v0.7.3

Accidentally left the reciprocalspaceship dependency pinned to the development version on github, which PyPI does not like. Switched that to require reciprocalspaceship 1.0.6, which was just released for exactly this purpose.

What's Changed

• Update pyproject.toml by @dennisbrookner in #90

Full Changelog: v0.7.2...v0.7.3

v0.7.2

Conflicts with gemmi 0.7.3 are resolved

This release represents a more permanent version of the patch implemented in matchmaps 0.7.1. Now, matchmaps calls the new gemmi.interpolate_grid_around_model, rather than the old gemmi.interpolate_grid_of_aligned_model2. Accordingly, gemmi 0.7.3 is now strictly required; attempting to run this release of matchmaps with gemmi 0.7.2 or earlier is guaranteed to not work.

What's Changed

• fix(deps): update dependency gemmi to >=0.7.0, <=0.7.3 by @renovate[bot] in #83
• fix(deps): update dependency numpy to >=2, <=2.3.1 by @renovate[bot] in #88
• fix(deps): update dependency rs-booster to >=0.1.2, <=0.1.3 by @renovate[bot] in #85

Full Changelog: v0.7.1...v0.7.2

v0.7.1

Patch: update dependencies to guarantee gemmi 0.7.3 is not used.

gemmi 0.7.3 renames the function gemmi.interpolate_grid_of_aligned_model2 to gemmi.interpolate_grid_around_model.

Soon, I will push a new release using the function name (or possibly a release which checks for both function names). In the mean time, this emergency release ensures that a pip install matchmaps will use gemmi 0.7.2 or earlier.

What's Changed

• Run pre-commit hooks, excluding tests/data files by @dennisbrookner in #78
• Fast-forward testing branch by @dennisbrookner in #79
• Add a test suite by @dennisbrookner in #80
• ci(pre-commit.ci): autoupdate by @pre-commit-ci in #81
• Upper bound dependencies by @dennisbrookner in #82
• fix(deps): update dependency reciprocalspaceship to >=1.0.4, <=1.0.5 by @renovate in #84

Full Changelog: v0.7.0...v0.7.1

v0.7.0

Two large updates

It's time for a larger versioning bump up to v0.7.0, as this release represents a significant modernization of the codebase and an expansion of functionality

matchmaps.diagnose utility

This release adds a new command-line utility, matchmaps.diagnose. This can be used to evaluate the (non-)isomorphism between two datasets and decide if matchmaps is a good fit. For more details, see the newly added online guide

A test suite, finally

This release also adds a rudimentary test suite. Now, any changes to call signatures in gemmi or reciprocalspaceship should be caught by these tests. To this end, I also added the Renovate bot to keep track of new versions.

Note that, at this time, it is not feasible to run tests on the external PHENIX and CCP4 utilities, thought that would of course be great. Maybe a real wizard at GitHub Actions could take a stab at this.

Release notes

• Add unit tests by @dennisbrookner in #73
• chore: Configure Renovate by @renovate in #74
• Add support for python 3.12 by @dennisbrookner in #75
• Fast forward plot branch with test suite, python 3.12, etc. by @dennisbrookner in #76
• Add a matchmaps.diagnose plotting utility by @dennisbrookner in #70

New Contributors

• @renovate made their first contribution in #74

Full Changelog: v0.6.6...v0.7.0

v0.6.6

What's Changed

• Include force_anomalous_to_be_equal_to = false with phenix 1.21 by @dennisbrookner in #62
• ci(dependabot): bump codecov/codecov-action from 4 to 5 by @dependabot in #68
• Patch bug with gemmi >= 0.7.0 by @dennisbrookner in #72

Full Changelog: v0.6.5...v0.6.6

v0.6.5

Important changes

• matchmaps now supports both phenix 1.20 (and earlier, though I'm not sure how far back it goes) and the new phenix 1.21 syntax. matchmaps will attempt to automatically detect the user's phenix version based on the output from phenix.version, but the user can override this if it is wrong for some reason (e.g. multiple phenixes are present or something).
• Previously, when CCP4 or phenix subprocesses failed, and the program was not being run in verbose mode, these failures were totally silent to the user. The failure then manifested as an unhelpful FileNotFoundError. Now, the user is specifically informed of which subprocess has failed, and is instructed to use verbose mode for more information.

Abbreviated Changelog

• Add support for phenix 1.21 by @dennisbrookner in #53
• Refactor argument parsers to avoid duplication by @dennisbrookner in #55
• clarify that phenix 1.21 is supported by @dennisbrookner in #56
• Helpful error detection from sub-processes by @dennisbrookner in #58

Full Changelog: v0.6.4...v0.6.5

v0.6.4

What's Changed

• Drop support for python 3.8 by @dennisbrookner in #52

Documentation changes

• Add troubleshooting / advanced usage page to website by @dennisbrookner in #49
• link to paper rather than preprint by @dennisbrookner in #50

Full Changelog: v0.6.3...v0.6.4

v0.6.3

What's Changed

• ci(dependabot): bump styfle/cancel-workflow-action from 0.12.0 to 0.12.1 by @dependabot in #40
• Expanding the "about the algorithm" page by @dennisbrookner in #41
• ci(dependabot): bump codecov/codecov-action from 3 to 4 by @dependabot in #42
• Suppress error where scaleit fails on non-isomorphous inputs by @dennisbrookner in #44
• ci(dependabot): bump softprops/action-gh-release from 1 to 2 by @dependabot in #45
• Make explicit that phenix 1.21 is not currently supported by @dennisbrookner in #47
• bump rs-booster dependency to version 0.1.2 by @dennisbrookner in #48

Full Changelog: v0.6.2...v0.6.3

v0.6.2

This version adds support for error-weighting to matchmaps and matchmaps.mr. Various bug fixes are also included and are described below.

What's Changed

• Fix bug: unmasked_radius was parsed but ignored by @dennisbrookner in #34
• fast-forward weights branch with bug fix from main by @dennisbrookner in #35
• Weights by @dennisbrookner in #36
• ci(dependabot): bump actions/setup-python from 4 to 5 by @dependabot in #33
• Create new visualization.md page by @dennisbrookner in #37
• Fix crash when --input-dir and --output-dir match by @dennisbrookner in #39

Full Changelog: v0.6.1...v0.6.2

v0.6.1

This PR doesn't include any material change to the outputs of matchmaps. The inclusion of FoFc coefficients in phenix outputs is purely for comparative analysis and does not affect the matchmaps difference map. Changes in map normalization lead to changes in contour levels, but not in the appearance of maps themselves when viewed at appropriate contour.

What's Changed

• Phenix outputs include FoFc map coefficients by @dennisbrookner in #30
• Implement proper map normalization by @dennisbrookner in #31

Full Changelog: v0.6.0...v0.6.1