0.7.8

THIS IS A RE-RELEASE OF 0.7.4
We ran into some issues with deployment to pypi, see #236

Major updates

• pytom_match_template.py now supports relion5 metadata input via the --relion5-tomograms-star option. You can provide a tomograms.star file from a tomogram reconstruction job and pytom-match-pick will fetch the necessary CTF and dose data by the tomogram id.
• Some small fixes were also made to relion5-style star file particle list to make reading back easier.

What's Changed

• dont publish attestation to pypi by @McHaillet in #240

Full Changelog: 0.7.7...0.7.8

0.7.7

Here we go again...

0.7.4

Major updates

• pytom_match_template.py now supports relion5 metadata input via the --relion5-tomograms-star option. You can provide a tomograms.star file from a tomogram reconstruction job and pytom-match-pick will fetch the necessary CTF and dose data by the tomogram id.
• Some small fixes were also made to relion5-style star file particle list to make reading back easier.

PR's

• deal with new ruff version and next try for tutorial-test.yml by @sroet in #224
• Update tutorial-tests.yml by @sroet in #225
• Get tutorial-tests.yml working by @sroet in #226
• Add bins option to both the tophat and extraction plot by @sroet in #227
• add check on input gpu indices by @sroet in #228
• Drop py 39 and upgrade by @sroet in #230
• Relion5 input by @McHaillet in #229

Full Changelog: 0.7.2...0.7.4

0.7.2

Major changes

• We have a site! (It's automatically deployed with changes to the main branch)
• The tomogram mask supplied to match_template can now be ignored upon extraction to prevent unnecessary recalculation.
• After some initial issues, reading of starfiles by RELION5 should be fixed when using relion5-compat mode in extract_candidates.
• Score maps can be written out in float16 mode to save some disk space.
• Defocus handedness can be specified in match_template and will calculate defocus offsets for each subvolume. The site got a new 'Benchmarks' tab where a post is added that illustrates the effects of specifying defocus handedness (check it out here).

PR overview

• fix defocus help unit by @sroet in #197
• mkdocs site for pytom-match-pick by @McHaillet in #199
• fix: strings signs in deploy docs workflow by @McHaillet in #203
• Update README.md by @sroet in #202
• update workflow permission to write github pages by @McHaillet in #204
• Update Developers.md by @McHaillet in #205
• Add ignore tomo mask option by @sroet in #209
• add check on tomo mask shape by @sroet in #210
• site updates: fix some usage informations, make authorship clear, add logo by @McHaillet in #212
• add E501 check by @sroet in #216
• Create CITATION.cff by @sroet in #218
• Fix relion5 compat by @McHaillet in #220
• logging.warn will be deprecated by @McHaillet in #223
• add comment to entry_points that tells developers not to use line bre… by @McHaillet in #219
• feature: calculate subvolume defocus values by @McHaillet in #217
• Enable 16 bit output by @sroet in #222

Full Changelog: 0.7.1...0.7.2

0.7.1

What's Changed

• Add DOI badge to readme.md by @sroet in #186
• add tutorial test yml by @sroet in #187
• Update pyproject.toml by @sroet in #188
• fix: make output directories for tutorial tests by @McHaillet in #189
• revert healpix workaround by @sroet in #193
• bug fix: add int type for rng seed by @McHaillet in #192

Full Changelog: 0.7.0...0.7.1

0.7.0

This is a big one! And probably containts the last major api changes before the 1.0.0 release.

Wiki/tutorial will soon be updated to correspond with latest changes.

Main updates

• Thanks to @sroet the package now supports arbitrary angular sampling generation via healpix. This now makes it possible to provide a particle diameter as input and let pytom-match-pick figure out the correct angular sampling!
• A bug was fixed in the particle symmetry option as it was not applied to the correct euler angle.
• CTF parameters can now only be applied in pytom_match_template.py and no longer for template creation. This makes input more consistent.
• Particle lists can be written out to be Relion5 compatible: centered coordinates in angstrom and proper star file column names.
• A tomogram mask can be provided during template matching and will skip subvolumes where the mask is empty.

PRs

• Readme illustration by @McHaillet in #156
• Phase plate ctf and argparse grouping by @McHaillet in #157
• default value phase shift entry point by @McHaillet in #159
• Update cupy requirements by @sroet in #167
• wait longer to prevent leaked processes by @McHaillet in #168
• small speedboost by precalculating conjugation by @McHaillet in #165
• expose tophat kernel connectivity and make n-false-postives a float by @McHaillet in #166
• make template non-symmetrical for template matching test to prevent race condition by @McHaillet in #169
• refactor volume splitting by @sroet in #170
• auto generate sampling angles by @sroet in #163
• Ctf rework by @McHaillet in #174
• Relion5 compatibility by @McHaillet in #178
• Center of mass fix by @McHaillet in #179
• calculate angular search from particle diameter by @McHaillet in #181
• fix symmetry bug by @sroet in #184
• phase randomization score map correction by @McHaillet in #182
• pre-commit with ruff by @McHaillet in #183
• Add tomogram mask option for template matching by @sroet in #185

Full Changelog: 0.6.0...0.7.0

0.6.0

What's Changed

• change default mrc permissive to false, try to autorecover, remove userfacing 'permissive' argument by @sroet in #140
• don't make plots for particle extraction in estimate_roc by @sroet in #147
• change 'ptm' to 'rln' in the extraction star file by @sroet in #149
• README.md now points to GitHub Discussions for usage questions by @sroet in #154

Full Changelog: 0.5.0...0.6.0

0.5.0

What's changed

• summations of tilts in the per-tilt weighted wedge has been improved leading to less aliasing
• center of tilt rotation is updated to properly correspond to Fourier space origin
• install instructions now point to the pip install from pypi!

Pull requests

• Try to fix race condition in testing by @sroet in #138
• update readme to point to pip install instead of source by @sroet in #139
• ctf model ramp weighting correction and updated interpolation center by @McHaillet in #141

Full Changelog: 0.4.3...0.5.0

0.4.3

Third (and hopefully last) release to try to get automatic pypi working

What's Changed

• Fix release workflow by @sroet in #137

Full Changelog: 0.4.2...0.4.3

0.4.2

second try for pypi release

What's Changed

• Update pyproject.toml by @sroet in #136

Full Changelog: 0.4.1...0.4.2