v.1.5.2

Release 1.5.2

Minor changes

• Remove datasets and the option to input dataset labels [#52]

Bug fixes

• Fix version error when installing with setuptools [c36080d]

Full Changelog: v1.5.1...v1.5.2

v1.5.1

Release 1.5.1

Minor changes

• Add information to certain log messages [624d31a]
• Add heuristic to decide which clustering method to use, which can improve performance for long genomes that may have suffered since a recent change [#51]

Bug fixes

• Fix several edge case bugs with small input [#47]
• Fix crash on macOS with Python 3.8 and above [#50]

Full Changelog: v1.5.0...v1.5.1

v1.5.0

Release 1.5.0

Notable new features and improvements

• Enhance usability for large, diverse input — including more efficient sequence clustering, additional parallelization for faster runtime, and a new design_large.py program with pragmatic defaults [#46]

Minor changes

• Rename 'blacklist' to 'avoid' [f363cb3]

Bug fixes

None

Full Changelog: v1.4.1...v1.5.0

v1.4.1

Release 1.4.1

Minor changes

• Advances Python versions in integration tests and advances fixed numpy and scipy versions [#44]
• Improves the behavior of the MinHash family's hash functions [#45]

Bug fixes

• Fixes bug where an initial guess could fall outside of parameter bounds [#43]

Full Changelog: v1.4.0...v1.4.1

v1.4.0

Release 1.4.0

Notable new features and improvements

• Adds option to download only a specified segment [cd0bfa2]
• Adds option to save a table of the number of sequences each probe maps to [#36]
• Improves handling when a cluster or grouping contains no candidate probes [#41]

Minor changes

• Adds option to specify NCBI API key [dfc35b1]
• Adds option to save a list of accessions used in the design [8b896e5]
• Allows fasta and taxid input to probe coverage analysis program [f72271a]
• Adds a critical log message when a taxonomy ID is likely incorrect [a9e6a2e]

Bug fixes

• Fixes bug when specifying --expand-n 0 [074ac1c]
• Fixes bug where --cluster-from-fragments uses a long fragment length [#38]
• Fixes bug fetching influenza C virus neighbors [b78725e]
• Fixes bug where parameter values could be rounded down to below a bound [ba4640a]

v1.3.1

Release 1.3.1

Minor changes

• Improves installation section of README [a8bec30, 54bf91a, 2228aba]
• Adds code of conduct document [#29]
• Updates NumPy/SciPy requirements [#30]
• Simplifies package version determination [#31]

Bug fixes

None

v1.3.0

Release 1.3.0

Notable new features and improvements

• Adds options to cluster input sequences using MinHash signatures and design separately on each cluster, improving runtime [#25]
• Adds feature to automatically download sequences for given taxonomies and use these as input [#27]

Bug fixes

None

v1.2.0

Release 1.2.0

Notable new features and improvements

• Adds option to specify k-mer length for mapping [#16]
• Contains more recent sequence data, available through git lfs [b12cc14]
• Adds option to specify the maximum number of N bases to expand [32147ef]
• Now requires -o/--output argument, which should avoid confusion [#20]
• Improves runtime during setup [#21]
• Adds option to filter poly(A) sequence [#22]
• Adds option to skip input sequences that are too short [#23]

Bug fixes

• Fixes bug in generated .py files for some segmented datasets [#17 and 18a7f90]
• Fixes bug when an input sequence is shorter than k-mer length [d33fda1]
• Fixes bug that left open too many file descriptors in some cases, preventing calls to os.fork() [80641b2]

v1.1.0

Release 1.1.0

Notable new features and improvements

• Can now use locality-sensitive hashing to reduce space of candidate probes [#13]
• Can now specify a custom hybridization model [#14]
• More clear explanations of why probe count constraint might not be satisfied [#15]

Bug fixes

• Fixes bug in parsing --add-reverse-complements option [a7f9277]

v1.0.0

Release 1.0.0

First official release of CATCH.