Skip to content

Releases: choderalab/yank

0.25.2 - Bugfix release

10 Nov 17:08
8c7bf01
Compare
Choose a tag to compare

Bugfixes

  • Fixed a bug that causes YANK to crash in when thermodynamic trailblazing is interrupted and then resumed with the option bidirectional_redistribution set (#1195).

0.25.1 - Bugfix and more robust yank status command

09 Nov 22:04
5400f46
Compare
Choose a tag to compare

Bugfixes

  • The command yank status now handles gracefully the case where it attempts to open a locked netcdf file (#855).
  • Fix bug causing YANK to crash sometimes when the option bidirectional_redistribution was activated (#1192).

0.25.0 - Moved multistate and mpi modules and new trailblaze algorithm

08 Nov 16:25
250e0de
Compare
Choose a tag to compare

API-breaking changes

  • The yank.mpi module and the objects in the yank.multistate package, which were deprecated in 0.24.0, have now been removed, and they can be found in the mpiplus and openmmtools libraries respectively.
  • The function yank.pipeline.trailblaze_alchemical_protocol has been renamed yank.pipeline.run_thermodynamic_trailblazing (#1180).

New features

  • The thermodynamic trailblazing algorithm used for the authomatic generation of the alchemical path is now capable of resuming after an unexpected interruption or crash. The samples generated during the process are used to initialize the replicas of the replica exchange or SAMS free energy calculation. This behavior can be controlled through the start_from_trailblaze_samples YAML option (#1176, #1180).
  • It is possible to control more options of the thermodynamic trailblazing algorithm and to discretize an alchemical path given through mathematical expressions enslaved to a generic variable (#1180).
  • Added a --setup-only flag in the yank script CLI command to run the automatic setup pipeline without running the free energy calculation (#1178).

Bugfixes

  • Fix a bug in which a list of experiments: [exp1, exp2] in the YAML file containing an unknown experiment name would fail silently without error (#1178).
  • Fixed a problem that would prevent YANK to work with Cerberus >= 1.2 (#1180).
  • Fixed the error message displayed when the anisotropic dispersion correction cutoff was too big for the box dimension (#1181).

Enhancements

  • By default, the automatic determination of the alchemical path now starts with the harmonic/flat-bottom restraint turned off and activate it in intermediate states instead of keeping the restraint activated throughout the calculation and reweighting in the analysis stage (#1176).

Known issues

  • Using parallel MPI processes causes poor mixing of the odd thermodynamic states while the mixing of the even states is normal. We're still investigating whether the issue is caused by a change in the MPI library or an internal bug. For now, we recommend running calculations using only 1 GPU (choderalab/openmmtools#449 and #1130).
  • Simulations restored from a checkpoint file have their velocities reset to zero (#1115).
  • Forward and backward convergence analysis free energy traces in the Jupter notebook are incorrect (#971).
  • Setup will fail if .mol2 atom substructure ID matches filename (#703).

0.24.1 - Bugfix release

01 Jun 15:37
3a2856b
Compare
Choose a tag to compare

Bugfixes

  • Fixed import of logsumexp() which was moved from scipy.misc to scipy.special (#1162).
  • Fixed a bug in which validation of molecules and system file paths did not consider paths relative to the YAML script path (#1164).
  • Improve the robustness of opening the netcdf file on resuming of the multi-state samplers by setting the environment variable HDF5_USE_FILE_LOCKING to 'FALSE' after 4 failed attempts (#1168).
  • Fixed a bug causing a crash during exception handling (#1168).
  • Fixed a bug causing a crash in selftest.py when loading OpenEye (#1170).

Enhancements

  • The class yank.utils.TLeap can now create octahedral boxes. This is not exposed in the YAML script yet, however (#1160).

0.24.0 - Experimental support for online status files

26 Feb 09:26
4dfcc8e
Compare
Choose a tag to compare
  • Added an experimental feature that allows --status to be provided with yank script invocations to emit a status.pkl file in the experiment directory at the conclusion of each experiment switch interval (#1135). This feature is experimental, and its invocation may be changed in a future release.
  • This release requires OpenMMTools >= 0.17.0, which includes a much faster way of implementing exact treatment of PME during alchemical calculations (#1136).
  • Some NetCDF robustness improvements.
  • SAMS improvements:
    • SAMS now properly restores t0 and stage, which should correctly persist the stage across phase switches and resumes.
    • Added SAMS logZ_history plot to notebook reports.
    • SAMS self._stage is now an integer to more easily store this in storage (0 is initial, 1 is asymptotic phase).
    • Suppress some verbose outputs.
    • If an asymptotically optimal stage is present in a SAMS simulation, the initial stage is discarded to equilibration and automatic equilibration detection is applied to the asymptotically optimal weight adjustment region only.
    • Jupyter notebook reports now show SAMS weight convergence if present.
  • The yank.multistate package was deprecated. In the future, this package will be available in OpenMMTools (#1146).

Known issues

  • The use of more than one process per experiment via MPI has known issues that are being debugged (#1130).
  • Simulations restored from a checkpoint file have their velocities reset to zero (#1115).
  • Forward and backward convergence analysis free energy traces in the Jupter notebook are incorrect (#971).
  • Peptide ligands are not currently supported (#376).
  • Setup will fail if .mol2 atom substructure ID matches filename (#703).

0.23.7 - Bugfix release

05 Oct 15:36
13d0717
Compare
Choose a tag to compare

Fix bug where resuming a simulation caused the checkpoint iteration to crash (#1103).

0.23.6 - Bugfix release

03 Oct 23:18
392ab21
Compare
Choose a tag to compare
  • Fix bug where redefining MCDisplacement/RotationMove in the YAML mcmc_moves block without specifying atom_subset caused the MC move to affect the whole system instead of only the ligand atoms (#1099).
  • Add net_charge option to the antechamber YAML pipeline for charged small molecules (#1099).
  • Added YAML syntax tutorial in the online documentation (#1099).

0.23.5 - Bugfix release

24 Sep 18:25
8ba0c2c
Compare
Choose a tag to compare
  • Fix bug where the stack trace of an exception raised by an MPI process would be logged incorrectly (#1084).
  • Improved robustness of NetCDF dataset opening and closing (#1084).
  • The algorithm identifying ions now detects also ions called "NA" and "CL" (#1086).
  • Fix bug that caused the standard state correction of Boresch restraints to be computed incorrectly (#1092).

0.23.4 - Bugfix release

25 Aug 23:58
eacee2a
Compare
Choose a tag to compare
  • Fix bug #1012 when determining automatically the group_size of replica exchange simulations with MPI (#1073).
  • Fix bug where passing WT to the modeller directive caused the automatic setup pipeline to crash (#1074).

0.23.3 Adds support for single mutations using Modeller

10 Aug 16:41
998d289
Compare
Choose a tag to compare
  • Adds an optional modeller directive to the molecules section of the YAML file through Modeller, a tool for comparative modeling of protein structures.
  • The following options are accessible through the modeller directive. (docs)
    • apply_mutations: Specify protein single mutations (e.g., T315I). (docs)
  • Multi-analyze feature can now create directories for multi-report output instead of raising an error.