`. + Bugfix: made further improvements to the single-end/paired-end locus overlapping algorithm. + Bugfix: fixed all depths being null in populations' VCF output. + Bugfix: Numerically tweaked the marukilow model to remove a limit case. + +Stacks 2.0 Beta 2 - Oct 19, 2017 +-------------------------------- + Feature: gstacks: Made it possible to read from multiple BAM files at the same time; modified the + interface accordingly. + Feature: gstacks: Parallelized the reference-based mode. + Feature: gstacks: Added various statistics & improvements to the log output. + Feature: gstacks: Improved how the forward & paired-end reads are merged (in denovo mode; no more trimming). + Feature: populations: Added code to calculate the overlap between RAD loci when a reference is available. + Feature: populations: Added VCF ouput (--vcf). + Feature: Updated the denovo_map.pl and ref_map.pl wrappers, samples must now be specified using --samples and --popmap. + Bugfix: Fixed three memory leaks in populations; improved reference-aligned batch logic. + Bugfix: Improved overlapping code in gstacks to merge more single and paired-end contigs together. + Bugfix: Now compiles on Apple OS X. + Bugfix: Fixed a bug that skewed the fixed-site (no-SNP) likelihood in the marukilow model. + +Stacks 2.0 Beta 1 - Oct 09, 2017 +-------------------------------- + Feature: Paired-end sequencing data can be utilized fully. In particular, when the shearing-based + protocol is used, the software will assemble a local contig from the paired reads across + the population, possibly overlap it with the forward-reads region, then align all reads to the + assembled contig. This new approach also fully supports double-digest protocols. + Feature: Haplotype calling and diploidy-violation dectection now rely on a novel, more powerful algorithm. + Feature: SNP and genotype-calling now uses the diploid models of Maruki and Lynch (2017). + Feature: The rxstacks program has been replaced with the gstacks program, and there is no need to re-run + some of the earlier steps of the pipeline anymore. + Feature: The memory footprint of the populations program has been considerably reduced and can be scaled + for any size data set. + Feature: The reference-based pipeline has been simplified, and now only comprises two steps: gstacks and populations. + Feature: Added --null_inline mode to clone_filter (and process_radtags) for previously unseen type + of oligo combination. + +Stacks 1.48 - Nov 20, 2017 +--------------------------- + Feature: Added HinP1I restriction enzyme. + Feature: Added --null_inline mode to clone_filter (and process_radtags) for previously unseen type + of oligo combination. + +Stacks 1.47 - Sept 06, 2017 +--------------------------- + Feature: Improved populations's fasta output options (especially, + added a option to export locus consensus sequences). + Feature: denovo_map.pl and red_map.pl now stop if a component + of the pipeline fails. + Feature: Improved the output of denovo_map.pl and ref_map.pl. + Bugfix: Added a format check in Fasta/GzFasta to avoid a potential + segfault when working on FastQ files. + Bugfix: Fixed a bug in count_fixed_catalog_snps.py that could cause + overwrites when working with uncompressed files.