forked from milaboratory/mixcr
-
Notifications
You must be signed in to change notification settings - Fork 0
/
CHANGELOG
288 lines (229 loc) · 14 KB
/
CHANGELOG
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
MiXCR 2.0.3 (26 Dec 2016)
========================
-- User-friendly export of mutations in germline sequences: nucleotide, amino-acid and detailed
(provides means to link nucleotide mutations with resulting amino-acid mutations; e.g. it allows to
easy differentiate between sense and nonsense nucleotide mutations). Mutations can be exported for
any gene feature (e.g. FR3) realtive to any gene feature (e.g. VRegion).
-- Fix for NPE in assemble action invoked with `-OmaxBadPointsPercent=0`
-- Support for Linux `free` version 3.3.10 in `mixcr` wrapper script (now MiXCR runs on Centos 7.2,
Ubuntu 16.04.1)
-- Unified way for gene feature translation (frame is inferred using closest reference point known
to be in-frame)
-- minor: fix for wrong filtering of stop codons in case of featuresToAlign with non-coding regions
-- minor: Backward compatibility option --diff-loci in align action
MiXCR 2.0.2 (10 Nov 2016)
========================
-- Fixes NPE in `exportClones` action
MiXCR 2.0.1 ( 8 Nov 2016)
========================
-- Fixes problem with absent library search folders
-- Added action `exportClonesPretty`, that prints well-formatted alignments for segments and clonal
sequence
-- Human readable report for assemblePartial action
-- Additional micro-filtering step for alignment hits to exclude wrong chain hits
-- RepSeq.IO: fixes bug when MIXCR fails to resolve partial library name for files without .gz
extension
-- Print warning when using java version <= 7
-- minor: Statistics about chains usage is added to `align` and `assemble` reports
-- minor: Added chains (`--chains` or `-c` option) filter for `exportAlignments`
-- minor: Execution time is now printed in reports
MiXCR 2.0 (13 Sep 2016)
========================
-- New JSON-based reference library format (see [RepSeq.IO](https://github.com/repseqio/repseqio))
-- Complete review of V/D/J/C gene library (see [repository](https://github.com/repseqio/library))
-- New simplified method to import IMGT library (see documentation)
-- All `--loci` options replaced with `--chain` (`-l` -> `-c`)
-- Removed option `--diff-loci` at `align` step
-- Added `-OallowChimeras=true` / `false` option at `align` step (better algorithm than was with
`--diff-loci`)
-- Removed: option `-u`/`--functional-only` in `align` action
-- Many small fixes
-- minor: Improved report content with absolute values for all rows and additional version info
-- minor: Now report with run statistics is additionally printed to stdout
-- minor: Execution time information added to report
MiXCR 1.8.3 ( 8 Sep 2016)
========================
-- Fix for C alignment now will be performed in R2 only for alignments with V gene in R1 (affects
only `align` with `-OallowPartialAlignments=true`)
-- Fix for wrong hits filtration in case of `-OallowPartialAlignments=true`
-- Alignment-to-clone index rewritten from scratch (see `-i` / `--index` option in `assemble`)
-- MapDB dependency removed
-- Fixes occasional NullPointerException during reading of alignemnt-clone index
-- Fixes NullPointerException with -descrR2 in exportAlignments
-- Better wording for command options descriptions in built-in help (printed by `mixcr -h`)
-- minor: Additional fixes for reproducibility of run-to-run results (internal rounding of `float`
hit scores, additional reseeding of random seed in mapping).
MiXCR 1.8.2 (31 Jul 2016)
========================
-- Fixed `floatingRightBound=false` for rna-seq alignment parameters
-- Added new action `filterAlignments`
-- Fixed wrong output for *gene family* export columns
-- Several fixes for alignment report
-- Default value for `qualityMergingAlgorithm` for paired-end merger changed to `MaxSubtraction`
-- Fixed `NPE` / `IllegalArgumentException` in case of complex assemblingFeatures
MiXCR 1.8.1 (29 Jun 2016)
========================
-- Revert quality filtering/mapping algorithm to MiXCR 1.7.x
-- Added different quality aggregation algorithms for assemble (Average, Min, Max, MiniMax) option
`-OqualityAggregationType=Max`
-- Added `clonesDiff` action to calculate descriptive statistics of differnece between two samples
-- Fixed wrong anchor point positions in Macaca mulatta IGL reference imported from IMGT
-- MiXCR returns exit code 1 if program terminated with error
-- Automatic correction of -OvParameters.geneFeatureToAlign in align in some cases (e.g. IMGT
reference and rna-seq paramenters)
MiXCR 1.8 (21 Jun 2016)
========================
-- Assembler of partial reads allows to assemble CDR3s from several short reads: final puzzle piece
for high quality repertoire data extraction from RNA-Seq data
-- Added support for P-segments; new gene features `VPSegment`, `DLeftPSegment`, `DRightPSegment`,
`JPSegment` now available both in `exportAlignments` and `exportClones`; feature is enabled by
default
-- Added new anchor points to encode information about P-segment positions
-- Clone assembler now can separates clones with different V/J/C genes but same clonal sequence
(e.g. CDR3); this behaviour is turned off by default (see `-OseparateByV/J/C` in `assemble` action
parameters to turn it on)
-- Quality aggregation algorithm in assembler changed (see description of assemble procedure)
-- Added clone filtration step: filter clones with low quality clonal sequence (see
-OminimalQuality parameter)
-- Added new action `alignmentsDiff`: run as `mixcr alignmentsDiff file1.vdjca file2.vdjca
report.txt` to find difference between two .vdjca files
-- Following export fields added: `-vFamily`, `-vFamilies` (same for D/J/C)
-- Added new options in `align`: `--not-aligned-R1` and `--not-aligned-R2` for exporting not
aligned reads at alignment step (useful for troubleshooting of raw data glitches / analysis
parameters)
-- Alignment seed is now fixed in order to have completely equal results from run to run on the
same dataset
-- Added option `-e`/`--cdr3-equals` to `exportAlignmentsPretty` action
-- Added option `-v`/`--write-all` to align action
-- Action `exportReads` renamed to `exportReadsForClones`
-- Added `exportReads` action to re-export raw reads from `*.vdjca` file; useful to filter TCR/IG
containing reads
-- Added new `-l`/`--filter-locus` option in `exportClones` action to limit the list of exported
clones by specified locus
-- Added new `-q`/`--minimal-clone-fraction` and `-c`/`--minimal-clone-count` options in
`exportClones` action to filter the list of exported clones by their fraction of reads number
-- minor: Refactoring of report in `assemble` action: better descriptions, fixes and new
informative lines
-- minor: Human-readable error messages for illegal analysis options with IMGT reference
-- minor: Added new export field: `-vBestIdentityPercent` and `-vIdentityPercents` (-j..., -c... in
the same way)
-- minor: Now IMGT import script produces more useful information
-- minor: Added human-readable error message for unknown species
-- minor: Better layout of anchor point marks in `exportAlignmentsPretty`
-- minor: Optional output of read description lines (if present in vdjca file) in
`exportAlignmentsPretty`
MiXCR 1.7.4 (20 Jun 2016)
========================
-- 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 1.7.3 (17 May 2016)
========================
-- 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
-- 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: Human-readable error messages for illegal analysis options with IMGT reference
-- minor: Added human-readable error message for unknown species
MiXCR 1.7.2 ( 5 May 2016)
========================
-- Fixed bug leading to mis-filtering of certain combinations of Alpha-Delta-V genes with Alpha-J
or Delta-J genes
MiXCR 1.7.1 ( 9 Mar 2016)
========================
-- Added two new export fields: `-aaFeatureFromLeft` and `-aaFeatureFromRight`
-- Fixed: bug with incorrect translation fo 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 1.7 (29 Dec 2015)
========================
-- 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: minor improvements and bug fixes
-- minor: version information embedded into jar file (try `mixcr -v`)
MiXCR 1.6 ( 6 Oct 2015)
========================
-- 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: some column names in output tab-delimited files slightly changed
-- minor: NPE in exportAlignmentsPretty fixed
-- minor: New anchor poins added to exportAlignmentsPretty output
MiXCR 1.5 (11 Sep 2015)
========================
-- 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: Additional micro filtration step for paired-end read aligner (work in very rear cases)
-- Minor: Fixed bug leading to incorrect behaviour if `readsLayout=Unknown`
-- Minor: Correct progress reporting with `-n` option in align action.
-- Minor: Added export options -vHitsWithoutScores (same for J and D genes)
-- Minor: 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.)
-- Minor: Correct handling of quotes in parameters overriding
MiXCR 1.4 (21 Aug 2015)
========================
-- 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 1.3 (13 Aug 2015)
========================
-- 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
MiXCR v1.2 (16 Jun 2015)
========================
-- new feature: added ability to export keep read descrptions in .vdjca file (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')
MiXCR v1.1 (29 Apr 2015)
========================
-- new feature: `exportAlignmentsPretty` action for exporting BLAST-style alignments
-- adjust D-gene aligner parameters
-- various performance improvements
-- bash script support for OS X and Linux
-- migration to MiLib 1.0
MiXCR v1.0 (1 Jan 2015)
=======================
-- first release