Skip to content

Releases: milaboratory/mixcr

MiXCR v1.7.4

20 Jun 05:37
Compare
Choose a tag to compare
  • Better compatibility with highly-paralelized execution of MiXCR on clusters: "more random" generation of random temp file names
  • Human-readable messages for two wrong input cases
  • NPE fixed in case of align with VGene, assemble VTranscript
  • Minor fix for documentation

MiXCR v1.7.3

17 May 16:39
Compare
Choose a tag to compare
Major changes
  • Fixed: wrong boundaries of CDR1-CDR3 in Rat IGH/IGK references imported from IMGT
  • Fixed: bug leading to incorrect parsing of input/output file paths with spaces
  • Automatic adjustment of featureToAlign in assemble action; fixes NPE in some use-cases of RNA data analysis with IMGT reference
  • Added support for FreeBSD in mixcr script
Minor changes
  • Human-readable error messages for illegal analysis options with IMGT reference
  • Added human-readable error message for unknown species

MiXCR v1.7.2

05 May 09:28
Compare
Choose a tag to compare
  • Fixed: bug leading to mis-filtering of certain combinations of Alpha-Delta-V genes with Alpha-J or Delta-J genes

MiXCR v1.7.1

09 Mar 16:58
Compare
Choose a tag to compare
  • Added two new export fields: -aaFeatureFromLeft and -aaFeatureFromRight
  • Fixed: bug with incorrect translation of FR4 gene feature in export
  • Fixed: wrong boundaries of FR3 and CDR3 in Mouse TRA/TRD references imported from IMGT
  • Fixed: duplicate-records exception while importing mouse IGLV from IMGT (exception suppressed)

MiXCR v1.7

29 Dec 09:58
Compare
Choose a tag to compare
Major changes
  • New feature (#8 and #36) that allows to extract actual reads that were used in each clone (for more details see documentation)
  • Import of custom gene segment sequences (mixcr importSegments ...)
  • Automated import of IMGT refenrence segment sequences (mixcr importFromIMGT ...)
  • Beta feature: new aligner with support of big gaps (see "New aligner" section in documentation).
  • Added new action exportReads which export initial reads aggregated by each clones to separate .fastq files
  • Added new export options for exportAlignments: -cloneId (prints clone ID that aggregated corresponding alignment) and -mapping (same, with additional info on mapping type)
  • Added new export option for exportClones: -mapping will pring all alignment IDs that were aggreafated by clone
  • Added option -n (print only first n rows) to all export commands
  • Added new assembler option: minimal clonal sequence length (-OminimalClonalSequenceLength=...)
Minor changes
  • minor improvements and bug fixes
  • version information embedded into jar file (try mixcr -v)

Update 13.01.2016: micro fix for importFromIMGT.sh script (see 10f7c87)

MiXCR v1.6

06 Oct 12:41
Compare
Choose a tag to compare
Major changes
  • New feature: optional short column names in export... action to simplify further data analysis using data table processing libraries like Pandas or R/DataFrames. (-s / --no-spaces in exportAlignments and exportClones)
  • New export fields: -defaultAnchorPoints outputs positions of default anchor point in aligned reads or clonal sequence (this column is added to default output format), -positionOf outputs position of specified anchor point, -lengthOf outputs lengt of specified gene feature
  • MiXCR execution script now passes java-specific options directly to JVM (e.g. you can execute mixcr -Xmx2g align ...)
  • Added V5UTRBeginTrimmed anchor point, V5UTR gene feature renamed to V5UTRGermline, trimmed V5UTR gene feature added
  • Added --filter-out-of-frames and --filter-stops options in exportClones
Minor changes
  • some column names in output tab-delimited files slightly changed
  • NPE in exportAlignmentsPretty fixed
  • New anchor poins added to exportAlignmentsPretty output

Update 10.12.2015: Script inside zip archive updated to support nested symlinks on Mac OS X (see 5a149cc)

MiXCR v1.5

10 Sep 23:17
Compare
Choose a tag to compare
Major changes
  • Analysis mode for RNA-Seq data. Use -p rna-seq in align action.
  • Possible set of D genes is limited by loci of V and J genes in current rearrangement
  • Default output format for exportAlignmentsPretty changed for the new one which is more compact and vivid
  • Two new filtering options in exportAlignmentsPretty: --cdr3-contains and --read-contains
  • Default value for -OjParameters.parameters.floatingRightBound changed to false
  • New parameters: -OincludeDScore and -OincludeCScore for align action. If set to true score of D or C gene will be added to total alignment score
  • New action: mergeAlignments to merge several *.vdjca files into one
Minor changes
  • Additional micro filtration step for paired-end read aligner (work in very rare cases)
  • Fixed bug leading to incorrect behaviour if readsLayout=Unknown
  • Correct progress reporting with -n option in align action.
  • Added export options -vHitsWithoutScores (same for J and D genes)
  • Added support for optional compression of .vdjca and .clns files (when specifying additional .gz extension: e.g. mixcr align inut.fastq output.vdjca.gz etc.)
  • Correct handling of quotes in parameters overriding

MiXCR v1.4

21 Aug 16:53
Compare
Choose a tag to compare
  • migration to MiLib 1.1
  • new parameter: vjAlignmentOrder (possible values VThenJ and JThenV). Usage mixcr align -OvjAlignmentOrder=JThenV ... . Previous behaviour of MiXCR was equivalent to JThenV, while now default value for the parameter is VThenJ. Parameter affects only Single-Read alignments and alignment of overlapped paired-end reads. Non-overlaping paired-end reads are always porcessed in VThenJ mode. JThenV can be used for short reads (~100bp) with full (or nearly full) J gene coverage.

MiXCR v1.3

13 Aug 08:19
Compare
Choose a tag to compare
  • new feature: alignments with V and J genes from different loci are dropped (this behaviour can be turned off by adding '--diff-loci' command line option)
  • default parameter value changed: for cloneClusteringParameters.specificMutationProbability value changed from 1E-4 to 1E-3
  • default parameter value changed: for cloneClusteringParameters.searchParameters value changed from oneMismatch to twoMismatchesOrIndels
  • bug-fix: overlapping of reads now preserves read description
  • bug-fix: -descrR2 now outputs information from R2 (was from R1)
  • minor bug-fixed: NPE in VDJCAlignerAbstract

⚠️ default assembler parameters changed: mixcr assemble ... now performs stronger error correction. To avoid any possible biases while comparing data sets, make sure that they are all analysed using the same MiXCR version.

MiXCR v1.2

16 Jun 15:30
Compare
Choose a tag to compare
  • new feature: added ability to export read descriptions with individual alignments result (option --save-description for align command and options -descrR1 and -descrR2 for exportAlignments command)
  • fixed crash caused by -OcloneClusteringParameters=null
  • bug with clonatype ordering in exportClones fixed (occurred with -OcloneClusteringParameters=null)
    note
  • removed -OmaxBadMismatchesAllowed on assemble step

⚠️ intermediate files format changed: *.vdjca files from previous versions of MiXCR are not supported