Phamclust

Latest version: v1.3.3

Safety actively analyzes 681866 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

1.3.3

---------------------------

- Bug fix: `nucclust` pipeline was miscalculating distances due to a
combination of two factors which have now been rectified:
- invoking `blastn` with a bad field name, resulting in missing evalue
and bitscore from intermediate files
- parsing blast HSPs as lists and using negative integers to slice them,
such that the 'send' field was being used instead of 'qlen' in the
denominator

1.3.2

---------------------------

- Added new pipeline `nucclust` to cluster genomes using nucleotide sequences
directly instead of pham data. This pipeline requires that `blastn` be
installed and in the PATH. The pipeline uses similar `blastn` parameters as
VIRIDIC's default (sensitive) settings, and the same math for calculating
intergenomic similarity/distance. In our experience, it is much faster than
VIRIDIC, but also much slower than pham-based clustering and distance
calculation.

1.3.1

---------------------------

- Bug fix: `--heatmap-midpoint` default value changed from None to 0.5 to
prevent `TypeError` if this value is not specified

1.3.0

---------------------------

- Exposed heatmap colors and the midpoint of 3-color scales as commandline
arguments so users can run wild with their creative preferences

- Joblib version 1.4.0 was officially released, so bumped dependency. Versions
below this one are no longer supported in PhamClust

1.2.0

---------------------------

- Added support for joblib 1.4.dev0 (installed from GitHub), unlocking
``Parallel(return_as="generator_unordered")``. This restores pre-joblib speed
while maintaining the very low memory footprint enabled by joblib.

- Matrices can now be sorted in tree order; call ``.reorder()`` on your matrix
without supplying a node-order and it will be put in tree order.

- PDF heatmaps traded for SVG format instead.

1.1.0

---------------------------

- Replaced internal parallelization with joblib.Parallel for drastically
reduced memory footprint. This results in approximately 10-20% reduction in
speed for the time being - see `In development`.

- Refactored `__main__.py` to separate the setup steps from phamclust logic

Links

Releases

© 2024 Safety CLI Cybersecurity Inc. All Rights Reserved.