Change log

Major changes to PhyKIT are summarized here.

2.1.6: Expanded plotting support for alignment/tree QC workflows:

• Added optional plotting to: pairwise_identity, saturation, covarying_evolutionary_rates, compositional_bias_per_site, evolutionary_rate_per_site, and alignment_entropy

• Added tip_to_tip_distance --all-pairs mode with optional clustered heatmap output (--plot)

• Standardized plotting arguments and JSON metadata: --plot, --plot-output, and plot_output in JSON payloads

• Updated CLI help text and online usage documentation for all newly plotted commands

• Expanded integration test coverage for plotting and JSON+plot behavior and reran full regression and documentation build checks

• Removed cython from runtime dependencies; repository has no Cython build pipeline (no .pyx/cythonize usage), so this was unnecessary

• Sunset Python 3.9 and 3.10 support; CI, packaging classifiers, and python_requires now target Python 3.11+

2.1.5: JSON output expansion and harmonization:

• Added --json support to the remaining CLI commands, including create_concatenation_matrix and nearest_neighbor_interchange

• Standardized JSON metadata key naming for improved consistency across commands

• Added canonical rows payloads for list-style JSON outputs while preserving legacy keys for backward compatibility

• Expanded integration test coverage for JSON payloads and performed full regression verification

2.1.4: New alignment utilities, masking support, and composition/RCV correctness updates:

• Added new alignment commands: - alignment_entropy (site entropy reporting) - occupancy_per_taxon (per-taxon valid-site occupancy) - composition_per_taxon (per-taxon symbol composition, excluding invalid symbols) - mask_alignment (column masking by gap fraction, occupancy, and optional entropy)

• Updated saturation slope calculation to use NumPy no-intercept least-squares (fit constrained through the origin), replacing sklearn in this code path

• Updated rcv and rcvt handling to be case-insensitive and to exclude gaps/ambiguous symbols from counts and normalization

• Clarified valid-length normalization in RCV calculations and related docs/help text

• Documentation maintenance updates to improve rendering consistency and remove duplicate changelog maintenance burden

2.1.0: Major performance improvements, expanded Python support, and bug fixes:

• Compatibility:

  • Added support for Python 3.12 and 3.13

  • Maintains compatibility with Python 3.9, 3.10, and 3.11

• Performance Optimizations:

  • Added multiprocessing support for computationally intensive functions (up to 8x faster): patristic distances, pairwise identity, polytomy test, LB score, bipartition support stats, sum of pairs score, saturation analysis, hidden paralogy check, and covarying evolutionary rates

  • Implemented tree caching with LRU cache to avoid re-parsing files

  • Added NumPy vectorization for alignment operations (5-10x faster)

  • Optimized file I/O with streaming for large concatenation operations

  • Added pickle-based fast tree copying for NNI operations

• Bug Fixes:

  • Fixed tree caching side effects that caused tree modifications to persist

  • Fixed spurious sequence detection to correctly use only terminal branches

  • Fixed DNA threader array broadcasting issues

  • Standardized error exit codes to 2

  • Fixed test infrastructure issues

  • Updated saturation slope fitting to use NumPy no-intercept least-squares (fit constrained through the origin), replacing sklearn in this code path

  • Updated rcv and rcvt to be case-insensitive and to exclude gaps/ambiguous symbols from composition counts and normalization; RCV now normalizes each taxon by valid (non-excluded) sequence length

2.0.2: Fixed bug in dna threading associated with how gaps were introduced in codons.

2.0.1: Added arguments to exclude sites with gaps in the pairwise identities and saturation functions.

2.0.0: Codebase overhaul to make PhyKIT more mem efficient and faster. For example, using list comprehension when appropriate.

1.21.0: The partition file outputted from the create_concat function has been updated to the following format: - column 1: alignment name - column 2: # of taxa present - column 3: # of taxa missing - column 4: fraction of occupancy - column 5: names of missing taxa (; separated)

1.20.0: Fixed bug for thread_dna function when using a ClipKIT log file. Input protein alignment must be the untrimmed alignment.

1.19.9: Saturation function now also reports the absolute value of 1-saturation. Lower values are indicative of less saturation.

1.19.4: Saturation function forces y-intercept to be zero when calculating slope

1.19.3: Saturation function now uses uncorrected distances instead of pairwise identities

1.19.2: Verbose pairwise identity reporting separates pairwise identities by tabs and not a dash

1.19.0: Added function to test for site-wise compositional biases in an alignment. See function compositional_bias_per_site.

1.18.0: Added function to estimate site-wise evolutionary rate in an alignment. See function evolutionary_rate_per_site.

1.15.0: Added function to recode alignments based on 8 different recoding schemes (7 for amino acids; 1 for nucleotides). See function recode.

1.14.0: Added an optional argument to the thread_dna function. Now, PhyKIT can thread nucleotide sequences onto a trimmed amino acid alignment. To do so, point PhyKIT to the ClipKIT outputted log file using the -c argument. The ClipKIT log file can be generated when trimming an alignment with ClipKIT by adding the -l argument (see here for more details: https://jlsteenwyk.com/ClipKIT/).

1.12.6: relative composition variability is now adapted for calculating compositional biases in individual taxa. The new function in rcvt (relative composition variability, taxon).

1.12.4: calculations of pairwise identity in alignment now supports excluding pairwise combinations with gaps.

1.12.3: hidden paralogy check now simply looks for monophyly or lack thereof for a set of taxa. Hidden paralogy check still reports insufficient taxon representation.

1.12.2: removed root.txt file from DVMC function. User's are now recommended to trim outgroup taxa beforehand

1.11.3: Added an optional argument to the prune_tree function wherein instead of pruning tips specified in the input file, those tips will be kept.

1.11.1: Modified sum of pairs score to divide the correct number of pairs by the number of pairs in the reference alignment rather than the query alignment alignment

1.11.0: Added terminal_branch_stats (alias: tbs) function to examine terminal branch lengths

1.10.1: Modified column score and sum of pairs score to divide the correct number of columns or pairs by the number of columns or pairs in the query alignment rather than the reference alignment

1.10.0: Added tip_to_tip_node_distance (alias: t2t_node_dist; t2t_nd) function to calculate the phylogenetic distance between two leaves in a phylogeny. Distance is measured in nodes between two leaves

1.9.0: Added monophyly_check (alias: is_monophyletic) function to examine monophyly among a specified set of taxa

1.8.0: Added hidden_paralogy_check (alias: clan_check) function to examine phylogenetic tree for issues of hidden paralogy

1.7.0: Added nearest_neighbor_interchange (alias: nni) function to generate all NNI moves for a binary rooted phylogeny

1.6.0: Added tip_to_tip_distance (alias: t2t_dist; t2t) function to calculate phylogenetic distance between two leaves in a phylogeny

1.5.0: Added root_tree (alias: root; rt) function to root a phylogenetic tree

1.4.0: PhyKIT is now Python version 3.9 and BioPython 1.79 compatible

1.3.0: Added function that estimates the evolutionary rate of a gene using tree-based properties. Function name is 'evolutionary_rate' or 'evo_rate'

1.2.2: added function to get the subtree of the last common ancestor among a set of taxa

1.2.0: added command line interfaces for all functions so that each command can easily be executed. For example, 'phykit aln_len -h' can now be called using 'pk_aln_len -h'

1.1.0: added faidx (alias: get_entry; ge) function to extract fasta entries from a multi-fasta file

1.0.3: added rooting procedure before calculating RF to handle comparing unrooted and rooted trees

1.0.2: function that calculates Robinson Foulds distance (robinson_foulds_distance; rf_distance; rf_dist; rf) now can take trees that differ in topology. PhyKIT will first determine shared tips between the two trees and prune both trees to a common set of tips. Next, PhyKIT will calculate the Robinson Foulds distance.

0.1.3: Added function (column_score; cs) to calculate the quality of an alignment given an input query alignment and a reference alignment to compare it to

0.1.2: Added function (sum_of_pairs_score; sops; sop) to calculate the quality of an alignment given an input query alignment and a reference alignment to compare it to

0.0.9: PhyKIT now handles error stemming from piping output