Molscore

MAJOR CHANGES
- Bloom filter scoring function
- Added rDock functionality
- Gypsum-DL bug fixes
- Improvements to GUI analysis
- Benchmark specific metrics (e.g., GuacaMol)
- Scoring function imports optional, unavailable upon fail

Bug fixes


v.1.1
MAJOR CHANGES
- Remove generative models & datasets these are separate and specific to generative models
- New benchmark mode with presets
- New class to automatically calculate metrics after benchmark
- Improved parallel/distributed computing API
- New scoring functions
- rDock
- Linker descriptors
- Reaction filters
- Silly bits
- Moved some scoring functions to server model to fix and separate dependencies
- LegacyQSAR functions (DRD2, GSK3B, JNK3)
- RAScore

This version was the version associated with "MolScore: A scoring and evaluation framework for de novo drug design. chemRxiv (2023)".

For more up-to-date code pull from the develop branch.

This version was used in "Augmented Hill-Climb increases reinforcement learning efficiency for language-based de novo molecule generation".

For more up-to-date code pull from the develop branch.

Code used in 'Comparison of structure- and ligand-based scoring functions for deep generative models: a GPCR case study'.

For more up-to-date code, pull from the develop branch.