-
Notifications
You must be signed in to change notification settings - Fork 3
/
ChangeLog
79 lines (64 loc) · 3.55 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
v1.04
* Added --addSpecIds flag which adds a column to the output files with the scan id/title (#26)
* Fixed parsing of scan number for mgf files (#22)
v1.03.1
* Update file indices when reusing dat files
v1.03
* Reuse converted raw to internal binary format files
v1.02.1
* Fix file_list.txt reading to not split on spaces in file paths
v1.02
* Update to newest ProteoWizard version (using boost 1.76 instead of 1.67)
* Add check for empty vector of spectra considered for clustering (#21)
* Split overlap clustering in batches
* Better fix for nativeId issue (#19)
v1.01.1
* Fixing several build issues
* Added OpenMP in osx build
* Undo nativeId fix which broke mgf and mzML output for consensus spectrum file (#19)
v1.01
* Fixing several build issues
* Fixed nativeId issue with new ProteoWizard version for MS2 format
* Consensus spectra for MS1 profile data (#9)
* Added tarball release as package option (#8, #18)
* Clarified names of temporary spectrum files
* Fixed bug with creation of output file in current dir.
v1.00.1
* Fixes infinite loop issue in counting cluster sizes (#17)
v1.00
* Added GUI packages using the Qt framework
* Changed output format to include cluster identifier and removed columns for peptide identifications
* Added support for producing consensus spectra directly using the 'maracluster batch' command
* Fixed issues for large precursor window clustering
* Improved logging and input arguments
* Updated ProteoWizard installation to use last successful build on TeamCity
v0.05
* Fixed several minor issues with race conditions resulting in differing consensus outputs
* Fixed problem with high memory usage in poisoned p-value clustering
v0.04.1
* Fixes consensus spectra merger with multiple bins
v0.04
* Fixed mzML issues with consensus spectra not retaining experiment configurations (#12)
* Added 'spectrum=123' as third option to detect scan numbers (#12)
v0.03.1
* Added macOS build.
* Added SCANS field to mgf consensus spectra output.
v0.03
* Fixed error with mgf consensus output (#5)
* Resolved most mzML validation issues, except a few that msconvert output also produces (#6)
* Fixed problem parsing scan numbers in mgf files (#7)
* Added CentOS, Fedora, Windows 32 and 64-bit builds
v0.02
* Fixed missing attribute 'defaultDataProcessingRef' in mzML output (#4).
* Switched to distributed clustering scheme and removed temporary files to reduce disk space usage. For a data set with 10 million spectra this reduced the disk space usage from 250GB to just 5GB.
* Added windows 32-bit build, with optional Vendor reader support. Vendor reader support is not yet available in the windows binary, but can be compiled from source.
* Added flag `-S`/`-splitMassChargeStates` for consensus spectra generation. This splits multiple candidate precursors in spectrum output file into separate spectrum copies with the same peak list, as some formats (e.g. mgf) and software packages (e.g. MS-GF+) do not support multiple charge states for a single peak list.
* Added flag `-M`/`-minClusterSize` to specify the minimum cluster size to create consensus spectra for.
v0.01
* Fixed segmentation fault during clustering on sample data.
* Fixed duplicate p-value writing during filter-sort procedure.
* Switched to undirected p-value edges to save disk space.
* Use spectrum indices if scan numbers are not available or cannot be parsed from the title (mzML, mgf).
* Fixed bug in CPack.txt in RedHat related REGEX command (jgriss).
* Added guard to detect insufficient disk space while writing binary files.
* Added support for precursor window in `Da`, in addition to the default `ppm`.