Covid-spike-classification

This is version 0.4.2 of covid-spike-classification.

Highlights of this version are:

* OCI container support via Docker Cloud builds (thanks cmkobel for
contributing)
* We now report the error probabilities for mutation calls based on the Phred
score of the mutated bases.

Detailed changes:

Carl Mathias Kobel (1):
Dockerfile (6)

Kai Blin (9):
README: Point docker cloud build status at the right docker hub repo
core: Track and report error probabilities for all reported mutations
README: Add note about OCI container route
README: Link to the source code for the full list of mutations
chore: Add CI automation, make flake8 happy
chore: Add CD automation to publish to pypi
Bump version to 0.4.2
core: Use more universal space as thousands separator in ratios
chore: Use right branch for CI tests

This is release 0.4.1 of covid-spike-classification.

This release adds support for using reads in nested input directories or
in zip files containing the reads in directories.

Detailed changes:

Kai Blin (3):
core: Support finding reads at arbitrary directory depths
chore: Add more gitignore patterns
Bump version to 0.4.1

This is release 0.4.0 of covid-spike-classification.

This release adds the telltale mutations for the new VOC/VUIs A.23.1 and
B.1.525 as far as they fall into the sequencing window.

It also reduces the samtools sort memory consumption.

Detailed changes:

Kai Blin (7):
README: Update tested versions
README: Clarify that we call mutations, not variants
core: Make important mutations check more flexible for the future
core: Add telltale mutations for A.23.1 and B.1.525
core: limit samtools sort memory consumption
README: Update install instructions now that we're on bioconda
Bump version to 0.4.0

This is release 0.3.1 of covid-spike-classification.

This is a bugfix release increasing the robustness of the pileup
parsing for some edge cases.

Detailed changes:

Kai Blin (3):
README: Add bioconda install link
core: More robust pileup parsing
Bump version to 0.3.1

This is release 0.3.0 of covid-spike-classification.

This version removes attempts to call the variants in favour of just pointing
out the important mutations (E484K and N501Y at this time).
It also improves compatibility with Python 3.7 and supports classification of
reads that were called in the reverse complement.

Detailed changes:

Kai Blin (4):
core: Don't try to call variants anymore.
core: use copy2 instead of copytree for 3.7 compatibility
core: Also support reads that are mapped in reverse complement
Bump version to 0.3.0

This is release 0.2.4 of covid-spike-classification.

This release is a bugfix release that stops the whole script from aborting if
a single input trace is bad and fails to generate a proper FASTQ output,
causing bowtie2 to fail.

Detailed changes:

Kai Blin (4):
README: add link to repository
core: More robustly deal with bowtie2 falling over on some inputs
core: add some spaces to labels, validated on simulated tests
Bump version to 0.2.4