Skip to content

Releases: ODiogoSilva/TriFusion

1.0.1

12 Aug 13:16
Compare
Choose a tag to compare

New Features

General

  • Limited taxa buttons in side panel to improve performance in data sets
    with a large number of taxa (> 1000).

Process

  • Added support for non-contiguous partition files in RAXML format.
  • Added support for reverse concatenation of multiple non-contiguous partitions
    without the need to specify the whole alignment in the partitions.

Statistics

  • Added character (nucleotide/residue) proportion for single genes plot.
    This new plot can be viewed in absolute counts or proportion character
    proportion.
  • Added gene conservation plot for single genes.

Improvements

General

  • Taxa information popup from the side panel are now dynamically fetched
    and should be noticeable faster for large data sets.
  • Added version option to all CLI interfaces via the "-v" or "--version"
    options.

Orthology

  • Improved handling and sanity checking of input and output directories
    for orthomcl_pipeline CLI interface.

Statistics

  • Automatic calculation of summary statistics has been disabled for
    data sets with a large number of taxa (> 1000). Calculation of summary
    statistics can still be triggered manually.

Bug Fixes

Orthology

  • Fixed orthomcl_pipeline error when the input proteomes where not
    properly sanitized (i.e., contain whitespace).
  • Fixed issue where ortholog sequences were incorrectly retrieved from database
    under some circunstances.

Statistics

  • Fixed crash on stacked barplots due to deprecated vega colormaps.
  • Fixed crash on sliding window plot for pairwise sequence similarity and
    segregating sites due to matplotlib update.

Dewit!

25 Jul 17:22
Compare
Choose a tag to compare

This is the first release candidate for TriFusion v1.0.0.

In this release a number of significant improvements were made to the performance and efficiency of the Process and Statistics modules. This resulted in up to 30x speed increase through most operations, particularly when dealing with thousands of alignments files (>50k files).

The other major improvement was in how partitions and substitution models are handled in TriFusion, particularly for output formats where they are required (MCMCtree, gphocs, ima2 and snapp). Partitions should now be defined, updated and modified correctly and seamlessly, even without user input.

The full list of changes is detailed below:

New Features

General

  • Progressbar2 is now bundled with TriFusion to prevent import issues [0.5.3].
  • Increased performance when updating app structures after loading alignment
    files into TriFusion [0.5.13].

Improvements

General

  • Added animation for dropdown menus [0.5.17].
  • Increased efficiency of filechoosers when opening directories with
    a great number of files [0.5.20].
  • Improved item search in the main side panel [0.5.21].
  • The order of partitions in the side panel and output files is now sorted
    according to the starting position of the position [0.5.23].
  • Added EPS, PS, JPEG and TIFF options as figure formats [0.5.33].

Orthology

  • Protein and nucleotide sequences exported by the Orthology module now have
    the same header for each taxa. This makes it easier for downstream processing
    of aligning concatenating, etc. [0.5.34].
  • Added a backstage output file that makes the correspondance between the
    unique headers of ortholog groups and their original header in the proteome
    file [0.5.34].
  • Added informative popups when comparing different group files with active
    filters and/or excluded taxa [0.5.34].
  • Added usearch and mcl checks before executing the orthomcl_pipeline [0.5.35].

Process

  • Refactored concatenation procedure increasing performance up to 29x in run
    time [0.5.4].
  • Added automatic recognition of "?" as missing data symbols, in addition to
    "n" and "x" [0.5.5].
  • Added option to invert selection of active Taxon/File buttons in the main
    side panel [0.5.6].
  • Greatly improved performance when importing partition scheme for large
    alignments [0.5.7].
  • Refactored Process backend to greatly improve performance of all operations,
    mainly when dealing with many input alignments. [0.5.10].
  • New output format dialog design that informs which formats are available for
    main operations [0.5.11].
  • Added check and warning when selecting output formats that are not valid
    for the current main operation [0.5.11].
  • Several minor UI improvements [0.5.12].
  • Increased memory efficiency of Process execution [0.5.14].
  • Output files are now correctly generated when non-contiguous partitions
    are defined. These partitions are effectively merged before
    writing the output files [0.5.23].
  • Added error message when there are no output formats after ignoring
    formats non-compliant with conversion/reverse concatenation [0.5.25].
  • Added custom SNAPP output format extension [0.5.25].
  • Added support for multiple sequence types (DNA and protein) for almost all
    operations [0.5.30].
  • Major improvement of the partition and substitution model setup in
    certain output file formats (Nexus and Phylip) [0.5.30].
  • Improved performance when toggling files/taxa on the side panel
    [0.5.39].
  • Improved performance on alignment reading for Unix operating systems
    [0.5.40].

Statistics

  • Improved legend positioning [0.5.1].
  • Homogenized axis labels and title aesthetic style across plots [0.5.1].
  • Color maps are dynamically selected based on the number of categories,
    resulting
    in more distinguishable color palettes [0.5.1].
  • Improved sliding window line plots. These line plots are now smoothed with
    an interpolation, allowing for a much better and fine grained color
    transition [0.5.1].
  • Improved dialogs and warnings for sliding window plots [0.5.27].

Bug fixes

General

  • Fixed total data set information for taxon informative popup [0.5.16].
  • Fixed text flickering for some informative popups [0.5.17].
  • Fixed partition label update in side panel after merging or splitting [0.5.22].
  • Fixed alignment counter for single partitions after splitting [0.5.23].
  • Fixed occasional crash when canceling the loading of input files [0.5.31].
  • Fixed removal of proteome files from selection in the side panel [0.5.33].

Orthology

  • Handled exception when orthology filters result in no data to be
    plotted [0.5.6].
  • Fixed import of some badly formatted group files in Orthology Explore
    section [0.5.24].
  • Fixed issue where the end of the orthology search operation would hold
    indefinitely [0.5.28].
  • Fixed minor issue with ortholog filter settings in explore section when
    proteome files are loaded [0.5.29].
  • Fixed issue where plots with no data would prevent the generation of a full
    report [0.5.34].
  • Fixed issues with orthomcl_pipeline that prevented the run from ending when
    invalid input files were present in the input directory [0.5.35].

Process

  • Fixed 'ntax' parameter of NEXUS format when converting with custom active
    taxa sets [0.5.2].
  • Fixed issue when reverse concatenating directly from an incorrect partitions
    file [0.5.8].
  • Fixed issue when converting after specifying an output file for
    concatenation. [0.5.11]
  • Fixed missing informative popups for inactive files/taxa [0.5.15].
  • Fixed several issues of the partition scheme when using custom file data sets
    [0.5.18].
  • Fixed issue with some Nexus badly formatted input files [0.5.19].
  • Fixed several issues with partition and substitution model setup in Nexus
    and Phylip output formats [0.5.30].
  • Fixed issue with overwrite/skip/cancel when executing process operations
    [0.5.32].
  • Fixed issue with consensus using the first sequence method [0.5.37].
  • Fixed issue preventing reverse concatenation from a single alignment
    when multiple alignments were provided [0.5.38].
  • Fixed issues with partitions when removing alignment files [0.5.38].

Statistics

  • Fixed color scheme for stacked bar plots that lumped together pairs
    of categories [0.5.1].
  • Fixed issue where removing taxa from the active taxa group would not
    take effect on some operations of the Statistics module [0.5.36].

El Concatenero

20 Apr 16:04
Compare
Choose a tag to compare

First official release of TriFusion for Linux, MacOS and Windows

V0.4.11

13 Jul 15:39
Compare
Choose a tag to compare
V0.4.11 Pre-release
Pre-release

This release fixes several important bugs from 0.4.0 before the release of 0.5.0.

Improvements

Process
  • More informative missing data filter dialog.

Bug Fixes

Process
  • Fixed issue with locus size update after filtering missing taxa.
  • Fixed duplicate detection in Process file loading.
  • Fixed bug when adding partitions to current AlignmentList.
  • Fixed exception handling when loading empty alignments.
  • Fixed float calculations when filtering by taxa minimum.
  • Added alignment cleaning for shelved alignments.
  • Fixed issue that prevented correct sequence collapsing.
  • Fixed issue with gap coding.
  • Fixed temporary directory setting for reverse concatenation.
  • Fixed 0 values for variation filters
  • Fixed issue when adding partitions to existing alignments

V0.4.0

06 Jul 16:10
Compare
Choose a tag to compare
V0.4.0 Pre-release
Pre-release
New Features
  • Added automatic calculation of summary statistics for the active data set when selecting the Statistics screen. All calculations are done in the background and can be canceled at any time. Summary statistics can be displayed for the overall data set or as a table with an entry for each gene
Improvements
  • Added Warning class of float checks
  • Duration of float checks can now be changed
  • Replaced icon view filechoosers with list view filechoosers with multiple selection support
  • Added refresh button to all filechoosers
  • Added Cancel option when prompting for files with the same name as output
  • Improved handling and termination of background process
  • Improved protein filter settings text check in Orthology Search
  • Improved ortholog filter settings text check in Orthology Search
  • Added option to remove upper limit in gene copy ortholog filter
  • Added option to provide minimum proportion of taxa for ortholog filter, instead of absolute value
  • Added Ctrl+f keybinding to bring focus to find text input in secondary filechooser
  • Completed Help texts in-App.
  • Added information to orthoMCL pipeline log
  • Added error handling when there is no unique field id on ortholog search
  • Removed PIL dependency for converting plots to grayscale
  • Prevented exit popup spamming.
  • Added setup.py and re-structured TriFusion package.
  • Added entry points for TriFusion GUI, and for command line versions of the three modules.
  • Added more information on Statistics plots
Bug Fixes
  • Corrected issue with table exportation in Statistics screen due to gray scale implementation
  • Fixed issue with single gene plots for non active alignments
  • Corrected behaviour of go to previous directory button in some filechoosers
  • Fixed issues with Exit popup.
  • Corrected bug with low e-values in Ortholog Search
  • Corrected crash when loading a new project while calculating summary statistics
  • Explicit disabling of stats plot type switch when a plot type is always inactive.
  • Suppress import warnings for command line versions
  • Handled secondary process exception when exiting the application
  • Fixed issue that calls main twice

TriFusion V0.3.0

03 Jun 08:52
Compare
Choose a tag to compare
TriFusion V0.3.0 Pre-release
Pre-release
New Features
  • Added Statistics plot with the cumulative distribution of missing genes for consecutive percentage values of missing data
  • Added Statistics plot for the detection of sequence size outliers for genes and species
  • Added Statistics plot for Allele Frequency Spectrum
Improvements
  • Added a general file checking system for main process output
  • Added support for Stockholm input and output formats
  • Added support for GPhoCS output format
  • Added support for automatic file loading via command line arguments when launching the app
  • Added full support for TriSeq.py, the command line version of the Process module
  • Added full support for orthomcl_pipeline.py, the command line version of the Orthology search module
  • Added full support for TriStats.py, the command line version of the Statistics module
  • Added support for wildcards in arguments for TriSeq.py and TriStats.py on windows
  • Added data check decorator when calling Statistics related method that can handle different types of exceptions
  • Added option to automatic convert all exported Statistics graphics to grayscale
  • Added full support for TriOrtho.py, the command line version of the Orthology explore module
Bug Fixes
  • Fixed issue with scrolling in Statistics Data exploration analyses panel
  • Fixed crash when using keybindings to change to Statistics screen repeatedly
  • Fixed path encoding issue when using Drag'N'Drop on Windows

TriFusion V0.2.0

19 May 15:43
Compare
Choose a tag to compare
TriFusion V0.2.0 Pre-release
Pre-release

[Improvements]
Missing data and minimum taxa filters are now disassociated and can be executed separately

[Bug Fixes]
Fixed issues with sequence case and interleave options in nexus output format
Fixed missing imports
Fixed faulty behaviour of shift+click on icon filechooser on Windows
Fixed issue when writing output alignments with less than 90 characters in interleave format
Fixed issue with number of output files when setting secondary operations

TriFusion V0.1.1

03 May 11:41
Compare
Choose a tag to compare
TriFusion V0.1.1 Pre-release
Pre-release

First official developmental release of TriFusion with most features implemented.