Official implementation for ActFound (Nature Machine Intelligence): A bioactivity foundation model using pairwise meta-learning
-
Updated
Oct 28, 2024 - Python
Official implementation for ActFound (Nature Machine Intelligence): A bioactivity foundation model using pairwise meta-learning
The official codebase for the paper "A Hitchhiker's Guide to Deep Chemical Language Processing for Bioactivity Prediction"
Hybrid Uncertainty Quantification for Bioactivity Assessment
BIOPREDICT: End-to-end QSAR framework for pIC50 prediction using Random Forest, PubChem fingerprints (PaDEL), and ChEMBL bioactivity data — registered intellectual property (IPO Pakistan, 2023)
QSAR Bioactivity Predictor is a Python application that allows users to create QSAR models to predict bioactivity for a specific target.
NOCTURNAL: Exploring the dark chemical space. A streamlined computational drug discovery platform from target identification to optimized drug visualization. Featuring a unique molecular optimization algorithm "MutaGen" and an interactive chemical space visualization module "ChemNet". All reinforced behind a modular, fault-tolerant architecture.
SMILES-based chemical language models (LSTM & Transformer) in PyTorch/🤗 Transformers for de novo drug design — beam search generation, perplexity-based molecule ranking, and bioactivity-conditioned fine-tuning.
QSAR + Streamlit app for predicting pIC50 of small molecules against SARS-CoV-2 Replicase Polyprotein — Random Forest, PubChem fingerprints, ChEMBL data, interactive web interface
Heterogeneous siamese neural network for bioactivity prediction using novel bioactivity representation
Predicting PARP 1 Inhibitors using Rep3Net
Bioactivity prediction of unknown chemical compounds for a new drug discovery for specific health problems.
Machine learning pipeline for QSAR modeling targeting human BACE1 inhibitors. Extracts chemical descriptors to predict bioactivity trends for virtual drug screening.
A modern, reproducible pipeline for molecular bioactivity prediction built as a final year research project. This repository integrates cheminformatics, advanced machine learning, and interactive visualization to accelerate drug discovery.
Bridging Predictive Reliability and Explainability: A Multi-Representation Deep Learning Framework for Chemical Space Analysis of Immune Bioassays
Per-target vs. multi-output LightGBM for 11-task bioactivity classification from SMILES (mean ROC AUC 0.668)
Code, curated data, and frozen outputs for 'Where and when a molecular property model can be trusted': random-forest disagreement ranks activity-model error in distribution, degrades unevenly under temporal shift, and should not be used as a conservative acquisition rule. Every reported number is machine-verified.
To associate your repository with the bioactivity-prediction topic, visit your repo's landing page and select "manage topics."