deepchem

GNN, GCN, Molecular Solubility, RDKit, Cheminformatics

Agent Skills for small-molecule and protein therapeutics: cheminformatics, molecular ML, docking and dynamics, protein design platforms, and target-discovery knowledge graphs.

XGBoost and GNN training and models for prediction of Hansen solubility parameters

Chempleter is a lightweight generative model to predict syntactically valid extensions of molecular structures.

A computational tutorial for predicting drug formulation properties using DeepChem

Using DeepChem with TensorFlow 2.x and Keras

HIV molecules inhibitor classification using GNN with attention

Molecule water solubility prediction using DeepChem.

Docker for reproducing solubility model

---------BBM420 -------- UNDERGRADUATE PROJECT FINAL REPORT

This is an introductory deep learning model which uses the Deepchem package. Lots of credit has to be given to the authors(Ramsundar B., Eastman P., Walters P., Pande V.) of the book "Deep Learning for the Life Sciences". This book has been a gold mine in terms of my exploration and journey in DeepLearning in the field of biochemistry.

some scripts using deepchem

A Graph Neural Network to classify human HIV inhibitor molecules using DeepChem feature extractor

End-to-end ML pipeline for drug toxicity and solubility prediction from SMILES. RDKit · DeepChem · XGBoost · Scaffold CV

practical notebooks for learning and applying machine learning in chemistry and cheminformatics.

This is a research project as part of the course CSE-700 : Independent Study, that I had taken up for Spring 2020 at the University at Buffalo.

Scaffold-split benchmark suite for ADMET property prediction (hERG, solubility, CYP3A4, Caco-2, clearance) with Morgan FP+RF and MPNN baselines, plus ECE/calibration metrics for uncertainty-aware evaluation. PyTorch · RDKit · DeepChem · scikit-learn.

Optimizing Mistral-7B for BACE-1 inhibitor prediction using Low-Rank Adaptation (r=8) to capture peak generalization on scaffold-split molecular data. Achieved 0.8034 ROC-AUC. Built for DeepChem GSoC '26.

Symbolic Regression layer implemented in PyTorch. This project is being developed as part of a proposal for GSoC 2026 with the DeepChem organization.

First open-source descriptor-augmented LLM for Neglected Tropical Disease drug discovery | OLMo-7B + QLoRA + DeepChem | Bioactivity & Toxicity prediction for Leishmaniasis, Chagas, Malaria, TB