scikit-bio changelog

Version 0.7.4-dev

Features

Added optional support for the Numba backend #2483, with Permanova and Mantel currently using it #2488and #2464.
Introduced transition probability matrix computation (#2496).
Added inverse robust center log ratio (rclr_inv) transformation (#2527).
TreeNode.prune and TreeNode.bifurcate now accept an inplace parameter (default True, preserving the previous in-place behavior) and return the resulting tree. This makes them consistent with other whole-tree methods such as shear and root_at_midpoint. Set inplace=False to leave the original tree unchanged and operate on a copy (#2495).

Performance enhancements

When using the Numba backend, Permanova is up to 8x faster #2488.
Improved TreeNode.copy such that it can handle node cross-references correctly: If a node attribute refers to another node in the tree, the copied node attribute will be redirected to the corresponding node in the new tree (#2497).

Bug Fixes

Fixed a subtle floating-point arithmetic issue in pair_align under a linear gap penalty. Previously it could be less tolerant than expected when atol was set smaller than the default (1e-5) and scores involved decimal numbers (#2513).
Fixed a bug in pair_align which raised a TypeError when called with atol=None. This should be equivalent to atol=0 (#2504).
Patched rclr such that it won't raise a zero division warning (#2526).

Version 0.7.3

Features

Implemented a global mechanism to document, dispatch and test functions supporting the Python array API standard. The documentation pages of those functions now display a support grid of five array libraries (NumPy, CuPy, PyTorch, JAX and Dask) by two devices (CPU and GPU). An example is clr. A new developer documentation page (https://scikit.bio/devdoc/array_api.html) was added to introduce this mechanism (#2412 and #2459).
Added mmvec (Microbe-Metabolite Vectors) to skbio.stats.ordination for learning joint embeddings of two feature sets from co-occurrence patterns. Supports L-BFGS and Adam optimizers, cross-validation via Q-squared scores, and prediction of one modality from another (#2360). Interface modernized and computational performance significantly enhanced (#2448).
Added gamma correction support for modeling among-site evolutionary rate heterogeneity in the JC69, F81, K2P, and TN93 models (#2460). Thank @keder for this contribution.
Added robust center log ratio (rclr) transformation for performing CLR-like transformation on only observed (non-zero) values, making it suitable for sparse compositional data (#2386).
Added support to TreeNode.from_taxonomy for parsing taxonomic lineage strings into trees with an optional extract_rank parameter (#2406). Thank @r0hansaxena for this and other contributions to this release.

Performance enhancements

Reduced import skbio time by ~37% by making requests and h5py lazy imports (#2427). Thank @MAUK9086 for contributing this.
Significantly accelerated gme, bme and nni of the skbio.tree module. Of which, bme received the most rigorous optimization. Parallelization is now enabled by default for this function (#2404).

Bug Fixes

Fixed nni's parsing of tree with single-child root node by explicitly raising in this situation (#2446).
Fixed permdisp mutating the input OrdinationResults object by adding a "grouping" column to its samples DataFrame (#2440).
Fixed permdisp emitting pcoa's RuntimeWarning on every call with a distance matrix larger than 10 samples (#2456). Thank @jissen706 for this fix and the one below.
Fixed the GenBank/EMBL/GFF3 feature-location parser to accept single-base locations with a fuzzy boundary (e.g. complement(<23231) as found in the NCBI lambda virus record J02459.1). Previously such locations raised FileFormatError: Could not parse location string. The corresponding serializer was updated symmetrically to emit <N/>N instead of dropping the fuzzy marker (#2473).

Miscellaneous

Replaced SciPy sparse matrix constructors with sparse array constructors to align with SciPy's sparse array APIs. This affects subsample_counts (csr_array) (no public-facing effect) and tree_basis (coo_array) (#2444).
Added a general _reader_kwargs mechanism for documenting class-specific parameters in auto-generated read() docstrings. Applied to TabularMSA to document the required constructor parameter (#2421). Thank @LiudengZhang for this addition.
beta_diversity now emits a UserWarning when unweighted_unifrac or weighted_unifrac are passed as a callable, advising users to pass the metric name as a string for the optimized implementation (#2462). Thank @SAY-5 for this patch.
Improved documentation for pip-based contribution instructions (#2433). Thank @AhmedKhairy22 for this and other contributions.

Version 0.7.2

Features

Added multiple metrics for calculating the evolutionary distances between aligned sequences: generic: pdist, logdet, paralin, jc69, f81, k2p, f84, tn93. They are hosted by the submodule skbio.sequence.distance. Also added align_dists, which consumes a multiple sequence alignment and generates a distance matrix using choice of metrics (#2336).
mantel and permanova can now run directly on condensed form DistanceMatrix objects (#2322, #2335).
Added new phylip_dm format for PHYLIP formatted distance matrices (#2345 and #2352).
Added support for reading relaxed PHYLIP formatted multiple sequence alignments (#2345, #2352, and #2361).
Added new plotting functionality for OrdinationResults objects, including plotting centroids, confidence ellipses, and 2D plots (#2362)

Performance enhancements

Improved TreeNode.from_linkage_matrix, including efficiency, compatibility (it now accepts array-like objects), and documentation (#2356).
Updated the documentation of dirmult_ttest to recommend input data supplied as raw counts rather than pre-normalized proportions, as the latter lose magnitude information and can lead to higher statistical uncertainty (#2358).

Bug Fixes

Fixed an IO issue in skbio.Table.write where tables containing metadata could not be written (#2338).
Fixed an issue in the from_iterable methods of the DistanceMatrix, SymmetricMatrix, and PairwiseMatrix classes where diagonals were being filled with garbage values (#2347).
Fixed a bug in TreeNode.root_at_midpoint which could fail on a tree with no-length branches (#2353).
Fixed an unexpected behavior in several TreeNode methods, where tips with .name is None were considered as taxa and included in the calculation. Nameless tips are unusual but not forbidden by the current data model. After fixation, those tips are excluded from the calculation. Affected methods are subset, subsets, bipart, bipart and cophenet (#2353).
Fixed an integer overflow issue in _nodes_by_counts that can occur on 32-bit systems, such as WASM ([#2369]#2369).

Miscellaneous

scikit-bio can now be cited by its journal publication in Nature Methods.
TreeNode.shear now returns the sheared tree even if inplace=True (#2353).
Added compatibility for pandas 3.0. Most functionality works, but the INSDC:missing scheme for encoding missing values is not yet supported with pandas 3.0. See issue #2375 for tracking (#2371).
Documentation was improved for the default_write_format attributes of supported file types (#2340)
Improved documentation for parallelization techniques across the package (#2343)

Backward-incompatible changes

Removed sokalmichener from the list of supported beta diversity metrics (see get_beta_diversity_metrics), as this metric was removed in SciPy 1.17.0. SciPy's documentation recommended using rogerstanimoto instead (#2367).

Version 0.7.1.post1

Features

Added support for building scikit-bio python wheels natively on Windows ARM64 (#2309).

Bug Fixes

Fixed source distribution packaging for Python 3.14 support.

Version 0.7.1

Features

Added function ancombc, a Python implementation of the ANCOM-BC (analysis of compositions of microbiomes with bias correction) method for differential abundance testing. We thank @FrederickHuangLin for advice (#2293, #2305, #2306, #2316).
Added function struc_zero, a statistical test for structural zeros. This function complements ancombc or can be used alone (#2305).
Enabled inheritance of IO methods (read, write, sniff) on all scikit-bio objects. Subclasses now inherit IO operations, significantly improving custom class creation and extensibility (#2301).
Added new PairwiseMatrix and SymmetricMatrix classes and an updated DistanceMatrix class. Underlying data of SymmetricMatrix and DistanceMatrix classes can now be stored in condensed form, reducing memory footprint by 50% (#2289).

Performance enhancements

Drastically speed up pcoa with the default eigh method (#2285).
Enriched documentation of permanova, explaining the underlying mathematics, the usage, and how to calculate R², a statistic that measures the proportion of data variance explained by the specified variable (#2294, #2294).
Added float32 support for minimum evolution phylogenetic tree building algorithms (gme, bme and nni) (#2291).
Added float32 support for reading labeled square matrix (lsmat) file format (io.format.lsmat) (#2230).

Bug Fixes

Fixed an unexpected behavior in the file I/O of TabularMSA class that fasta and fastq formats populate .metadata['id'] of each sequence whereas clustal, phylip and stockholm formats populate .index of the entire alignments. Now all file formats populate both pieces of information (#2320).
Fixed a bug that pair_align with trim_ends=True on completely misaligned sequences would raise an IndexError instead of returning an empty path (#2284).
Fixed IO format binary_dm implementation (#2282, #2283).
Fixed a bug in reading binary dissimilarity matrix format (io.format.binary_dm), that a float32 data file would be unnecessarily casted into float64 (#2230).
Replace "Correspondance Analysis" with "Correspondence Analysis" in creating OrdinationResults objects for the accuracy of terminology. This term has been "Correspondence Analysis" otherwise in the project.
Fixed a runtime error when computing phylogenetic diversity metrics on 32-bit architectures (#2321).

Miscellaneous

Renamed DissimilarityMatrix as PairwiseMatrix to better reflect its nature. The old name is kept as an alias (#2289).
Replace "Correspondance Analysis" with "Correspondence Analysis" in creating OrdinationResults objects for the accuracy of terminology. This term has been "Correspondence Analysis" otherwise in the project (#2318).
Added environment variable DISABLE_OPENMP, which lets a user optionally disable OpenMP when compiling the Cython code if needed (#2319).
Dropped support for Python 3.9 and added support for Python 3.14 (#2311).
Dropped support for NumPy versions older than 2.0 (#2314).

Version 0.7.0

Features

Added a development roadmap for scikit-bio to the website (#2251).
Added function dirmult_lme, a differential abundance test for longitudinal data through fitting a Dirichlet-multinomial linear mixed effects model (#2080 and #2250).
Added function pair_align, a re-designed pairwise sequence alignment engine that is versatile, efficient, and generalizable (#2226 and #2196). It is meant to replace the old slow Python engine and the SSW wrapper. It supports:
- Global, local and semi-global alignments (with all four ends customizable).
- Nucleotide, protein, and un-grammared sequences, plain strings (ASCII and Unicode), words/tokens, and numbers.
- Match/mismatch scores or substitution matrix.
- Linear and affine gap penalties.
- Integer, decimal and infinite scores.
- Returning one, multiple or all optimal alignment paths.
Added wrapper functions pair_align_prot and pair_align_nucl which are preloaded with scoring schemes consistent with BLASTP and BLASTN, respectively (#2234).
Added function align_score to calculate the score of a pairwise or multiple sequence alignment (#2201, #2192).
Implemented a dispatch system in scikit-bio to handle a variety of table formats. Currently, it handles arrays, Pandas and Polars dataframes, BIOM tables, and AnnData. Scikit-bio functions can now operate on any of these table formats and outputs an object of the same format, or a format designated by the user (#2187, #2203, #2246, #2258, #2260).
Added four augmentation methods: phylomix, compos_cutmix, aitchison_mixup and mixup to enable generation of synthetic samples (#2214, #2190, #2253).
Added support for scikit-bio-binaries, a separate package which currently increases performance of the pcoa and permanova functions within scikit-bio (#2247).
Added pre-built wheels of scikit-bio to PyPI for easier installation across platforms (#2233, #2232, #2228, #2252).
Adopting the Python array API standard in scikit-bio to enable GPU support for select functions. Further expansion of GPU support within scikit-bio is expected (#2239 and #2250).
Added AlignPath.to_aligned and AlignPath.from_aligned to extract aligned regions of the original sequences, and to reconstruct a path from aligned sequences (#2226).
Added parameter starts to AlignPath.from_tabular to specify starting positions in the original sequences (#2226).

Performance enhancements

Enriched the tutorials of modules sequence, alignment, and table (#2263).
Improved the performance of dirmult_ttest (#2250).
Improved the performance of ancom. This is primarily due to exploiting vectorization of the statistical testing function (such as f_oneway). As a consequence, a custom testing function now must accept 2-D arrays as input and return 1-D arrays. Function names available under scipy.stats are not impacted (#2250).
Added attributes ranges and stops to AlignPath. They facilitate locating the aligned part of each sequence as seq[start:stop] (#2226 and #2201).
Improved the performance of SubstitutionMatrix.identity.
Enhanced TabularMSA.from_path_seqs. It now can extract the aligned region from the middle of a sequence. Also added docstring and doctests (#2201).
Enhanced and changed the default behavior of AlignPath.to_bits, which now returns a bit array representing positions instead of segments. This is desired because with the old default behavior, to_bits and from_bits are not consistent with each other (#2201).

Bug Fixes

Fixed a bug that PairAlignPath.from_cigar would ignore the first insertion (I) of a CIGAR string (#2236).
Fixed an inaccurate statement that one can specify gap as np.inf or np.nan in AlignPath.to_indices. These cases are impossible because the output is integer type.
Fixed an inaccurate statement in the documentation of SubstitutionMatrix.is_ascii. This attribute is True when all characters in the alphabet are ASCII codes (0 to 127), not extended ASCII codes (0 to 255) (#2226).
Fixed a bug that a SubstitutionMatrix cannot be copied (#2226).
Fixed a bug in AlignPath.to_indices which would throw an error if the alignment path has only one segment (#2201).
Fixed a bug in the documentation in which the source button would link to decorator code, instead of the relevant function (#2184).

Miscellaneous

In TreeNode.root_at and TreeNode.root_at_midpoint, the default value of branch_attrs was changed to an empty list; that of root_name was changed to None; that of reset was changed to True (#2259).
In TreeNode.unrooted_copy, the default value of branch_attrs was changed to {"length", "support"}. Specifically, "name" was removed from this set, as a node label is often an attribute of the node instead of the branch. The default value of root_name was changed to None (#2259).
In TreeNode.copy, the default value of deep was set to False. Now tree.copy() returns a shallow copy instead of a deep copy (#2259).
In TreeNode.compare_cophenet, the default value of ignore_self was set to True. Therefore the estimated cophenetic distance between trees better correlates with their discrepancy (#2259).
Renamed column "Reject null hypothesis" as "Signif" in ancom and dirmult_ttest's report tables for conciseness (#2250).
Renamed the parameter significance_test as sig_test in ancom for conciseness. The old name is preserved as an alias (#2250).
Set the default data type of SubstitutionMatrix as np.float32 (previous it was float, which is equivalent to np.float64). Made dtype an optional parameter in from_dict and identity methods.
Adjusted the __repr__ of AlignPath and PairAlignPath (#2226 and #2235).
Changed AlignPath.shape's type from a named tuple to a normal tuple (#2235). Let the values be native Python int rather than np.int64 (#2201).
Changed AlignPath.lengths and AlignPath.starts's dtype from int64 to intp, as these attributes are to facilitate indexing (#2226).
Changed Sequence.to_indices's output index array dtype from uint8 to intp, which is the native NumPy indexing type (#2226).
Enriched the documentation of SubstitutionMatrix (#2226).
Let AlignPath.states be uniformly 2-D, even if there are 8 or less sequences in the alignment (#2201).
Updated documentation to include description of how to stream data through stdin with scikit-bio's read function (#2185).
Improved documentation for the DistanceMatrix object (#2204).
Remove autoplotting functionality to enable inplace operations on large in-memory objects, and improve documentation of existing plotting methods (#2216, #2223).
Initiated efforts to add type annotations to scikit-bio's codebase, starting with the stats.distance module (#2219)
Restored functionality to scikit-bio's benchmarking system and introduced a new repository for storing, running, and hosting benchmarks to prevent performance regression (#2245).
Renamed the parameter number_of_dimensions to dimensions for the pcoa and permdisp functions. number_of_dimensions will remain a valid alias of the parameter, such that either option may be used. (#2257).
Enriched the tutorial for skbio.Sequence (#2243).
The tree.nj function can now operate on DistanceMatrix objects containing float32 or float64 values (#2217).
Improved documentation for conversion between scikit-bio sequence alignments and Biopython and Biotite alignments (#2229, #2230).
Rewrote the install page for the website to reflect availability of wheels and to explicitly state scikit-bio's version support windows (#2254).
Renamed the parameter distance_matrix to distmat in pcoa, bioenv, anosim, permanova, and permdisp. distance_matrix will remain a valid alias of the parameter, such that either option may be used. (#2261)

Backward-incompatible changes

Removed TreeNode.unrooted_deepcopy. Use TreeNode.unrooted_copy(deep=True) instead (#2259).
Removed TreeNode.deepcopy. Use TreeNode.copy(deep=True) instead (#2259).
Removed TreeNode.subtree. It was a placehold but never implemented (#2259).
Removed the wrapper for the Striped Smith Waterman (SSW) library (#2240, #2241). Specifically, this removes local_pairwise_align_ssw, StripedSmithWaterman, and AlignmentStructure under skbio.alignment. We recommend using the new skbio.alignment.pair_align function for pairwise sequence alignment, or other packages that provide production-ready alignment algorithms. See #1814 for discussions.
Removed skbio.alignment.make_identity_substitution_matrix. This has been replaced with skbio.sequence.SubstitutionMatrix.identity.

Version 0.6.3

Features

Python 3.13+ is now supported (#2146).
Added Balanced Minimum Evolution (BME) function for phylogenetic reconstruction and balanced option for NNI (#2105 and #2169).
Added functions rf_dists, wrf_dists and path_dists under skbio.tree to calculate multiple pariwise distance metrics among an arbitrary number of trees. They correspond to TreeNode methods compare_rfd, compare_wrfd and compare_cophenet for two trees (#2166).
Added height and depth methods under TreeNode to calculate the height and depth of a given node.
Added TreeNode.compare_wrfd to calculate the weighted Robinson-Foulds distance or its variants between two trees (#2144).
Wrapped UPGMA and WPGMA from SciPy's linkage method (#2094).
Added TreeNode methods: bipart, biparts and compare_biparts to encode and compare bipartitions in a tree (#2144).
Added TreeNode.has_caches to check if a tree has caches (#2103).
Added TreeNode.is_bifurcating to check if a tree is bifurcating (i.e., binary) (#2117).
Added support for Python's pathlib module in the IO system (#2119).
Added TreeNode.path to return a list of nodes representing the path from one node to another (#2131).
Exposed vectorize_counts_and_tree function from the diversity module to allow use for improving ML accuracy in downstream pipelines (#2173)

Performance enhancements

Significantly improved the performance of the neighbor joining (NJ) algorithm (nj) (#2147) and the greedy minimum evolution (GME) algorithm (gme) for phylogenetic reconstruction, and the NNI algorithm for tree rearrangement (#2169).
Significantly improved the performance of TreeNode.cophenet (renamed from tip_tip_distances) for computing a patristic distance matrix among all or selected tips of a tree (#2152).
Supported Robinson-Foulds distance calculation (TreeNode.compare_rfd) based on bipartitions (equivalent to compare_biparts). This is automatically enabled when the input tree is unrooted. Otherwise the calculation is still based on subsets (equivalent to compare_subsets). The user can override this behavior using the rooted parameter (#2144).
Re-wrote the underlying algorithm of TreeNode.compare_subsets because it is equivalent to the Robinson-Foulds distance on rooted trees. Added parameter proportion. Renamed parameter exclude_absent_taxa as shared_only (#2144).
Added parameter include_self to TreeNode.subset. Added parameters within, include_full and include_tips to TreeNode.subsets (#2144).
Improved the performance and customizability of TreeNode.total_length (renamed from descending_branch_length). Added parameters include_stem and include_self.
Improved the performance of TreeNode.lca (#2132).
Improved the performance of TreeNode methods: ancestors, siblings, and neighbors (#2133, #2135).
Improved the performance of tree traversal algorithms (#2093).
Improved the performance of tree copying (#2103).
Further improved the caching mechanism of TreeNode. Specifically: 1. Node attribute caches are only registered at the root node, which improves memory efficiency. 2. Method clear_caches can be customized to clear node attribute and/or lookup caches, or specified attribute caches (#2099). 3. Added parameter uncache to multiple methods that involves tree manipulation. Default is True. When one knows that caches are not present or relevant, one may set this parameter as False to skip cache clearing to significantly improve performance (#2103).
Expanded the functionality of TreeNode.cache_attr. It can now take a custom function to combine children and self attributes. This makes it possible to cache multiple useful clade properties such as node count and total branch length. Also enriched the method's docstring to provide multiple examples of caching clade properties (#2099).
Added parameter inplace to methods shear, root_at, root_at_midpoint and root_by_outgroup of TreeNode to enable manipulating the tree in place (True), which is more efficient that making a manipulated copy of the tree (False, default) (#2103).
TreeNode.extend can accept any iterable type of nodes as input (#2103).
Added parameter strict to TreeNode.shear (#2103).
Added parameter exclude_attrs to TreeNode.unrooted_copy (#2103).
Added support for legacy random generator to get_rng, such that outputs of scikit-bio functions become reproducible with code that starts with np.random.seed or uses RandomState (#2130).
Allowed shuffle and compare_cophenet (renamed from compare_tip_distances) of TreeNode to accept a random seed or random generator to generate the shuffling function, which ensures output reproducibility (#2118).
Replaced accumulate_to_ancestor with depth under TreeNode. The latter has expanded functionality which covers the default behavior of the former.
Added beta diversity metric jensenshannon, which calculates Jensen-Shannon distance. Thank @quliping for suggesting this in #2125.
Added parameter include_self to TreeNode.ancestors to optionally include the initial node in the path (default: False) (#2135).
Added parameter seed to functions pcoa, anosim, permanova, permdisp, randdm, lladser_pe, lladser_ci, isubsample, subsample_power, subsample_paired_power, paired_subsamples and hommola_cospeciation to accept a random seed or random generator to ensure output reproducibility (#2120 and #2129).
Made the IORegistry sniffer only attempt file formats which are logical given a specific object, thus improving reading efficiency.
Allowed the number_of_dimensions parameter in the function pcoa to accept float values between 0 and 1 to capture fractional cumulative variance.

Bug fixes

Fixed a bug in TreeNode.find which returns the input node object even if it's not in the current tree (#2153).
Fixed a bug in TreeNode.get_max_distance which returns tip names instead of tip instances when there are single-child nodes in the tree (#2144).
Fixed an issue in subsets and cophenet (renamed from tip_tip_distances) of TreeNode which leaves remnant attributes at each node after execution (#2144).
Fixed a bug in TreeNode.compare_rfd which raises an error if taxa of the two trees are not subsets of each other (#2144).
Fixed a bug in TreeNode.compare_subsets which includes the full set (not a subset) of shared taxa between two trees if a basal clade of either tree consists of entirely unshared taxa (#2144).
Fixed a bug in TreeNode.lca which returns the parent of input node X instead of X itself if X is ancestral to other input nodes (#2132).
Fixed a bug in TreeNode.find_all which does not look for other nodes with the same name if a TreeNode instance is provided, as in contrast to what the documentation claims (#2099).
Fixed a bug in skbio.io.format.embed which was not correctly updating the idptr sizing. (#2100).
Fixed a bug in TreeNode.unrooted_move which does not respect specified branch attributes (#2103).
Fixed a bug in skbio.diversity.get_beta_diversity_metrics which does not display metrics other than UniFrac (#2126).
Raises an error when beta diversity metric mahalanobis is called but sample number is smaller than or equal to feature number in the data. Thank @quliping for noting this in #2125.
Fixed a bug in io.format.fasta that improperly handled sequences containing spaces. (#2156)

Miscellaneous

Added a parameter warn_neg_eigval to pcoa and permdisp to control when to raise a warning when negative eigenvalues are encountered. The default setting is more relaxed than the previous behavior, therefore warnings will not be raised when the negative eigenvalues are small in magnitude, which is the case in many real-world scenarios #2154.
Refactored dirmult_ttest to use a separate function for fitting data to Dirichlet-multinomial distribution (#2113)
Remodeled documentation. Special methods (previously referred to as built-in methods) and inherited methods of a class no longer have separate stub pages. This significantly reduced the total number of webpages in the documentation (#2110).
Renamed invalidate_caches as clear_caches under TreeNode, because the caches are indeed deleted rather than marked as obsolete. The old name is preserved as an alias (#2099).
Renamed remove_deleted as remove_by_func under TreeNode. The old name is preserved as an alias (#2103).
Renamed descending_branch_length as total_length under TreeNode. The old name is preserved as an alias.
Under TreeNode, renamed get_max_distance as maxdist. Renamed tip_tip_distances as cophenet. Renamed compare_tip_distances as compare_cophenet. The new names are consistent with SciPy's relevant functions and the main body of the literature. The old names are preserved as aliases.

Deprecated functionality

Method TreeNode.subtree is deprecated. It will become a private member in version 0.7.0 (#2103).

Backward-incompatible changes

Dropped support for Python 3.8 as it has reached end-of-life (EOL). scikit-bio may still be installed under Python 3.8 and will likely work, but the development team no longer guarantee that all functionality will work as intended.
Removed skbio.util.SkbioWarning. Now there are no specific warnings to scikit-bio.
Removed skbio.util.EfficiencyWarning. Previously it was only used in the Python implementations of pairwise sequence alignment algorithms. The new code replaced it with PendingDeprecationWarning.
Removed skbio.util.RepresentationWarning. Previously it was only used in TreeNode.tip_tip_distances when a node has no branch length. The new code removed this behavior (#2152).

Version 0.6.2

Features

Added Greedy Minimum Evolution (GME) function for phylogenetic reconstruction (#2087).
Added support for Microsoft Windows operating system. (#2071, #2068, #2067, #2061, #2046, #2040, #2036, #2034, #2032, #2005)
Added alpha diversity metrics: Hill number (hill), Renyi entropy (renyi) and Tsallis entropy (tsallis) (#2074).
Added rename method for OrdinationResults and DissimilarityMatrix classes (#2027, #2085).
Added nni function for phylogenetic tree rearrangement using nearest neighbor interchange (NNI) (#2050).
Added method TreeNode.unrooted_move, which resembles TreeNode.unrooted_copy but rearranges the tree in place, thus avoid making copies of the nodes (#2073).
Added method TreeNode.root_by_outgroup, which reroots a tree according to a given outgroup (#2073).
Added method TreeNode.unroot, which converts a rooted tree into unrooted by trifucating its root (#2073).
Added method TreeNode.insert, which inserts a node into the branch connecting self and its parent (#2073).

Performance enhancements

The time and memory efficiency of TreeNode has been significantly improved by making its caching mechanism lazy (#2082).
Treenode.copy and TreeNode.unrooted_copy can now perform shallow copy of a tree in addition to deep copy.
TreeNode.unrooted_copy can now copy all attributes of the nodes, in addition to name and length (#2073).
Paremter above was added to TreeNode.root_at, such that the user can root the tree within the branch connecting the given node and its parent, thereby creating a rooted tree (#2073).
Parameter branch_attrs was added to the unrooted_copy, root_at, and root_at_midpoint methods of TreeNode, such that the user can customize which node attributes should be considered as branch attributes and treated accordingly during the rerooting operation. The default behavior is preserved but is subject ot change in version 0.7.0 (#2073).
Parameter root_name was added to the unrooted_copy, root_at, and root_at_midpoint methods of TreeNode, such that the user can customize (or omit) the name to be given to the root node. The default behavior is preserved but is subject ot change in version 0.7.0 (#2073).

Bug fixes

Cleared the internal node references after performing midpoint rooting (TreeNode.root_at_midpoint), such that a deep copy of the resulting tree will not result in infinite recursion (#2073).
Fixed the Zenodo link in the README to always point to the most recent version (#2078).

Miscellaneous

Added statsmodels as a dependency of scikit-bio. It replaces some of the from-scratch statistical analyses in scikit-bio, including Welch's t-test (with confidence intervals), Benjamini-Hochberg FDR correction, and Holm-Bonferroni FDR correction (#2049, (#2063)).

Deprecated functionality

Methods deepcopy and unrooted_deepcopy of Treenode are deprecated. Use copy and unrooted_copy instead.

Version 0.6.1

Features

NumPy 2.0 is now supported (#2051). We thank @rgommers 's advice on this (#1964).
Added module skbio.embedding to provide support for storing and manipulating embeddings for biological objects, such as protein embeddings outputted from protein language models (#2008).
Added an efficient sequence alignment path data structure AlignPath and its derivative PairAlignPath to provide a uniform interface for various multiple and pariwise alignment formats (#2011).
Added simpson_d as an alias for dominance (Simpson's dominance index, a.k.a. Simpson's D) (#2024).
Added inv_simpson (inverse Simpson index), which is equivalent to enspie (#2024).
Added parameter exp to shannon to calculate the exponential of Shannon index (i.e., perplexity, or effective number of species) (#2024).
Added parameter finite to Simpson's D (dominance) and derived metrics (simpson, simpson_e and inv_simpson) to correct for finite samples (#2024).
Added support for dictionary and pandas DataFrame as input for TreeNode.from_taxonomy (#2042).

Performance enhancements

subsample_counts now uses an optimized method from biom-format (#2016).
Improved efficiency of counts matrix and vector validation prior to calculating community diversity metrics (#2024).

Miscellaneous

Default logarithm base of Shannon index (shannon) was changed from 2 to e. This is to ensure consistency with other Shannon-based metrics (pielou_e), and with literature and implementations in the field. Meanwhile, parameter base was added to pielou_e such that the user can control this behavior (#2024). See discussions in 1884 and 2014.
Improved treatment of empty communities (i.e., all taxa have zero counts, or there is no taxon) when calculating alpha diversity metrics. Most metrics will return np.nan and do not raise a warning due to zero division. Exceptions are metrics that describe observed counts, includng sobs, singles, doubles and osd, which return zero (#2024). See discussions in #2014.
Return values of pielou_e and heip_e were set to 1.0 for one-taxon communities, such that NaN is avoided, while honoring the definition (evenness of taxon abundance(s)) and the rationale (ratio between observed and maximum) (#2024).
Removed hdmedians as a dependency by porting its geomedian function (geometric median) into scikit-bio (#2003).
Removed 98% warnings issued during the test process (#2045 and #2037).

Version 0.6.0

Performance enhancements

Launched the new scikit-bio website: https://scikit.bio. The previous domain names scikit-bio.org and skbio.org continue to work and redirect to the new website.
Migrated the scikit-bio website repo from the gh-pages branch of the scikit-bio repo to a standalone repo: scikit-bio.github.io.
Replaced the Bootstrap theme with the PyData theme for building documentation using Sphinx. Extended this theme to the website. Customized design elements (#1934).
Improved the calculation of Fisher's alpha diversity index (fisher_alpha). It is now compatible with optimizers in SciPy 1.11+. Edge cases such as all singletons can be handled correctly. Handling of errors and warnings was improved. Documentation was enriched (#1890).
Allowed delimiter=None which represents whitespace of arbitrary length in reading lsmat format matrices (#1912).

Features

Added biom-format Table import and updated corresponding requirement files (#1907).
Added biom-format 2.1.0 IO support (#1984).
Added Table support to alpha_diversity and beta_diversity drivers (#1984).
Implemented a mechanism to automatically build documentation and/or homepage and deploy them to the website (#1934).
Added the Benjamini-Hochberg method as an option for FDR correction (in addition to the existing Holm-Bonferroni method) for ancom and dirmult_ttest (#1988).
Added function dirmult_ttest, which performs differential abundance test using a Dirichilet multinomial distribution. This function mirrors the method provided by ALDEx2 (#1956).
Added method Sequence.to_indices to convert a sequence into a vector of indices of characters in an alphabet (can be from a substitution matrix) or unique characters observed in the sequence. Supports gap masking and wildcard substitution (#1917).
Added class SubstitutionMatrix to support substitution matrices for nucleotides, amino acids are more general cases (#1913).
Added alpha diversity metric sobs, which is the observed species richness (S_{obs}) of a sample. sobs will replace observed_otus, which uses the historical term "OTU". Also added metric observed_features to be compatible with the QIIME 2 terminology. All three metrics are equivalent (#1902).
beta_diversity now supports use of Pandas a DataFrame index, issue #1808.
Added alpha diversity metric phydiv, which is a generalized phylogenetic diversity (PD) framework permitting unrooted or rooted tree, unweighted or weighted by abundance, and an exponent parameter of the weight term (#1893).
Adopted NumPy's new random generator np.random.Generator (see NEP 19) (#1889).
SciPy 1.11+ is now supported (#1887).
Removed IPython as a dependency. Scikit-bio continues to support displaying plots in IPython, but it no longer requires importing IPython functionality (#1901).
Made Matplotlib an optional dependency. Scikit-bio no longer requires Matplotlib except for plotting, during which it attempts to import Matplotlib if it is present in the system, and raises an error if not (#1901).
Ported the QIIME 2 metadata object into skbio. (#1929)
Python 3.12+ is now supported, thank you @actapia (#1930)
Introduced native character conversion ([#1971])(#1971)

Backward-incompatible changes [experimental]

Beta diversity metric kulsinski was removed. This was motivated by that SciPy replaced this distance metric with kulczynski1 in version 1.11 (see SciPy issue #2009), and that both metrics do not return 0 on two identical vectors (#1887).

Bug fixes

Fixed documentation interface of vlr and relevant functions (#1934).
Fixed broken link in documentation of Simpson's evenness index. See issue #1923.
Safely handle Sequence.iter_kmers where k is greater than the sequence length (#1723)
Re-enabled OpenMP support, which has been mistakenly disabled in 0.5.8 (#1874)
permanova and permdist operate on a DistanceMatrix and a grouping object. Element IDs must be synchronized to compare correct sets of pairwise distances. This failed in case the grouping was provided as a pandas.Series, because it was interpreted as an ordered list and indices were ignored (see issue #1877 for an example). Note: pandas.DataFrame was handled correctly. This behavior has been fixed with PR #1879
Fixed slicing for TabularMSALoc on Python 3.12. See issue #1926.

Miscellaneous

Replaced the historical term "OTU" with the more generic term "taxon" (plural: "taxa"). As a consequence, the parameter "otu_ids" in phylogenetic alpha and beta diversity metrics was replaced by "taxa". Meanwhile, the old parameter "otu_ids" is still kept as an alias of "taxa" for backward compatibility. However it will be removed in a future release.
Revised contributor's guidelines.
Renamed function multiplicative_replacement as multi_replace for conciseness (#1988).
Renamed parameter multiple_comparisons_correction as p_adjust of function ancom for conciseness (#1988).
Enabled code coverage reporting via Codecov. See #1954.
Renamed the default branch from "master" to "main". See #1953.
Enabled subclassing of DNA, RNA and Protein classes to allow secondary development.
Dropped support for NumPy < 1.17.0 in order to utilize the new random generator.
Use CYTHON by default during build (#1874)
Implemented augmented assignments proposed in issue #1789
Incorporated Ruff's formatting and linting via pre-commit hooks and GitHub Actions. See PR #1924.
Improved docstrings for functions accross the entire codebase. See #1933 and #1940
Removed API lifecycle decorators in favor of deprecation warnings. See #1916

Version 0.5.9

Features

Adding Variance log ratio estimators in skbio.stats.composition.vlr and skbio.stats.composition.pairwise_vlr (#1803)
Added skbio.stats.composition.tree_basis to construct ILR bases from TreeNode objects. (#1862)
IntervalMetadata.query now defaults to obtaining all results, see #1817.

Backward-incompatible changes [experimental]

With the introduction of the tree_basis object, the ILR bases are now represented in log-odds coordinates rather than in probabilities to minimize issues with numerical stability. Furthermore, the ilr and ilr_inv functions now takes the basis input parameter in terms of log-odds coordinates. This affects the skbio.stats.composition.sbp_basis as well. (#1862)

Important

Complex multiple axis indexing operations with TabularMSA have been removed from testing due to incompatibilities with modern versions of Pandas. (#1851)
Pinning scipy <= 1.10.1 (#1851)

Bug fixes

Fixed a bug that caused build failure on the ARM64 microarchitecture due to floating-point number handling. (#1859)
Never let the Gini index go below 0.0, see #1844.
Fixed bug #1847 in which the edge from the root was inadvertantly included in the calculation for descending_branch_length

Miscellaneous

Replaced dependencies CacheControl and lockfile with requests to avoid a dependency inconsistency issue of the former. (See #1863, merged in #1859)
Updated installation instructions for developers in CONTRIBUTING.md (#1860)

Version 0.5.8

Features

Added NCBI taxonomy database dump format (taxdump) (#1810).
Added TreeNode.from_taxdump for converting taxdump into a tree (#1810).
scikit-learn has been removed as a dependency. This was a fairly heavy-weight dependency that was providing minor functionality to scikit-bio. The critical components have been implemented in scikit-bio directly, and the non-criticial components are listed under "Backward-incompatible changes [experimental]".
Python 3.11 is now supported.

Backward-incompatible changes [experimental]

With the removal of the scikit-learn dependency, three beta diversity metric names can no longer be specified. These are wminkowski, nan_euclidean, and haversine. On testing, wminkowski and haversine did not work through skbio.diversity.beta_diversity (or sklearn.metrics.pairwise_distances). The former was deprecated in favor of calling minkowski with a vector of weights provided as kwarg w (example below), and the latter does not work with data of this shape. nan_euclidean can still be accessed fron scikit-learn directly if needed, if a user installs scikit-learn in their environment (example below).

counts = [[23, 64, 14, 0, 0, 3, 1],
        [0, 3, 35, 42, 0, 12, 1],
        [0, 5, 5, 0, 40, 40, 0],
        [44, 35, 9, 0, 1, 0, 0],
        [0, 2, 8, 0, 35, 45, 1],
        [0, 0, 25, 35, 0, 19, 0],
        [88, 31, 0, 5, 5, 5, 5],
        [44, 39, 0, 0, 0, 0, 0]]

# new mechanism of accessing wminkowski
from skbio.diversity import beta_diversity
beta_diversity("minkowski", counts, w=[1,1,1,1,1,1,2])

# accessing nan_euclidean through scikit-learn directly
import skbio
from sklearn.metrics import pairwise_distances
sklearn_dm = pairwise_distances(counts, metric="nan_euclidean")
skbio_dm = skbio.DistanceMatrix(sklearn_dm)

Deprecated functionality [experimental]

skbio.alignment.local_pairwise_align_ssw has been deprecated (#1814) and will be removed or replaced in scikit-bio 0.6.0.

Bug fixes

Use oldest-supported-numpy as build dependency. This fixes problems with environments that use an older version of numpy than the one used to build scikit-bio (#1813).

Version 0.5.7

Features

Introduce support for Python 3.10 (#1801).
Tentative support for Apple M1 (#1709).
Added support for reading and writing a binary distance matrix object format. (#1716)
Added support for np.float32 with DissimilarityMatrix objects.
Added support for method and number_of_dimensions to permdisp reducing the runtime by 100x at 4000 samples, issue #1769.
OrdinationResults object is now accepted as input for permdisp.

Performance enhancements

Avoid an implicit data copy on construction of DissimilarityMatrix objects.
Avoid validation on copy of DissimilarityMatrix and DistanceMatrix objects, see PR #1747
Use an optimized version of symmetry check in DistanceMatrix, see PR #1747
Avoid performing filtering when ids are identical, see PR #1752
center_distance_matrix has been re-implemented in cython for both speed and memory use. Indirectly speeds up pcoa PR #1749
Use a memory-optimized version of permute in DistanceMatrix, see PR #1756.
Refactor pearson and spearman skbio.stats.distance.mantel implementations to drastically improve memory locality. Also cache intermediate results that are invariant across permutations, see PR #1756.
Refactor permanova to remove intermediate buffers and cythonize the internals, see PR #1768.

Bug fixes

Fix windows and 32bit incompatibility in unweighted_unifrac.

Miscellaneous

Python 3.6 has been removed from our testing matrix.
Specify build dependencies in pyproject.toml. This allows the package to be installed without having to first manually install numpy.
Update hdmedians package to a version which doesn't require an initial manual numpy install.
Now buildable on non-x86 platforms due to use of the SIMD Everywhere library.
Regenerate Cython wrapper by default to avoid incompatibilities with installed CPython.
Update documentation for the skbio.stats.composition.ancom function. (#1741)

Version 0.5.6

Features

Added option to return a capture group compiled regex pattern to any class inheriting GrammaredSequence through the to_regex method. (#1431)
Added Dissimilarity.within and .between to obtain the respective distances and express them as a DataFrame. (#1662)
Added Kendall Tau as possible correlation method in the skbio.stats.distance.mantel function (#1675).
Added support for IUPAC amino acid codes U (selenocysteine), O (pyrrolysine), and J (leucine or isoleucine). (#1576

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]

Changed skbio.tree.TreeNode.support from a method to a property.
Added assign_supports method to skbio.tree.TreeNode to extract branch support values from node labels.
Modified the way a node's label is printed: support:name if both exist, or support or name if either exists.

Performance enhancements

Bug fixes

Require Sphinx <= 3.0. Newer Sphinx versions caused build errors. #1719
- skbio.stats.ordination tests have been relaxed. (#1713)
Fixes build errors for newer versions of NumPy, Pandas, and SciPy.
Corrected a criticial bug in skbio.alignment.StripedSmithWaterman/skbio.alignment.local_pairwise_align_ssw which would cause the formatting of the aligned sequences to misplace gap characters by the number of gap characters present in the opposing aligned sequence up to that point. This was caused by a faulty implementation of CIGAR string parsing, see #1679 for full details.
Fixes build errors for newer versions of NumPy, Pandas, and SciPy.
Corrected a criticial bug in skbio.alignment.StripedSmithWaterman/skbio.alignment.local_pairwise_align_ssw which would cause the formatting of the aligned sequences to misplace gap characters by the number of gap characters present in the opposing aligned sequence up to that point. This was caused by a faulty implementation of CIGAR string parsing, see #1679 for full details.

Deprecated functionality [stable]

Deprecated functionality [experimental]

Miscellaneous

skbio.diversity.beta_diversity now accepts a pandas DataFrame as input.
Avoid pandas 1.0.0 import warning (#1688)
Added support for Python 3.8 and dropped support for Python 3.5.
This version now depends on scipy >= 1.3 and pandas >= 1.0.

Version 0.5.5 (2018-12-10)

Features

skbio.stats.composition now has methods to compute additive log-ratio transformation and inverse additive log-ratio transformation (alr, alr_inv) as well as a method to build a basis from a sequential binary partition (sbp_basis).

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]

Performance enhancements

Bug fixes

Deprecated functionality [stable]

Deprecated functionality [experimental]

Miscellaneous

Python 3.6 and 3.7 compatibility is now supported
A pytest runner is shipped with every installation (#1633)
The nosetest framework has been replaced in favor of pytest (#1624)
The numpy docs are deprecated in favor of Napoleon (#1629)
This version is now compatible with numpy >= 1.17.0 and Pandas >= 0.23. (#1627)

Version 0.5.4 (2018-08-23)

Features

Added FSVD, an alternative fast heuristic method to perform Principal Coordinates Analysis, to skbio.stats.ordination.pcoa.

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]

Performance enhancements

Added optimized utility methods f_matrix_inplace and e_matrix_inplace which perform f_matrix and e_matrix computations in-place and are used by the new center_distance_matrix method in skbio.stats.ordination.

Bug fixes

Deprecated functionality [stable]

Deprecated functionality [experimental]

Miscellaneous

Version 0.5.3 (2018-08-07)

Features

Added unpack and unpack_by_func methods to skbio.tree.TreeNode to unpack one or multiple internal nodes. The unpack operation removes an internal node and regrafts its children to its parent while retaining the overall length. (#1572)
Added support to skbio.tree.TreeNode to return the support value of a node.
Added permdisp to skbio.stats.distance to test for the homogeniety of groups. (#1228).
Added pcoa_biplot to skbio.stats.ordination to project descriptors into a PCoA plot.
Fixed pandas to 0.22.0 due to this: pandas-dev/pandas#20527

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]

Performance enhancements

Bug fixes

Relaxing type checking in diversity calculations. (#1583).

Deprecated functionality [stable]

Deprecated functionality [experimental]

Miscellaneous

Version 0.5.2 (2018-04-18)

Features

Added skbio.io.format.embl for reading and writing EMBL files for DNA, RNA and Sequence classes.
Removing ValueError check in skbio.stats._subsample.subsample_counts when replace=True and n is greater than the number of items in counts. #1527
Added skbio.io.format.gff3 for reading and writing GFF3 files for DNA, Sequence, and IntervalMetadata classes. (#1450)
skbio.metadata.IntervalMetadata constructor has a new keyword argument, copy_from, for creating an IntervalMetadata object from an existing IntervalMetadata object with specified upper_bound.
skbio.metadata.IntervalMetadata constructor allows None as a valid value for upper_bound. An upper_bound of None means that the IntervalMetadata object has no upper bound.
skbio.metadata.IntervalMetadata.drop has a new boolean parameter negate to indicate whether to drop or keep the specified Interval objects.

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]

Performance enhancements

skbio.tree.nj wall-clock runtime was decreased by 99% for a 500x500 distance matrix and 93% for a 100x100 distance matrix. (#1512, #1513)

Bug fixes

The include_self parameter was not being honored in skbio.TreeNode.tips. The scope of this bug was that if TreeNode.tips was called on a tip, it would always result in an empty list when unrolled.
In skbio.stats.ordination.ca, proportion_explained was missing in the returned OrdinationResults object. (#1345)
skbio.diversity.beta_diversity now handles qualitative metrics as expected such that beta_diversity('jaccard', mat) == beta_diversity('jaccard', mat > 0). Please see #1549 for further detail.
skbio.stats.ordination.rda The occasional column mismatch in output biplot_scores is fixed (#1519).

Deprecated functionality [stable]

Deprecated functionality [experimental]

Miscellaneous

scikit-bio now depends on pandas >= 0.19.2, and is compatible with newer pandas versions (e.g. 0.20.3) that were previously incompatible.
scikit-bio now depends on numpy >= 1.17.0, < 1.14.0 for compatibility with Python 3.4, 3.5, and 3.6 and the available numpy conda packages in defaults and conda-forge channels.
added support for running tests from setup.py. Both python setup.py nosetests and python setup.py test are now supported, however python setup.py test will only run a subset of the full test suite. (#1341)

Version 0.5.1 (2016-11-12)

Features

Added IntervalMetadata and Interval classes in skbio.metadata to store, query, and manipulate information of a sub-region of a sequence. (#1414)
Sequence and its child classes (including GrammaredSequence, RNA, DNA, Protein) now accept IntervalMetadata in their constructor API. Some of their relevant methods are also updated accordingly. (#1430)
GenBank parser now reads and writes Sequence or its subclass objects with IntervalMetadata. (#1440)
DissimilarityMatrix now has a new constructor method called from_iterable. (#1343).
DissimilarityMatrix now allows non-hollow matrices. (#1343).
DistanceMatrix.from_iterable now accepts a validate=True parameter. (#1343).
DistanceMatrix now has a new method called to_series to create a pandas.Series from a DistanceMatrix (#1397).
Added parallel beta diversity calculation support via skbio.diversity.block_beta_diversity. The issue and idea is discussed in (#1181, while the actual code changes are in #1352).

Backward-incompatible changes [stable]

The constructor API for Sequence and its child classes (including GrammaredSequence, RNA, DNA, Protein) are changed from (sequence, metadata=None, positional_metadata=None, lowercase=False) to (sequence, metadata=None, positional_metadata=None, interval_metadata=None, lowercase=False)

The changes are made to allow these classes to adopt IntervalMetadata object for interval features on the sequence. The interval_metadata parameter is added imediately after positional_metadata instead of appended to the end, because it is more natural and logical and, more importantly, because it is unlikely in practice to break user code. A user's code would break only if they had supplied metadata, postional_metadata, and lowercase parameters positionally. In the unlikely event that this happens, users will get an error telling them a bool isn't a valid IntervalMetadata type, so it won't silently produce buggy behavior.

Backward-incompatible changes [experimental]

Modifying basis handling in skbio.stats.composition.ilr_inv prior to checking for orthogonality. Now the basis is strictly assumed to be in the Aitchison simplex.
DistanceMatrix.from_iterable default behavior is now to validate matrix by computing all pairwise distances. Pass validate=False to get the previous behavior (no validation, but faster execution).(#1343).
GenBank I/O now parses sequence features into the attribute of interval_metadata instead of positiona_metadata. And the key of FEATURES is removed from metadata attribute.

Performance enhancements

TreeNode.shear was rewritten for approximately a 25% performance increase. (#1399)
The IntervalMetadata allows dramatic decrease in memory usage in reading GenBank files of feature rich sequences. (#1159)

Bug fixes

skbio.tree.TreeNode.prune and implicitly skbio.tree.TreeNode.shear were not handling a situation in which a parent was validly removed during pruning operations as may happen if the resulting subtree does not include the root. Previously, an AttributeError would raise as parent would be None in this situation.
numpy linking was fixed for installation under El Capitan.
A bug was introduced in #1398 into TreeNode.prune and fixed in #1416 in which, under the special case of a single descendent existing from the root, the resulting children parent references were not updated. The cause of the bug was a call made to self.children.extend as opposed to self.extend where the former is a list.extend without knowledge of the tree, while the latter is TreeNode.extend which is able to adjust references to self.parent.

Miscellaneous

Removed deprecated functions from skbio.util: is_casava_v180_or_later, remove_files, and create_dir.
Removed deprecated skbio.Sequence.copy method.

Version 0.5.0 (2016-06-14)

IMPORTANT: scikit-bio is no longer compatible with Python 2. scikit-bio is compatible with Python 3.4 and later.

Features

Added more descriptive error message to skbio.io.registry when attempting to read without specifying into and when there is no generator reader. (#1326)
Added support for reference tags to skbio.io.format.stockholm reader and writer. (#1348)
Expanded error message in skbio.io.format.stockholm reader when constructor is not passed, in order to provide better explanation to user. (#1327)
Added skbio.sequence.distance.kmer_distance for computing the kmer distance between two sequences. (#913)
Added skbio.sequence.Sequence.replace for assigning a character to positions in a Sequence. (#1222)
Added support for pandas.RangeIndex, lowering the memory footprint of default integer index objects. Sequence.positional_metadata and TabularMSA.positional_metadata now use pd.RangeIndex as the positional metadata index. TabularMSA now uses pd.RangeIndex as the default index. Usage of pd.RangeIndex over the previous pd.Int64Index should be transparent, so these changes should be non-breaking to users. scikit-bio now depends on pandas >= 0.18.0 (#1308)
Added reset_index=False parameter to TabularMSA.append and TabularMSA.extend for resetting the MSA's index to the default index after appending/extending.
Added support for partial pairwise calculations via skbio.diversity.partial_beta_diversity. (#1221, #1337). This function is immediately deprecated as its return type will change in the future and should be used with caution in its present form (see the function's documentation for details).
TemporaryFile and NamedTemporaryFile are now supported IO sources for skbio.io and related functionality. (#1291)
Added tree_node_class=TreeNode parameter to skbio.tree.majority_rule to support returning consensus trees of type TreeNode (the default) or a type that has the same interface as TreeNode (e.g. TreeNode subclasses) (#1193)
TreeNode.from_linkage_matrix and TreeNode.from_taxonomy now support constructing TreeNode subclasses. TreeNode.bifurcate now supports TreeNode subclasses (#1193)
The ignore_metadata keyword has been added to TabularMSA.iter_positions to improve performance when metadata is not necessary.
Pairwise aligners in skbio.alignment now propagate per-sequence metadata objects (this does not include positional_metadata).

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]

TabularMSA.append and TabularMSA.extend now require one of minter, index, or reset_index to be provided when incorporating new sequences into an MSA. Previous behavior was to auto-increment the index labels if minter and index weren't provided and the MSA had a default integer index, otherwise error. Use reset_index=True to obtain the previous behavior in a more explicit way.
skbio.stats.composition.ancom now returns two pd.DataFrame objects, where it previously returned one. The first contains the ANCOM test results, as before, and the second contains percentile abundances of each feature in each group. The specific percentiles that are computed and returned is controlled by the new percentiles parameter to skbio.stats.composition.ancom. In the future, this second pd.DataFrame will not be returned by this function, but will be available through the contingency table API. (#1293)
skbio.stats.composition.ancom now performs multiple comparisons correction by default. The previous behavior of not performing multiple comparisons correction can be achieved by passing multiple_comparisons_correction=None.
The reject column in the first pd.DataFrame returned from skbio.stats.composition.ancom has been renamed Reject null hypothesis for clarity. (#1375)

Bug fixes

Fixed row and column names to biplot_scores in the OrdinationResults object from skbio.stats.ordination. This fix affect the cca and rda methods. (#1322)
Fixed bug when using skbio.io.format.stockholm reader on file with multi-line tree with no id. Previously this raised an AttributeError, now it correctly handles this type of tree. (#1334)
Fixed bug when reading Stockholm files with GF or GS features split over multiple lines. Previously, the feature text was simply concatenated because it was assumed to have trailing whitespace. There are examples of Stockholm files with and without trailing whitespace for multi-line features, so the skbio.io.format.stockholm reader now adds a single space when concatenating feature text without trailing whitespace to avoid joining words together. Multi-line trees stored as GF metadata are concatenated as they appear in the file; a space is not added when concatenating. (#1328)
Fixed bug when using Sequence.iter_kmers on empty Sequence object. Previously this raised a ValueError, now it returns an empty generator.
Fixed minor bug where adding sequences to an empty TabularMSA with MSA-wide positional_metadata would result in a TabularMSA object in an inconsistent state. This could happen using TabularMSA.append or TabularMSA.extend. This bug only affects a TabularMSA object without sequences that has MSA-wide positional_metadata (for example, TabularMSA([], positional_metadata={'column': []})).
TreeNode.distance now handles the situation in which self or other are ancestors. Previosly, a node further up the tree was used resulting in inflated distances. (#807)
TreeNode.prune can now handle a root with a single descendent. Previously, the root was ignored from possibly having a single descendent. (#1247)
Providing the format keyword to skbio.io.read when creating a generator with an empty file will now return an empty generator instead of raising StopIteration. (#1313)
OrdinationResults is now importable from skbio and skbio.stats.ordination and correctly linked from the documentation (#1205)
Fixed performance bug in pairwise aligners resulting in 100x worse performance than in 0.2.4.

Deprecated functionality [stable]

Deprecated use of the term "non-degenerate", in favor of "definite". GrammaredSequence.nondegenerate_chars, GrammaredSequence.nondegenerates, and GrammaredSequence.has_nondegenerates have been renamed to GrammaredSequence.definite_chars, GrammaredSequence.definites, and GrammaredSequence.has_definites, respectively. The old names will be removed in scikit-bio 0.5.2. Relevant affected public classes include GrammaredSequence, DNA, RNA, and Protein.

Deprecated functionality [experimental]

Deprecated function skbio.util.create_dir. This function will be removed in scikit-bio 0.5.1. Please use the Python standard library functionality described here. (#833)
Deprecated function skbio.util.remove_files. This function will be removed in scikit-bio 0.5.1. Please use the Python standard library functionality described here. (#833)
Deprecated function skbio.util.is_casava_v180_or_later. This function will be removed in 0.5.1. Functionality moved to FASTQ sniffer. (#833)

Miscellaneous

When installing scikit-bio via pip, numpy must now be installed first (#1296)

Version 0.4.2 (2016-02-17)

Minor maintenance release. This is the last Python 2.7 compatible release. Future scikit-bio releases will only support Python 3.

Features

Added skbio.tree.TreeNode.bifurcate for converting multifurcating trees into bifurcating trees. (#896)
Added skbio.io.format.stockholm for reading Stockholm files into a TabularMSA and writing from a TabularMSA. (#967)
scikit-bio Sequence objects have better compatibility with numpy. For example, calling np.asarray(sequence) now converts the sequence to a numpy array of characters (the same as calling sequence.values).
Added skbio.sequence.distance subpackage for computing distances between scikit-bio Sequence objects (#913)
Added skbio.sequence.GrammaredSequence, which can be inherited from to create grammared sequences with custom alphabets (e.g., for use with TabularMSA) (#1175)
Added skbio.util.classproperty decorator

Backward-incompatible changes [stable]

When sniffing or reading a file (skbio.io.sniff, skbio.io.read, or the object-oriented .read() interface), passing newline as a keyword argument to skbio.io.open now raises a TypeError. This backward-incompatible change to a stable API is necessary because it fixes a bug (more details in bug fix section below).
When reading a FASTQ or QSEQ file and passing variant='solexa', ValueError is now raised instead of NotImplementedError. This backward-incompatible change to a stable API is necessary to avoid creating a spin-locked process due to a bug in Python. See #1256 for details. This change is temporary and will be reverted to NotImplementedError when the bug is fixed in Python.

Backward-incompatible changes [experimental]

skbio.io.format.genbank: When reading GenBank files, the date field of the LOCUS line is no longer parsed into a datetime.datetime object and is left as a string. When writing GenBank files, the locus date metadata is expected to be a string instead of a datetime.datetime object (#1153)
Sequence.distance now converts the input sequence (other) to its type before passing both sequences to metric. Previous behavior was to always convert to Sequence.

Bug fixes

Fixed bug when using Sequence.distance or DistanceMatrix.from_iterable to compute distances between Sequence objects with differing metadata/positional_metadata and passing metric=scipy.spatial.distance.hamming (#1254)
Fixed performance bug when computing Hamming distances between Sequence objects in DistanceMatrix.from_iterable (#1250)
Changed skbio.stats.composition.multiplicative_replacement to raise an error whenever a large value of delta is chosen (#1241)
When sniffing or reading a file (skbio.io.sniff, skbio.io.read, or the object-oriented .read() interface), passing newline as a keyword argument to skbio.io.open now raises a TypeError. The file format's newline character will be used when opening the file. Previous behavior allowed overriding the format's newline character but this could cause issues with readers that assume newline characters are those defined by the file format (which is an entirely reasonable assumption). This bug is very unlikely to have surfaced in practice as the default newline behavior is universal newlines mode.
DNA, RNA, and Protein are no longer inheritable because they assume an IUPAC alphabet.
DistanceMatrix constructor provides more informative error message when data contains NaNs (#1276)

Miscellaneous

Warnings raised by scikit-bio now share a common subclass skbio.util.SkbioWarning.

Version 0.4.1 (2015-12-09)

Features

The TabularMSA object was added to represent and operate on tabular multiple sequence alignments. This satisfies RFC 1. See the TabularMSA docs for full details.
Added phylogenetic diversity metrics, including weighted UniFrac, unweighted UniFrac, and Faith's Phylogenetic Diversity. These are accessible as skbio.diversity.beta.unweighted_unifrac, skbio.diversity.beta.weighted_unifrac, and skbio.diversity.alpha.faith_pd, respectively.
Addition of the function skbio.diversity.alpha_diversity to support applying an alpha diversity metric to multiple samples in one call.
Addition of the functions skbio.diversity.get_alpha_diversity_metrics and skbio.diversity.get_beta_diversity_metrics to support discovery of the alpha and beta diversity metrics implemented in scikit-bio.
Added skbio.stats.composition.ancom function, a test for OTU differential abundance across sample categories. (#1054)
Added skbio.io.format.blast7 for reading BLAST+ output format 7 or BLAST output format 9 files into a pd.DataFrame. (#1110)
Added skbio.DissimilarityMatrix.to_data_frame method for creating a pandas.DataFrame from a DissimilarityMatrix or DistanceMatrix. (#757)
Added support for one-dimensional vector of dissimilarities in skbio.stats.distance.DissimilarityMatrix constructor. (#6240)
Added skbio.io.format.blast6 for reading BLAST+ output format 6 or BLAST output format 8 files into a pd.DataFrame. (#1110)
Added inner, ilr, ilr_inv and clr_inv, skbio.stats.composition, which enables linear transformations on compositions (#892
Added skbio.diversity.alpha.pielou_e function as an evenness metric of alpha diversity. (#1068)
Added to_regex method to skbio.sequence._iupac_sequence ABC - it returns a regex object that matches all non-degenerate versions of the sequence.
Added skbio.util.assert_ordination_results_equal function for comparing OrdinationResults objects in unit tests.
Added skbio.io.format.genbank for reading and writing GenBank/GenPept for DNA, RNA, Protein and Sequence classes.
Added skbio.util.RepresentationWarning for warning about substitutions, assumptions, or particular alterations that were made for the successful completion of a process.
TreeNode.tip_tip_distances now supports nodes without an associated length. In this case, a length of 0.0 is assumed and an skbio.util.RepresentationWarning is raised. Previous behavior was to raise a NoLengthError. (#791)
DistanceMatrix now has a new constructor method called from_iterable.
Sequence now accepts lowercase keyword like DNA and others. Updated fasta, fastq, and qseq readers/writers for Sequence to reflect this.
The lowercase method has been moved up to Sequence meaning all sequence objects now have a lowercase method.
Added reverse_transcribe class method to RNA.
Added Sequence.observed_chars property for obtaining the set of observed characters in a sequence. (#1075)
Added Sequence.frequencies method for computing character frequencies in a sequence. (#1074)
Added experimental class-method Sequence.concat which will produce a new sequence from an iterable of existing sequences. Parameters control how positional metadata is propagated during a concatenation.
TreeNode.to_array now supports replacing nan branch lengths in the resulting branch length vector with the value provided as nan_length_value.
skbio.io.format.phylip now supports sniffing and reading strict, sequential PHYLIP-formatted files into skbio.Alignment objects. (#1006)
Added default_gap_char class property to DNA, RNA, and Protein for representing gap characters in a new sequence.

Backward-incompatible changes [stable]

Sequence.kmer_frequencies now returns a dict. Previous behavior was to return a collections.Counter if relative=False was passed, and a collections.defaultdict if relative=True was passed. In the case of a missing key, the Counter would return 0 and the defaultdict would return 0.0. Because the return type is now always a dict, attempting to access a missing key will raise a KeyError. This change may break backwards-compatibility depending on how the Counter/defaultdict is being used. We hope that in most cases this change will not break backwards-compatibility because both Counter and defaultdict are dict subclasses.

If the previous behavior is desired, convert the dict into a Counter/defaultdict:
```
import collections
from skbio import Sequence
seq = Sequence('ACCGAGTTTAACCGAATA')

# Counter
freqs_dict = seq.kmer_frequencies(k=8)
freqs_counter = collections.Counter(freqs_dict)

# defaultdict
freqs_dict = seq.kmer_frequencies(k=8, relative=True)
freqs_default_dict = collections.defaultdict(float, freqs_dict)
```
Rationale: We believe it is safer to return dict instead of Counter/defaultdict as this may prevent error-prone usage of the return value. Previous behavior allowed accessing missing kmers, returning 0 or 0.0 depending on the relative parameter. This is convenient in many cases but also potentially misleading. For example, consider the following code:
```
from skbio import Sequence
seq = Sequence('ACCGAGTTTAACCGAATA')
freqs = seq.kmer_frequencies(k=8)
freqs['ACCGA']
```
Previous behavior would return 0 because the kmer 'ACCGA' is not present in the Counter. In one respect this is the correct answer because we asked for kmers of length 8; 'ACCGA' is a different length so it is not included in the results. However, we believe it is safer to avoid this implicit behavior in case the user assumes there are no 'ACCGA' kmers in the sequence (which there are!). A KeyError in this case is more explicit and forces the user to consider their query. Returning a dict will also be consistent with Sequence.frequencies.

Backward-incompatible changes [experimental]

Replaced PCoA, CCA, CA and RDA in skbio.stats.ordination with equivalent functions pcoa, cca, ca and rda. These functions now take pd.DataFrame objects.
Change OrdinationResults to have its attributes based on pd.DataFrame and pd.Series objects, instead of pairs of identifiers and values. The changes are as follows:
- species and species_ids have been replaced by a pd.DataFrame named features.
- site and site_ids have been replaced by a pd.DataFrame named samples.
- eigvals is now a pd.Series object.
- proportion_explained is now a pd.Series object.
- biplot is now a pd.DataFrame object named biplot_scores.
- site_constraints is now a pd.DataFrame object named sample_constraints.
short_method_name and long_method_name are now required arguments of the OrdinationResults object.
Removed skbio.diversity.alpha.equitability. Please use skbio.diversity.alpha.pielou_e, which is more accurately named and better documented. Note that equitability by default used logarithm base 2 while pielou_e uses logarithm base e as described in Heip 1974.
skbio.diversity.beta.pw_distances is now called skbio.diversity.beta_diversity. This function no longer defines a default metric, and metric is now the first argument to this function. This function can also now take a pairwise distances function as pairwise_func.
Deprecated function skbio.diversity.beta.pw_distances_from_table has been removed from scikit-bio as scheduled. Code that used this should be adapted to use skbio.diversity.beta_diversity.
TreeNode.index_tree now returns a 2-D numpy array as its second return value (the child node index) instead of a 1-D numpy array.
Deprecated functions skbio.draw.boxplots and skbio.draw.grouped_distributions have been removed from scikit-bio as scheduled. These functions generated plots that were not specific to bioinformatics. These types of plots can be generated with seaborn or another general-purpose plotting package.
Deprecated function skbio.stats.power.bootstrap_power_curve has been removed from scikit-bio as scheduled. Use skbio.stats.power.subsample_power or skbio.stats.power.subsample_paired_power followed by skbio.stats.power.confidence_bound.
Deprecated function skbio.stats.spatial.procrustes has been removed from scikit-bio as scheduled in favor of scipy.spatial.procrustes.
Deprecated class skbio.tree.CompressedTrie and function skbio.tree.fasta_to_pairlist have been removed from scikit-bio as scheduled in favor of existing general-purpose Python trie packages.
Deprecated function skbio.util.flatten has been removed from scikit-bio as scheduled in favor of solutions available in the Python standard library (see here and here for examples).
Pairwise alignment functions in skbio.alignment now return a tuple containing the TabularMSA alignment, alignment score, and start/end positions. The returned TabularMSA's index is always the default integer index; sequence IDs are no longer propagated to the MSA. Additionally, the pairwise alignment functions now accept the following input types to align:
- local_pairwise_align_nucleotide: DNA or RNA
- local_pairwise_align_protein: Protein
- local_pairwise_align: IUPACSequence
- global_pairwise_align_nucleotide: DNA, RNA, or TabularMSA[DNA|RNA]
- global_pairwise_align_protein: Protein or TabularMSA[Protein]
- global_pairwise_align: IUPACSequence or TabularMSA
- local_pairwise_align_ssw: DNA, RNA, or Protein. Additionally, this function now overrides the protein kwarg based on input type. constructor parameter was removed because the function now determines the return type based on input type.
Removed skbio.alignment.SequenceCollection in favor of using a list or other standard library containers to store scikit-bio sequence objects (most SequenceCollection operations were simple list comprehensions). Use DistanceMatrix.from_iterable instead of SequenceCollection.distances (pass key="id" to exactly match original behavior).
Removed skbio.alignment.Alignment in favor of skbio.alignment.TabularMSA.
Removed skbio.alignment.SequenceCollectionError and skbio.alignment.AlignmentError exceptions as their corresponding classes no longer exist.

Bug Fixes

Sequence objects now handle slicing of empty positional metadata correctly. Any metadata that is empty will no longer be propagated by the internal _to constructor. (#1133)
DissimilarityMatrix.plot() no longer leaves a white border around the heatmap it plots (PR #1070).
TreeNode.root_at_midpoint`` no longer fails when a node with two equal length child branches exists in the tree. (#1077)
TreeNode._set_max_distance, as called through TreeNode.get_max_distance or TreeNode.root_at_midpoint would store distance information as lists in the attribute MaxDistTips on each node in the tree, however, these distances were only valid for the node in which the call to _set_max_distance was made. The values contained in MaxDistTips are now correct across the tree following a call to get_max_distance. The scope of impact of this bug is limited to users that were interacting directly with MaxDistTips on descendant nodes; this bug does not impact any known method within scikit-bio. (#1223)
Added missing nose dependency to setup.py's install_requires. (#1214)
Fixed issue that resulted in legends of OrdinationResult plots sometimes being truncated. (#1210)

Deprecated functionality [stable]

skbio.Sequence.copy has been deprecated in favor of copy.copy(seq) and copy.deepcopy(seq).

Miscellaneous

Doctests are now written in Python 3.
make test now validates MANIFEST.in using check-manifest. (#461)
Many new alpha diversity equations added to skbio.diversity.alpha documentation. (#321)
Order of lowercase and validate keywords swapped in DNA, RNA, and Protein.

Version 0.4.0 (2015-07-08)

Initial beta release. In addition to the changes detailed below, the following subpackages have been mostly or entirely rewritten and most of their APIs are substantially different (and improved!):

skbio.sequence
skbio.io

The APIs of these subpackages are now stable, and all others are experimental. See the API stability docs for more details, including what we mean by stable and experimental in this context. We recognize that this is a lot of backward-incompatible changes. To avoid these types of changes being a surprise to our users, our public APIs are now decorated to make it clear to developers when an API can be relied upon (stable) and when it may be subject to change (experimental).

Features

Added skbio.stats.composition for analyzing data made up of proportions
Added new skbio.stats.evolve subpackage for evolutionary statistics. Currently contains a single function, hommola_cospeciation, which implements a permutation-based test of correlation between two distance matrices.
Added support for skbio.io.util.open_file and skbio.io.util.open_files to pull files from HTTP and HTTPS URLs. This behavior propagates to the I/O registry.
FASTA/QUAL (skbio.io.format.fasta) and FASTQ (skbio.io.format.fastq) readers now allow blank or whitespace-only lines at the beginning of the file, between records, or at the end of the file. A blank or whitespace-only line in any other location will continue to raise an error #781.
scikit-bio now ignores leading and trailing whitespace characters on each line while reading FASTA/QUAL and FASTQ files.
Added ratio parameter to skbio.stats.power.subsample_power. This allows the user to calculate power on groups for uneven size (For example, draw twice as many samples from Group B than Group A). If ratio is not set, group sizes will remain equal across all groups.
Power calculations (skbio.stats.power.subsample_power and skbio.stats.power.subsample_paired_power) can use test functions that return multiple p values, like some multivariate linear regression models. Previously, the power calculations required the test to return a single p value.
Added skbio.util.assert_data_frame_almost_equal function for comparing pd.DataFrame objects in unit tests.

Performance enhancements

The speed of quality score decoding has been significantly improved (~2x) when reading fastq files.
The speed of NucleotideSequence.reverse_complement has been improved (~6x).

Bug fixes

Changed Sequence.distance to raise an error any time two sequences are passed of different lengths regardless of the distance_fn being passed. (#514)
Fixed issue with TreeNode.extend where if given the children of another TreeNode object (tree.children), both trees would be left in an incorrect and unpredictable state. (#889)
Changed the way power was calculated in subsample_paired_power to move the subsample selection before the test is performed. This increases the number of Monte Carlo simulations performed during power estimation, and improves the accuracy of the returned estimate. Previous power estimates from subsample_paired_power should be disregarded and re-calculated. (#910)
Fixed issue where randdm was attempting to create asymmetric distance matrices.This was causing an error to be raised by the DistanceMatrix constructor inside of the randdm function, so that randdm would fail when attempting to create large distance matrices. (#943)

Deprecated functionality

Deprecated skbio.util.flatten. This function will be removed in scikit-bio 0.3.1. Please use standard python library functionality described here Making a flat list out of lists of lists, Flattening a shallow list (#833)
Deprecated skbio.stats.power.bootstrap_power_curve will be removed in scikit-bio 0.4.1. It is deprecated in favor of using subsample_power or sample_paired_power to calculate a power matrix, and then the use of confidence_bounds to calculate the average and confidence intervals.

Backward-incompatible changes

Removed the following deprecated functionality:
- skbio.parse subpackage, including SequenceIterator, FastaIterator, FastqIterator, load, parse_fasta, parse_fastq, parse_qual, write_clustal, parse_clustal, and FastqParseError; please use skbio.io instead.
- skbio.format subpackage, including fasta_from_sequence, fasta_from_alignment, and format_fastq_record; please use skbio.io instead.
- skbio.alignment.SequenceCollection.int_map; please use SequenceCollection.update_ids instead.
- skbio.alignment.SequenceCollection methods to_fasta and toFasta; please use SequenceCollection.write instead.
- constructor parameter in skbio.alignment.Alignment.majority_consensus; please convert returned biological sequence object manually as desired (e.g., str(seq)).
- skbio.alignment.Alignment.to_phylip; please use Alignment.write instead.
- skbio.sequence.BiologicalSequence.to_fasta; please use BiologicalSequence.write instead.
- skbio.tree.TreeNode methods from_newick, from_file, and to_newick; please use TreeNode.read and TreeNode.write instead.
- skbio.stats.distance.DissimilarityMatrix methods from_file and to_file; please use DissimilarityMatrix.read and DissimilarityMatrix.write instead.
- skbio.stats.ordination.OrdinationResults methods from_file and to_file; please use OrdinationResults.read and OrdinationResults.write instead.
- skbio.stats.p_value_to_str; there is no replacement.
- skbio.stats.subsample; please use skbio.stats.subsample_counts instead.
- skbio.stats.distance.ANOSIM; please use skbio.stats.distance.anosim instead.
- skbio.stats.distance.PERMANOVA; please use skbio.stats.distance.permanova instead.
- skbio.stats.distance.CategoricalStatsResults; there is no replacement, please use skbio.stats.distance.anosim or skbio.stats.distance.permanova, which will return a pandas.Series object.
skbio.alignment.Alignment.majority_consensus now returns BiologicalSequence('') if the alignment is empty. Previously, '' was returned.
min_observations was removed from skbio.stats.power.subsample_power and skbio.stats.power.subsample_paired_power. The minimum number of samples for subsampling depends on the data set and statistical tests. Having a default parameter to set unnecessary limitations on the technique.

Miscellaneous

Changed testing procedures
- Developers should now use make test
- Users can use python -m skbio.test
- Added skbio.util._testing.TestRunner (available through skbio.util.TestRunner). Used to provide a test method for each module init file. This class represents a unified testing path which wraps all skbio testing functionality.
- Autodetect Python version and disable doctests for Python 3.
numpy is no longer required to be installed before installing scikit-bio!
Upgraded checklist.py to check source files non-conforming to new header style. (#855)
Updated to use natsort >= 4.0.0.
The method of subsampling was changed for skbio.stats.power.subsample_paired_power. Rather than drawing a paired sample for the run and then subsampling for each count, the subsample is now drawn for each sample and each run. In test data, this did not significantly alter the power results.
checklist.py now enforces __future__ imports in .py files.

Version 0.2.3 (2015-02-13)

Features

Modified skbio.stats.distance.pwmantel to accept a list of filepaths. This is useful as it allows for a smaller amount of memory consumption as it only loads two matrices at a time as opposed to requiring that all distance matrices are loaded into memory.
Added skbio.util.find_duplicates for finding duplicate elements in an iterable.

Bug fixes

Fixed floating point precision bugs in Alignment.position_frequencies, Alignment.position_entropies, Alignment.omit_gap_positions, Alignment.omit_gap_sequences, BiologicalSequence.k_word_frequencies, and SequenceCollection.k_word_frequencies (#801).

Backward-incompatible changes

Removed feature_types attribute from BiologicalSequence and all subclasses (#797).
Removed find_features method from BiologicalSequence and ProteinSequence (#797).
BiologicalSequence.k_word_frequencies now returns a collections.defaultdict of type float instead of type int. This only affects the "default" case, when a key isn't present in the dictionary. Previous behavior would return 0 as an int, while the new behavior is to return 0.0 as a float. This change also affects the defaultdicts that are returned by SequenceCollection.k_word_frequencies.

Miscellaneous

DissimilarityMatrix and DistanceMatrix now report duplicate IDs in the DissimilarityMatrixError message that can be raised during validation.

Version 0.2.2 (2014-12-04)

Features

Added plot method to skbio.stats.distance.DissimilarityMatrix for creating basic heatmaps of a dissimilarity/distance matrix (see #684). Also added _repr_png_ and _repr_svg_ methods for automatic display in the IPython Notebook, with png and svg properties for direct access.
Added __str__ method to skbio.stats.ordination.OrdinationResults.
Added skbio.stats.distance.anosim and skbio.stats.distance.permanova functions, which replace the skbio.stats.distance.ANOSIM and skbio.stats.distance.PERMANOVA classes. These new functions provide simpler procedural interfaces to running these statistical methods. They also provide more convenient access to results by returning a pandas.Series instead of a CategoricalStatsResults object. These functions have more extensive documentation than their previous versions. If significance tests are suppressed, p-values are returned as np.nan instead of None for consistency with other statistical methods in scikit-bio. #754
Added skbio.stats.power for performing empirical power analysis. The module uses existing datasets and iteratively draws samples to estimate the number of samples needed to see a significant difference for a given critical value.
Added skbio.stats.isubsample for subsampling from an unknown number of values. This method supports subsampling from multiple partitions and does not require that all items be stored in memory, requiring approximately O(N*M)`` space where Nis the number of partitions andM` is the maximum subsample size.
Added skbio.stats.subsample_counts, which replaces skbio.stats.subsample. See deprecation section below for more details (#770).

Bug fixes

Fixed issue where SSW wouldn't compile on i686 architectures (#409).

Deprecated functionality

Deprecated skbio.stats.p_value_to_str. This function will be removed in scikit-bio 0.3.0. Permutation-based p-values in scikit-bio are calculated as (num_extreme + 1) / (num_permutations + 1), so it is impossible to obtain a p-value of zero. This function historically existed for correcting the number of digits displayed when obtaining a p-value of zero. Since this is no longer possible, this functionality will be removed.
Deprecated skbio.stats.distance.ANOSIM and skbio.stats.distance.PERMANOVA in favor of skbio.stats.distance.anosim and skbio.stats.distance.permanova, respectively.
Deprecated skbio.stats.distance.CategoricalStatsResults in favor of using pandas.Series to store statistical method results. anosim and permanova return pandas.Series instead of CategoricalStatsResults.
Deprecated skbio.stats.subsample in favor of skbio.stats.subsample_counts, which provides an identical interface; only the function name has changed. skbio.stats.subsample will be removed in scikit-bio 0.3.0.

Backward-incompatible changes

Deprecation warnings are now raised using DeprecationWarning instead of UserWarning (#774).

Miscellaneous

The pandas.DataFrame returned by skbio.stats.distance.pwmantel now stores p-values as floats and does not convert them to strings with a specific number of digits. p-values that were previously stored as "N/A" are now stored as np.nan for consistency with other statistical methods in scikit-bio. See note in "Deprecated functionality" above regarding p_value_to_str for details.
scikit-bio now supports versions of IPython < 2.0.0 (#767).

Version 0.2.1 (2014-10-27)

This is an alpha release of scikit-bio. At this stage, major backwards-incompatible API changes can and will happen. Unified I/O with the scikit-bio I/O registry was the focus of this release.

Features

Added strict and lookup optional parameters to skbio.stats.distance.mantel for handling reordering and matching of IDs when provided DistanceMatrix instances as input (these parameters were previously only available in skbio.stats.distance.pwmantel).
skbio.stats.distance.pwmantel now accepts an iterable of array_like objects. Previously, only DistanceMatrix instances were allowed.
Added plot method to skbio.stats.ordination.OrdinationResults for creating basic 3-D matplotlib scatterplots of ordination results, optionally colored by metadata in a pandas.DataFrame (see #518). Also added _repr_png_ and _repr_svg_ methods for automatic display in the IPython Notebook, with png and svg properties for direct access.
Added skbio.stats.ordination.assert_ordination_results_equal for comparing OrdinationResults objects for equality in unit tests.
BiologicalSequence (and its subclasses) now optionally store Phred quality scores. A biological sequence's quality scores are stored as a 1-D numpy.ndarray of nonnegative integers that is the same length as the biological sequence. Quality scores can be provided upon object instantiation via the keyword argument quality, and can be retrieved via the BiologicalSequence.quality property. BiologicalSequence.has_quality is also provided for determining whether a biological sequence has quality scores or not. See #616 for more details.
Added BiologicalSequence.sequence property for retrieving the underlying string representing the sequence characters. This was previously (and still is) accessible via BiologicalSequence.__str__. It is provided via a property for convenience and explicitness.
Added BiologicalSequence.equals for full control over equality testing of biological sequences. By default, biological sequences must have the same type, underlying sequence of characters, identifier, description, and quality scores to compare equal. These properties can be ignored via the keyword argument ignore. The behavior of BiologicalSequence.__eq__/__ne__ remains unchanged (only type and underlying sequence of characters are compared).
Added BiologicalSequence.copy for creating a copy of a biological sequence, optionally with one or more attributes updated.
BiologicalSequence.__getitem__ now supports specifying a sequence of indices to take from the biological sequence.
Methods to read and write taxonomies are now available under skbio.tree.TreeNode.from_taxonomy and skbio.tree.TreeNode.to_taxonomy respectively.
Added SequenceCollection.update_ids, which provides a flexible way of updating sequence IDs on a SequenceCollection or Alignment (note that a new object is returned, since instances of these classes are immutable). Deprecated SequenceCollection.int_map in favor of this new method; it will be removed in scikit-bio 0.3.0.
Added skbio.util.cardinal_to_ordinal for converting a cardinal number to ordinal string (e.g., useful for error messages).
New I/O Registry: supports multiple file formats, automatic file format detection when reading, unified procedural skbio.io.read and skbio.io.write in addition to OOP interfaces (read/write methods) on the below objects. See skbio.io for more details.
- Added "clustal" format support:
  - Has sniffer
  - Readers: Alignment
  - Writers: Alignment
- Added "lsmat" format support:
  - Has sniffer
  - Readers: DissimilarityMatrix, DistanceMatrix
  - Writers: DissimilarityMatrix, DistanceMatrix
- Added "ordination" format support:
  - Has sniffer
  - Readers: OrdinationResults
  - Writers: OrdinationResults
- Added "newick" format support:
  - Has sniffer
  - Readers: TreeNode
  - Writers: TreeNode
- Added "phylip" format support:
  - No sniffer
  - Readers: None
  - Writers: Alignment
- Added "qseq" format support:
  - Has sniffer
  - Readers: generator of BiologicalSequence or its subclasses, SequenceCollection, BiologicalSequence, NucleotideSequence, DNASequence, RNASequence, ProteinSequence
  - Writers: None
- Added "fasta"/QUAL format support:
  - Has sniffer
  - Readers: generator of BiologicalSequence or its subclasses, SequenceCollection, Alignment, BiologicalSequence, NucleotideSequence, DNASequence, RNASequence, ProteinSequence
  - Writers: same as readers
- Added "fastq" format support:
  - Has sniffer
  - Readers: generator of BiologicalSequence or its subclasses, SequenceCollection, Alignment, BiologicalSequence, NucleotideSequence, DNASequence, RNASequence, ProteinSequence
  - Writers: same as readers

Bug fixes

Removed constructor parameter from Alignment.k_word_frequencies, BiologicalSequence.k_words, BiologicalSequence.k_word_counts, and BiologicalSequence.k_word_frequencies as it had no effect (it was never hooked up in the underlying code). BiologicalSequence.k_words now returns a generator of BiologicalSequence objects instead of strings.
Modified the Alignment constructor to verify that all sequences have the same length, if not, raise an AlignmentError exception. Updated the method Alignment.subalignment to calculate the indices only once now that identical sequence length is guaranteed.

Deprecated functionality

Deprecated constructor parameter in Alignment.majority_consensus in favor of having users call str on the returned BiologicalSequence. This parameter will be removed in scikit-bio 0.3.0.
Existing I/O functionality deprecated in favor of I/O registry, old functionality will be removed in scikit-bio 0.3.0. All functionality can be found at skbio.io.read, skbio.io.write, and the methods listed below:
- Deprecated the following "clustal" readers/writers:
  - write_clustal -> Alignment.write
  - parse_clustal -> Alignment.read
- Deprecated the following distance matrix format ("lsmat") readers/writers:
  - DissimilarityMatrix.from_file -> DissimilarityMatrix.read
  - DissimilarityMatrix.to_file -> DissimilarityMatrix.write
  - DistanceMatrix.from_file -> DistanceMatrix.read
  - DistanceMatrix.to_file -> DistanceMatrix.write
- Deprecated the following ordination format ("ordination") readers/writers:
  - OrdinationResults.from_file -> OrdinationResults.read
  - OrdinationResults.to_file -> OrdinationResults.write
- Deprecated the following "newick" readers/writers:
  - TreeNode.from_file -> TreeNode.read
  - TreeNode.from_newick -> TreeNode.read
  - TreeNode.to_newick -> TreeNode.write
- Deprecated the following "phylip" writers:
  - Alignment.to_phylip -> Alignment.write
- Deprecated the following "fasta"/QUAL readers/writers:
  - SequenceCollection.from_fasta_records -> SequenceCollection.read
  - SequenceCollection.to_fasta -> SequenceCollection.write
  - fasta_from_sequences -> skbio.io.write(obj, into=<file>, format='fasta')
  - fasta_from_alignment -> Alignment.write
  - parse_fasta -> skbio.io.read(<fasta>, format='fasta')
  - parse_qual -> skbio.io.read(<fasta>, format='fasta', qual=<file>)
  - BiologicalSequence.to_fasta -> BiologicalSequence.write
- Deprecated the following "fastq" readers/writers:
  - parse_fastq -> skbio.io.read(<fastq>, format='fastq')
  - format_fastq_record -> skbio.io.write(<fastq>, format='fastq')

Backward-incompatible changes

skbio.stats.distance.mantel now returns a 3-element tuple containing correlation coefficient, p-value, and the number of matching rows/cols in the distance matrices (n). The return value was previously a 2-element tuple containing only the correlation coefficient and p-value.
skbio.stats.distance.mantel reorders input DistanceMatrix instances based on matching IDs (see optional parameters strict and lookup for controlling this behavior). In the past, DistanceMatrix instances were treated the same as array_like input and no reordering took place, regardless of ID (mis)matches. array_like input behavior remains the same.
If mismatched types are provided to skbio.stats.distance.mantel (e.g., a DistanceMatrix and array_like), a TypeError will be raised.

Miscellaneous

Added git timestamp checking to checklist.py, ensuring that when changes are made to Cython (.pyx) files, their corresponding generated C files are also updated.
Fixed performance bug when instantiating BiologicalSequence objects. The previous runtime scaled linearly with sequence length; it is now constant time when the sequence is already a string. See #623 for details.
IPython and six are now required dependencies.

Version 0.2.0 (2014-08-07)

This is an initial alpha release of scikit-bio. At this stage, major backwards-incompatible API changes can and will happen. Many backwards-incompatible API changes were made since the previous release.

Features

Added ability to compute distances between sequences in a SequenceCollection object (#509), and expanded Alignment.distance to allow the user to pass a function for computing distances (the default distance metric is still scipy.spatial.distance.hamming) (#194).
Added functionality to not penalize terminal gaps in global alignment. This functionality results in more biologically relevant global alignments (see #537 for discussion of the issue) and is now the default behavior for global alignment.
The python global aligners (global_pairwise_align, global_pairwise_align_nucleotide, and global_pairwise_align_protein) now support aligning pairs of sequences, pairs of alignments, and a sequence and an alignment (see #550). This functionality supports progressive multiple sequence alignment, among other things such as adding a sequence to an existing alignment.
Added StockholmAlignment.to_file for writing Stockholm-formatted files.
Added strict=True optional parameter to DissimilarityMatrix.filter.
Added TreeNode.find_all for finding all tree nodes that match a given name.

Bug fixes

Fixed bug that resulted in a ValueError from local_align_pairwise_nucleotide (see #504) under many circumstances. This would not generate incorrect results, but would cause the code to fail.

Backward-incompatible changes

Removed skbio.math, leaving stats and diversity to become top level packages. For example, instead of from skbio.math.stats.ordination import PCoA you would now import from skbio.stats.ordination import PCoA.
The module skbio.math.gradient as well as the contents of skbio.math.subsample and skbio.math.stats.misc are now found in skbio.stats. As an example, to import subsample: from skbio.stats import subsample; to import everything from gradient: from skbio.stats.gradient import *.
The contents of skbio.math.stats.ordination.utils are now in skbio.stats.ordination.
Removed skbio.app subpackage (i.e., the application controller framework) as this code has been ported to the standalone burrito Python package. This code was not specific to bioinformatics and is useful for wrapping command-line applications in general.
Removed skbio.core, leaving alignment, genetic_code, sequence, tree, and workflow to become top level packages. For example, instead of from skbio.core.sequence import DNA you would now import from skbio.sequence import DNA.
Removed skbio.util.exception and skbio.util.warning (see #577 for the reasoning behind this change). The exceptions/warnings were moved to the following locations:

FileFormatError, RecordError, FieldError, and EfficiencyWarning have been moved to skbio.util
BiologicalSequenceError has been moved to skbio.sequence
SequenceCollectionError and StockholmParseError have been moved to skbio.alignment
DissimilarityMatrixError, DistanceMatrixError, DissimilarityMatrixFormatError, and MissingIDError have been moved to skbio.stats.distance
TreeError, NoLengthError, DuplicateNodeError, MissingNodeError, and NoParentError have been moved to skbio.tree
FastqParseError has been moved to skbio.parse.sequences
GeneticCodeError, GeneticCodeInitError, and InvalidCodonError have been moved to skbio.genetic_code

The contents of skbio.genetic_code formerly skbio.core.genetic_code are now in skbio.sequence. The GeneticCodes dictionary is now a function genetic_code. The functionality is the same, except that because this is now a function rather than a dict, retrieving a genetic code is done using a function call rather than a lookup (so, for example, GeneticCodes[2] becomes genetic_code(2).
Many submodules have been made private with the intention of simplifying imports for users. See #562 for discussion of this change. The following list contains the previous module name and where imports from that module should now come from.

skbio.alignment.ssw to skbio.alignment
skbio.alignment.alignment to skbio.alignment
skbio.alignment.pairwise to skbio.alignment
skbio.diversity.alpha.base to skbio.diversity.alpha
skbio.diversity.alpha.gini to skbio.diversity.alpha
skbio.diversity.alpha.lladser to skbio.diversity.alpha
skbio.diversity.beta.base to skbio.diversity.beta
skbio.draw.distributions to skbio.draw
skbio.stats.distance.anosim to skbio.stats.distance
skbio.stats.distance.base to skbio.stats.distance
skbio.stats.distance.permanova to skbio.stats.distance
skbio.distance to skbio.stats.distance
skbio.stats.ordination.base to skbio.stats.ordination
skbio.stats.ordination.canonical_correspondence_analysis to skbio.stats.ordination
skbio.stats.ordination.correspondence_analysis to skbio.stats.ordination
skbio.stats.ordination.principal_coordinate_analysis to skbio.stats.ordination
skbio.stats.ordination.redundancy_analysis to skbio.stats.ordination
skbio.tree.tree to skbio.tree
skbio.tree.trie to skbio.tree
skbio.util.misc to skbio.util
skbio.util.testing to skbio.util
skbio.util.exception to skbio.util
skbio.util.warning to skbio.util

Moved skbio.distance contents into skbio.stats.distance.

Miscellaneous

Relaxed requirement in BiologicalSequence.distance that sequences being compared are of equal length. This is relevant for Hamming distance, so the check is still performed in that case, but other distance metrics may not have that requirement. See #504).
Renamed powertrip.py repo-checking script to checklist.py for clarity.
checklist.py now ensures that all unit tests import from a minimally deep API. For example, it will produce an error if skbio.core.distance.DistanceMatrix is used over skbio.DistanceMatrix.
Extra dimension is no longer calculated in skbio.stats.spatial.procrustes.
Expanded documentation in various subpackages.
Added new scikit-bio logo. Thanks Alina Prassas!

Version 0.1.4 (2014-06-25)

This is a pre-alpha release. At this stage, major backwards-incompatible API changes can and will happen.

Features

Added Python implementations of Smith-Waterman and Needleman-Wunsch alignment as skbio.core.alignment.pairwise.local_pairwise_align and skbio.core.alignment.pairwise.global_pairwise_align. These are much slower than native C implementations (e.g., skbio.core.alignment.local_pairwise_align_ssw) and as a result raise an EfficencyWarning when called, but are included as they serve as useful educational examples as they’re simple to experiment with.
Added skbio.core.diversity.beta.pw_distances and skbio.core.diversity.beta.pw_distances_from_table. These provide convenient access to the scipy.spatial.distance.pdist beta diversity metrics from within scikit-bio. The skbio.core.diversity.beta.pw_distances_from_table function will only be available temporarily, until the biom.table.Table object is merged into scikit-bio (see #489), at which point skbio.core.diversity.beta.pw_distances will be updated to use that.
Added skbio.core.alignment.StockholmAlignment, which provides support for parsing Stockholm-formatted alignment files and working with those alignments in the context RNA secondary structural information.
Added skbio.core.tree.majority_rule function for computing consensus trees from a list of trees.

Backward-incompatible changes

Function skbio.core.alignment.align_striped_smith_waterman renamed to local_pairwise_align_ssw and now returns an Alignment object instead of an AlignmentStructure
The following keyword-arguments for StripedSmithWaterman and local_pairwise_align_ssw have been renamed:
- gap_open -> gap_open_penalty
- gap_extend -> gap_extend_penalty
- match -> match_score
- mismatch -> mismatch_score
Removed skbio.util.sort module in favor of natsort package.

Miscellaneous

Added powertrip.py script to perform basic sanity-checking of the repo based on recurring issues that weren't being caught until release time; added to Travis build.
Added RELEASE.md with release instructions.
Added intersphinx mappings to docs so that "See Also" references to numpy, scipy, matplotlib, and pandas are hyperlinks.
The following classes are no longer namedtuple subclasses (see #359 for the rationale):
- skbio.math.stats.ordination.OrdinationResults
- skbio.math.gradient.GroupResults
- skbio.math.gradient.CategoryResults
- skbio.math.gradient.GradientANOVAResults
Added coding guidelines draft.
Added new alpha diversity formulas to the skbio.math.diversity.alpha documentation.

Version 0.1.3 (2014-06-12)

This is a pre-alpha release. At this stage, major backwards-incompatible API changes can and will happen.

Features

Added enforce_qual_range parameter to parse_fastq (on by default, maintaining backward compatibility). This allows disabling of the quality score range-checking.
Added skbio.core.tree.nj, which applies neighbor-joining for phylogenetic reconstruction.
Added bioenv, mantel, and pwmantel distance-based statistics to skbio.math.stats.distance subpackage.
Added skbio.math.stats.misc module for miscellaneous stats utility functions.
IDs are now optional when constructing a DissimilarityMatrix or DistanceMatrix (monotonically-increasing integers cast as strings are automatically used).
Added DistanceMatrix.permute method for randomly permuting rows and columns of a distance matrix.
Added the following methods to DissimilarityMatrix: filter, index, and __contains__ for ID-based filtering, index lookup, and membership testing, respectively.
Added ignore_comment parameter to parse_fasta (off by default, maintaining backward compatibility). This handles stripping the comment field from the header line (i.e., all characters beginning with the first space) before returning the label.
Added imports of BiologicalSequence, NucleotideSequence, DNA, DNASequence, RNA, RNASequence, Protein, ProteinSequence, DistanceMatrix, align_striped_smith_waterman, SequenceCollection, Alignment, TreeNode, nj, parse_fasta, parse_fastq, parse_qual, FastaIterator, FastqIterator, SequenceIterator in skbio/__init__.py for convenient importing. For example, it's now possible to from skbio import Alignment, rather than from skbio.core.alignment import Alignment.

Bug fixes

Fixed a couple of unit tests that could fail stochastically.
Added missing __init__.py files to a couple of test directories so that these tests won't be skipped.
parse_fastq now raises an error on dangling records.
Fixed several warnings that were raised while running the test suite with Python 3.4.

Backward-incompatible changes

Functionality imported from skbio.core.ssw must now be imported from skbio.core.alignment instead.

Miscellaneous

Code is now flake8-compliant; added flake8 checking to Travis build.
Various additions and improvements to documentation (API, installation instructions, developer instructions, etc.).
__future__ imports are now standardized across the codebase.
New website front page and styling changes throughout. Moved docs site to its own versioned subdirectories.
Reorganized alignment data structures and algorithms (e.g., SSW code, Alignment class, etc.) into an skbio.core.alignment subpackage.

Version 0.1.1 (2014-05-16)

Fixes to setup.py. This is a pre-alpha release. At this stage, major backwards-incompatible API changes can and will happen.

Version 0.1.0 (2014-05-15)

Initial pre-alpha release. At this stage, major backwards-incompatible API changes can and will happen.

Uh oh!

FilesExpand file tree

CHANGELOG.md

Latest commit

History

CHANGELOG.md

File metadata and controls

scikit-bio changelog

Version 0.7.4-dev

Features

Performance enhancements

Bug Fixes

Version 0.7.3

Features

Performance enhancements

Bug Fixes

Miscellaneous

Version 0.7.2

Features

Performance enhancements

Bug Fixes

Miscellaneous

Backward-incompatible changes

Version 0.7.1.post1

Features

Bug Fixes

Version 0.7.1

Features

Performance enhancements

Bug Fixes

Miscellaneous

Version 0.7.0

Features

Performance enhancements

Bug Fixes

Miscellaneous

Backward-incompatible changes

Version 0.6.3

Features

Performance enhancements

Bug fixes

Miscellaneous

Deprecated functionality

Backward-incompatible changes

Version 0.6.2

Features

Performance enhancements

Bug fixes

Miscellaneous

Deprecated functionality

Version 0.6.1

Features

Performance enhancements

Miscellaneous

Version 0.6.0

Performance enhancements

Features

Backward-incompatible changes [experimental]

Bug fixes

Miscellaneous

Version 0.5.9

Features

Backward-incompatible changes [experimental]

Important

Bug fixes

Miscellaneous

Version 0.5.8

Features

Backward-incompatible changes [experimental]

Deprecated functionality [experimental]

Bug fixes

Version 0.5.7

Features

Performance enhancements

Bug fixes

Miscellaneous

Version 0.5.6

Features

Backward-incompatible changes [stable]

Backward-incompatible changes [experimental]