v1.1.6

New release to fix Zenodo bug.

1.1.4

New features

• Added two new mini alignment options - identity and similarity plots
  Identity plots show which positions in your alignment are identical to the consensus, different to the consensus or gaps. Created using the --plot_consensus_identity parameter.
  

Similarity plots calculate a substitution score for your alignment vs the consensus, using a substitution matrix of your choice, and colour the plot according to this score. Created using the --plot_consensus_similarity parameter.

• Added additional statistics plots, plotted by default when --plot_stats_input or --plot_stats_output is specified.

  • The nucleotide or amino acid frequency in the alignment and, for nucleotide alignments, the A/T (U) vs C/G ratio
    
  • For nucleotide alignments, the proportion of each "change" in nucleotide relative to the consensus.
    

• Added the option to remove sequences with duplicate names from the alignment, keeping either the first or last instance of each sequence name, with the parameter --deduplicate_ids.

Testing, Packaging and Documentation

• Updated CI workflow to cover more recent versions of Python
• Changed all requirements in package to require a minimum rather than fixed version of software
• Various updates to tests using images to work with latest version of pytest
• Minor updates to readthedocs build

Debugging

• Minor debug to crop_divergent to run remove_gap_only by default afterwards
• Improved markup for remove_gap_only
• Adjustment to move Q amino acids slightly higher in sequence logos
• Removed error sometimes raised if no cleaning functions are run
• Updated utilityFunctions.listFonts() as the matplotlib method used here was deprecated

1.1.0

A detailed description of all changes associated with the release is in the attached PDF.
CIAlign_v1.1.0_Improvements.pdf

Briefly:
The “remove insertions” function now runs much faster, particularly when the maximum insertion size is very large
A new “crop divergent” function is now available, which allows the user to isolate the most conserved region of the alignment.
CIAlign can now generate position weight, position probability and position frequency matrices.
Users can now specify, using the retain functions, sequences which they would like CIAlign not to remove.
The plot_coverage function is now the plot_stats function and will generate three plots, showing coverage, Shannon entropy and information content per column.
Users can now extract part of an alignment based on their own coordinates, using the get_section function.
An additional colour palette is now available for mini alignments and sequence logos,
It is now possible to convert T to U in alignments, as well as U to T.
The CIAlign manual is now hosted at ReadTheDocs - https://cialign.readthedocs.io/ - and is more readable and comprehensive.
Unit test coverage is now 100% across all module files (excluding import lines and logging)

What's Changed

• Retain sequences by @KatyBrown in #45
• Crop divergent by @KatyBrown in #46
• Multiple changes for release of version 1.1.0 by @KatyBrown in #47

Full Changelog: v1.0.17...v1.1.0

1.0.17

Testing for conda release

1.0.16

Changelog:
Added an option to set a threshold proportion of sequences below which an insertion is removed
Added guidance for adding new CIAlign parameters
Pep8 compliance

1.0.15

Changlog:
Changed colour palette to be colour bilndness safe
Updated test images to reflect the new palette
Minor updates to online tables and figures - added linear regression results
Added continuous integration testing

v1.0.14

Changlog:

• Minor updates to setup.py, MANIFEST.in to allow pip installation including .txt files
• Minor updates to timing data

v1.0.11

Changelog:

• Changed how the main CIAlign.py script works internally (but not the results) by moving the argument parser to argP.py and the code to run CIAlign from CIAlign.py to runCIAlign.py and split it up into functions, added extra comments and docstrings
• Added an option to the miniAlignments to keep the numbering from the initial alignment in the mini alignment for the output (e.g. if the input is 1, 2, 3, 4 and 3 is removed the rows in the output will be numbered 1, 2, 4 rather than 1, 2, 3.
• Added default, minimum and maximum values to all the numeric parameters
• Added two additional amino acids - U and O - to the mini alignment colour dictionary
• Added an option to parse non-CIAlign log files to the benchmarking AlignmentStats find_removed_cialign function
• Added --clean, --interpret, --visualise options to run multiple functions
• Added --logo_start, --logo_end to make a sequence logo for partial sequences
• Additional scripts to show how benchmarking was performed
• Generated online figures and online tables for manuscript
• Minor fixes to unit tests to accomodate changes in the main code
• Minor bug fixes and additional documentation

v1.0.10

Changelog
Added unit tests
Added benchmarking scripts
Changed default parameters for remove_insertions and remove_divergent
Minor bug fixes

v1.0.9

Changlog

Minor edit to pip3 installation from v1.0.8