29 projects
chemsplit
A self-contained, scikit-learn-compatible library of dataset-splitting strategies for cheminformatics machine learning.
papyrus-scripts
A collection of scripts to handle the Papyrus bioactivity dataset
ml2json
A safe, transparent way to share and deploy scikit-learn models.
jax-grover
A JAX implementation of the GROVER molecular representation model (dual-view GTransformer encoder)
papyrus-structure-pipeline
Papyrus Structure Pipeline
dimorphite-ojmb
An open-source program for enumerating the ionization states of drug-like small molecules
pydompekeygen
A Python implementation of DompéKeys
molspotter
Identify silly molecules
CDK-pywrapper
Python wrapper for CDK molecular descriptors and fingerprints
BlueDesc-pywrapper
Python wrapper for BlueDesc molecular descriptors
jcompoundmapper-pywrapper
Python wrapper for jCompoundMapper molecular fingerprints
mordred-ojmb
molecular descriptor calculator
PaDEL-pywrapper
Python wrapper for the PaDEL descriptors
Mold2-pywrapper
Python wrapper for Mold2 descriptors
chemopy2
A Python library calculating molecular descriptors.
prodec
A package to calculate protein sequence descriptors
syzygy-cfa
Combinatorial Fusion Analysis for Python
uncorrupt-smiles
Train and apply Transformer models that translate invalid SMILES back into valid ones.
jax-cddd
JAX port of CDDD (Continuous Data-Driven Descriptors) for molecular SMILES embedding/reconstruction
scaffound
Python implementation of Dompé's 'Molecular Anatomy'
solphedge
pH-dependent solubility predictions for small molecules
sa-filter
Identify structural alerts using the RDKit and ChEMBL.
mlchemad
Applicability domains for cheminformactics.
enumerate-smiles
Enumerate heterocycles, stereoisomers, tautomers and SMILES of molecules.
pepsift
Identify peptides and derivatives from small molecule datasets
map4-ojmb
MinHashed AtomPair Fingerprint of Radius 2
Signature-pywrapper
Python wrapper for signature molecular descriptors
BlueDesc
BlueDesc - Molecular Descriptor Calculator
munkres-rmsd
Proper RMSD calculation between molecules using the Kuhn-Munkres Hungarian algorithm.
