BiomedSciAI / biomed-multi-view
This repository contains the implementation of the Multi-view Molecular Embedding with Late Fusion (MMELON) architecture. MMELON combines molecular representations from three views — image, graph, and text —to learn a joint embedding that can be finetuned for downstream tasks in chemical and biological property prediction.
☆25Updated last month
Alternatives and similar repositories for biomed-multi-view:
Users that are interested in biomed-multi-view are comparing it to the libraries listed below
- SELFormer: Molecular Representation Learning via SELFIES Language Models☆90Updated 4 months ago
- Molecular Out-Of-Distribution☆37Updated last week
- Working collection of papers, repos and models of transformer based language models trained or tuned for the Chemical domain, from natura…☆59Updated last year
- Recursion's molecular foundation model☆44Updated 6 months ago
- Composing Unbalanced Flows for Flexible Docking and Relaxation (ICLR 2025)☆27Updated this week
- BARTSmiles, generative masked language model for molecular representations☆33Updated last year
- Sample-efficient Generative Molecular Design using Memory Manipulation☆33Updated 5 months ago
- Machine Learning dataset splitting for life sciences.☆27Updated 9 months ago
- MolKGNN is a deep learning model for predicting biological activity or molecular properties. It features in 1. SE(3)-invariance 2. confor…☆19Updated last year
- ☆23Updated 11 months ago
- Fine-tuning of ChemBERTA LLMs to predict molecules' HIV inhibition replication.☆26Updated last year
- Autoregressive fragment-based diffusion for target-aware ligand design☆28Updated 11 months ago
- Molecular Set Representation Learning☆48Updated 5 months ago
- ☆34Updated last month
- Code for our Nature Scientific Reports paper "A universal framework for accurate and efficient geometric deep learning of molecular syste…☆68Updated 5 months ago
- Diffusion-based molecule conformer generation☆40Updated last year
- Reagent prediction with Molecular Transformer. Improvement of data for reaction product prediction in a self-supervised fashion.☆24Updated 6 months ago
- Multi-target de novo molecular generator conditioned on AlphaFold's latent protein embeddings.☆65Updated 3 weeks ago
- ☆20Updated 3 weeks ago
- A Benchmarking Platform for Realistic And Practical Inverse Molecular Design☆74Updated last year
- Converts clipboard content to smiles and much more☆60Updated 10 months ago
- ☆24Updated this week
- Atom-in-SMILES tokenizer for SMILES strings.☆36Updated 8 months ago
- a multi-property optimization method.☆32Updated 4 months ago
- Training and inference code for ShEPhERD: Diffusing shape, electrostatics, and pharmacophores for bioisosteric drug design [ICLR 2025]☆62Updated this week
- 🔥 PyTorch implementation of GNINA scoring function for molecular docking☆61Updated last month
- Generative Pre-Training from Molecules☆20Updated 2 years ago
- A geometric flow matching model for generative protein-ligand docking and affinity prediction. (ISMB 2025)☆88Updated 2 weeks ago
- ☆42Updated 10 months ago
- MiniMol is a 10M-parameters molecular fingerprinting model pre-trained on >3300 biological and quantum tasks☆16Updated 6 months ago