forked from al2na/methylKit
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathNEWS
466 lines (314 loc) · 17.4 KB
/
NEWS
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
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
methylKit 1.3.1
--------------
IMPROVEMENTS AND BUG FIXES
* fix conversion rate error in methCall()
methylKit 1.1.8
--------------
IMPROVEMENTS AND BUG FIXES
* fix methSeg error when only one segment is returned from fastseg,
add case handling for methSeg2bed
* check for user interruption in methCall to enable stop in execution
* changes to select() function: check for out-of-bound indices to
prevent downstream errors
methylKit 1.1.7
--------------
IMPROVEMENTS AND BUG FIXES
* fix methCall segementation fault, added tests and test files from bismark
* fixed missing p.value at coercion of methylBase to GRanges
* changes to the pool() function: save.db=TRUE for methylBaseDB by default,
differing lengths of given sample.ids and unique treatment lead to error,
added tests
* fix osx related error when reading gzipped files with methRead
* change deprecated function names in test files
* fixed bug with dataSim() function updated the manual
methylKit 1.1.6
--------------
IMPROVEMENTS AND BUG FIXES
* fixed a bug where tileMethylCounts() function did not work with
small chromosomes/scaffolds with few bases covered.
methylKit 1.1.5
--------------
IMPROVEMENTS AND BUG FIXES
* fixes missing error messages during methRead : https://github.com/al2na/methylKit/pull/57
methylKit 1.1.4
--------------
IMPROVEMENTS AND BUG FIXES
* merging tabix files is fixed: https://github.com/al2na/methylKit/pull/56
methylKit 1.1.3
--------------
IMPROVEMENTS AND BUG FIXES
* Typos in the vignette are fixed, thanks to Marcin Kosinski
methylKit 1.1.2
--------------
IMPROVEMENTS AND BUG FIXES
* Bug fixes in SAM file reading process. If there were more than one
header line there were problems in reading. Now this is fixed.
https://github.com/al2na/methylKit/pull/51
methylKit 1.1.1
--------------
IMPROVEMENTS AND BUG FIXES
* Fisher's exact test now works as described in the manual. It is automatically
applied in calculateDiffMeth() when there are only two groups with one
replicate each.
* During logistic regression modeling, the samples without counts are removed
from the model but the same filtering is not applied for covariates
data.frame, which can cause errors if min.per.group argument is used.
Now this is fixed: https://github.com/al2na/methylKit/issues/50
methylKit 0.99.4
--------------
IMPROVEMENTS AND BUG FIXES
* all annotation functions are now removed and are available
through genomation package. See the vignette for examples. Most
functions have similar names and functionality in genomation.
The users need to convert methylKit objects to GRanges using
as(methylKit.obj,"GRanges") before they can use genomation
functions.
methylKit 0.99.3
--------------
IMPROVEMENTS AND BUG FIXES
* Fixed a bug in processBismark() C++ function where BAM files are should be treated as 0-based.
Now this is fixed.
* Bug in calculateDiffMethDSS is fixed (https://github.com/al2na/methylKit/issues/49)
methylKit 0.99.2
--------------
IMPROVEMENTS AND BUG FIXES
* Fixed a bug in methRead() introduced after the addition of mincov argument.
the bug occurred only reading files legacy text files that have CpGs
with coverage below 10
* Changes to vignette for better description of the tests.
* Compiler error that occurs in older compilers are fixed via this PR
https://github.com/al2na/methylKit/pull/43
methylKit 0.99.1
--------------
IMPROVEMENTS AND BUG FIXES
* mostly changes to meet BioCcheck() requirements and reccomendations
* C++ code compiles on windows now, regex requirement is no longer there.
methylKit 0.9.6
--------------
IMPROVEMENTS AND BUG FIXES
* changes to following function names:
read() to methRead()
read.bismark to processBismarkAln()
adjust.methylC() to adjustMethylC()
get.methylDiff() to getMethylDiff()
annotate.WithFeature() to annotateWithFeature()
annotate.WithFeature.Flank() to annotateWithFeatureFlank()
annotate.WithGenicParts() to annotateWithGenicParts()
read.bed() to readBed()
read.feature.flank() to readFeatureFlank()
read.transcript.features() to readTranscriptFeatures()
* Improved documentation for methRead() (old read())
* Now, bismark cytosine report and coverage files can be read using methRead()
pipeline argument. see ?methRead
* Ported the Perl script for methylation base calling to C/C++ via Rcpp.
Contributed by Alexander Gosdschan.
* methRead() uses data.table::fread() to read files faster.
* methRead() has a new argument mincov, which sets the minimum number of
reads that needs to cover a base. Positions with coverage below this
number are discarded.
NEW FUNCTIONS AND FEATURES
* new function methSeg() can segment methylation (methylRaw objects) and
differential methylation (methylDiff objects) profiles to segments.
Associated function methSeg2bed() creates BED files from segments.
see ?methSeg. A test is added to check this in R CMD check.
Contributed by Arsene Wabo and Alexander Gosdschan.
* new tabix based classes methylRawDB, methylRawListDB, methylBaseDB,
methylDiffDB and respective methods implemented. Tests are updated to
check proper function in R CMD check. Contributed by Alexander Gosdschan.
* calculateDiffMeth() now supports basic overdispersion correction and multiple
methods for pvalue correction. The function also now handles covariates
such as age,sex etc. A test is added to check this in R CMD check.
Contributed by Adrian Bierling.
* New function calculateDiffMethDSS() is using beta-binomial model from DSS
package to calculate differential methylation. Contributed by
Dhruva Chandramohan with modifications from Altuna Akalin. This is a modified
version of the function from DSS package so that it can work with methylKit
objects.
* dataSim creates a methylBase object with simulated methylation data.
A test is added to check this in R CMD check. Contributed by Adrian Bierling.
methylKit 0.9.5
--------------
IMPROVEMENTS AND BUG FIXES
* travis CI build shield added
methylKit 0.9.4
--------------
IMPROVEMENTS AND BUG FIXES
* tileMethylCounts now works on methylBase objects, affected by BioC 3.0 upgrade.
A test is added to check this in R CMD check.
methylKit 0.9.3
--------------
IMPROVEMENTS AND BUG FIXES
* compatibility with BioC 3.0. Multiple BioC functions moved to other packages
within BioC,which broke some code and caused installation issues.
Now this is fixed.
* data.table::merge is now stable when all=TRUE. Removed my(altuna) version of data.table::merge
the code base.
methylKit 0.9.2.5
--------------
IMPROVEMENTS AND BUG FIXES
* calculateDiffMeth slim=FALSE argument works correctly when there are multiple
samples per group.
methylKit 0.9.2.4
--------------
IMPROVEMENTS AND BUG FIXES
* install_github() now works correctly. Removed blank lines at the end of
DESCRIPTION file.
* calculateDiffMeth and tileMethylCounts now works correctly,
typos were introduced in code with 0.9.2.2. Now these are fixed
methylKit 0.9.2.3
--------------
IMPROVEMENTS AND BUG FIXES
* regionCounts() bug occurring when the first argument is a class of methylBase is fixed.
The bug introduced an additional column to the resulting methylBase object.
https://groups.google.com/forum/#!topic/methylkit_discussion/p19K-pgavAI
methylKit 0.9.2.2
--------------
IMPROVEMENTS AND BUG FIXES
* unite() issues when destrand=FALSE is resolved. The issue appeared due to numeric vs.
integer conflict when merging data sets using chr,start and end locations.
* typo in calculateDiffMeth() argument "weighted.mean" is fixed.
* unite() is faster when destrand=TRUE, due to improvements in internal function
.CpG.dinuc.unify()
* regionCounts() strand.aware argument now works correctly. This bug had no effect
on tileMethylCounts. With the default arguments, strand aware setting was not
in effect at all, so every region was treated as strandless.
* Conversion to GRanges objects now removes seqlevels(chromosome names)
that are not used. Not removing these could cause problems in regionCounts()
function and functions depending on regionCounts().
* read() function in some cases was treating strand columns of the flat CpG
files as logical if strand had F or R letters and most of the first CpGs were
having F strand (Forward strand). Now this is fixed.
methylKit 0.9.2.1
--------------
IMPROVEMENTS AND BUG FIXES
* Fixes a bug in getMethylationStats() function that generates incorrect label numbers when the user overrides the default number
of breaks for the histogram. Contributed by Bonnie Barrilleaux.
methylKit 0.9.2
--------------
IMPROVEMENTS AND BUG FIXES
* A bug introduced with 0.9.1 is fixed. The bug occured when unite() function
is used with destrand=TRUE argument. It returned less number of CpGs then it was
supposed to, although returned CpGs had correct methylation and coverage values.
methylKit 0.9.1
--------------
NEW FUNCTIONS AND FEATURES
* objects takes less memory and they are reorganized. updateMethObject() updates
the objects from previous versions the latest version.
* New batch effect control functions are implemented. You can control if certain principal
components are associated with batch effects and remove those components from your
data. see ?assocComp amd ?removeComp. In addition, if you have corrected for the
batch effects via other methods, you can reconstruct a corrected methylBase object (
see ?reconstruct). Check the "batch effects" section in the vignette.
IMPROVEMENTS AND BUG FIXES
* unite() function takes less time due to use of data.table::merge
* fixes a bug appeared with R 3.02, where getting exon and intron coordinates
from BED12 files produced an error.
* tested with R 3.2 and matching bioconductor packages
* data(methylKit) has the new objects
methylKit 0.5.7
--------------
IMPROVEMENTS AND BUG FIXES
* tested with R 3.0 and matching bioconductor packages
* deprecated matchMatrix import from IRanges was causing a problem with package installation. v0.5.7 fixes that problem.
* Now there is no need for "chr" string in BED files when reading them as annotation files. Some assemblies do not have the "chr"" string in their chromosome names.
methylKit 0.5.6
--------------
NEW FUNCTIONS AND FEATURES
* new arguments for clusterSamples() and PCASamples() functions are added. With
the new options "sd.filter","sd.threshold","filterByQuantile" are added. These
options help finetune how low variation bases/regions are discarded prior to
clustering or PCA. See ?PCASamples and ?clusterSamples() for details on the
new options
* FAQ section added to the vignette
* show methods added for each class. Now, typing the variable name containing the
object will display concise information about the contents of the object.
* Subsetting objects via "[" notation is now enabled. You can subset rows of the
objects and it will return a new object rather than just a data frame.
IMPROVEMENTS AND BUG FIXES
* tileMethylCounts() error is fixed. Error occurred when tilling sparsely
covered small chromosomes like chrM in human RRBS data.
* read.bismark() can deal with Bismark output with Bowtie2. Bowtie2 can put gaps
in the alignment, now read.bismark() can deal with those gaps when parsing the
SAM file.
* Coverage columns are coerced to integer when reading generic methylation per
base files. BSMAP scripts can produce a methylation ratio file where coverages
(or "effective CT counts" as they are called) are not always integers, which
causes a problem in the downstream analysis. Now, these non-integer columns
are rounded to nearest integer while reading.
See http://zvfak.blogspot.com/2012/10/how-to-read-bsmap-methylation-ratio.html
for example usage of this functionality.
methylKit 0.5.5
--------------
IMPROVEMENTS AND BUG FIXES
* Differential methylation percentage calculation bug fixed. The bug occurred when "min.per.group" argument used in unite() function and when "weighted.mean=TRUE" in calculateDiffMeth() function.
* plotTargetAnnotation bug is fixed. Bug occured when "precedence" argument set to FALSE.
methylKit 0.5.4
--------------
IMPROVEMENTS AND BUG FIXES
* Examples added to help pages
* Changes to DESCRIPTION for complying with the bioconductor guidelines
* unused "cor" option removed from PCASamples() function
* some irrelevant functions are not exported (i.e they are not public) anymore.
* getContext() looks for the correct slot name now
methylKit 0.5.3
--------------
NEW FUNCTIONS AND FEATURES
* new function adjust.methylC() can be used to adjust measured 5mC levels by measured 5hmC levels
IMPROVEMENTS AND BUG FIXES
* pool() function bug fixed where one of the groups have one sample the pool function was not returning correct values
* calculateMethDiff() function option "SLIM" is now working. If set to TRUE SLIM method for q-value calculation will be used. If FALSE, p.adjust with method="BH" option will be used for P-value correction.
* read.bismark() function now works correctly under Windows.
* read.bismark() bug occurring when the reads are paired and overlapping is fixed now.
* read.bismark() bug occurring when the alignment is done in non-directional manner is fixed now. However, illumina sequencing protocols are directional and you are unlikely to have encountered this error if you aligned your sequences in a directional manner.
* getCorrelation() function has a new option called "method" can take the value of "spearman","person" or "kendall"
methylKit 0.5.2
--------------
NEW FUNCTIONS AND FEATURES
* new function pool() sums up coverage, numCs and numTs values within each group so one representative sample for each group will be created in a new methylBase object.
* new function normalizeCoverage() normalizes coverage and associated number of Cs and number of Ts values between samples using a scaling factor derived from the ratio between mean or median of coverage distributions of samples.
methylKit 0.5.1
--------------
IMPROVEMENTS AND BUG FIXES
* calculateDiffMeth() can now deal with differential methylation calculations where one group has multiple samples but the other one has only one.
methylKit 0.5
--------------
NEW FUNCTIONS AND FEATURES
* new function reorganize() can be used to reorganize methylRawList and methylBase objects to create new objects from subset of samples or can be used to change the order of samples and treatment vector.
* new function bedgraph() can output UCSC bedgraph files for methylRaw and methylDiff objects
* new function percMethylation() extracts percent methylation values from a methylBase object and returns a matrix
* unite() now can merge bases/regions that are not covered by all samples by setting "min.per.group" option
* PCASamples() has new options "transpose" and "sd.threshold". Now, one can do PCA analysis on the transposed % methylation values. "sd.threshold" is for removing rows with small variation prior to PCA analysis .see ?PCASamples for details
IMPROVEMENTS AND BUG FIXES
* regionCounts() bug that appeared with data.table 1.8.0 is fixed
* unite() bug that appeared with R 2.15 is fixed
* calculateDiffMeth() can deal with NA values introduced by setting "min.per.replicate" option of unite() function
* PCASamples() now uses "prcomp" function to do the PCA analysis
* external annotation data (cpgi.hg18.bed.txt and refseq.hg18.bed.txt) that comes with the package is now only a subset of full datasets. Do not use them for your own analysis, download the complete annotation data from UCSC or other sources in BED format.
methylKit 0.4.1
--------------
NEW FUNCTIONS AND FEATURES
* new function select() can be used to subset methylRaw, methylBase and methylDiff objects to create new objects with a subset of methylation data useful if you want to use only a particular portion of methylation events on the genome.
IMPROVEMENTS AND BUG FIXES
* read.bismark bug fixed, now it should be able to save methylation call files with no problem
methylKit 0.4
--------------
NEW FUNCTIONS AND FEATURES
* New read.bismark() function can read directly from a sorted SAM file output by Bismark aligner, the function can save methylation calls as text files or read them as methylRaw object to be used in analysis
* calculateDiffMeth() can do all differential methylation calculations using multiple-cores if 'num.cores' is set to an integer denoting how many cores should be used.
* getCoverageStats() & getMethylationStats() have a new option called 'labels', if set to FALSE, no labels will be drawn on top of the histogram bars.
IMPROVEMENTS AND BUG FIXES
* cov.bases option in tileMethylCounts() now works
* methylRaw,methylBase and methylDiff objects have a new slot 'resolution', which designates
whether methylation information is base-pair resolution or regional resolution. allowed values 'base' or 'region'
* getCoverageStats() & getMethylationStats() now print the sample ids and methylation context automatically in the
title when plotting
* getCoverageStats() & getMethylationStats() takes extra options to be passed to hist() function
* destrand option of unite() function will be over-ridden when methylRawList to be united contains regions rather than bases
* unite() function checks if supplied elements of methylRawList object have the same context,assembly and resolution before
uniting data sets.
* validity function checking the format of the data for a methylRaw object is implemented
* clusterSamples() and PCASamples() take methylation context information automatically and use it in plot titles
methylKit 0.3.1
--------------
IMPROVEMENTS AND BUG FIXES
* syntax error when fisher.test applied at calculateMethDiff() is fixed.