🌟 BioNeMo: A Framework for Developing AI Models for Drug Design.

NVIDIA BioNeMo2 Framework is a set of tools, libraries, and models for computational drug discovery and design.

It accelerates the most time-consuming and expensive steps in building and adapting biomolecular AI models by providing optimized models and tools that are easily integrated into GPU-based computing resources.

The framework enables the creation, training and tuning of models, and its capabilities span a variety of workloads and therapeutic mechanisms: molecule generation, protein structure prediction, protein-ligand prediction and representation learning.

In addition to pipeline code, scripts and utilities, BioNeMo2 Framework contains:

▶️ Pre-trained models:

🟢 ESM-2 is a pre-trained bidirectional encoder (BERT-like) for amino acid sequences. BioNeMo2 includes checkpoints with parameters 650M and 3B;

🟢 Geneformer is a tabular scoring model that generates a dense representation of a cell's scRNA by examining co-expression patterns in individual cells.


▶️ Datasets:

🟠 CELLxGENE is a collection of publicly available single-cell datasets collected by the CZI (Chan Zuckerberg Initiative) with a total volume of 24 million cells;


🟠 UniProt is a database of clustered sets of protein sequences from UniProtKB, created on the basis of translated genomic data.


📌 Licensing: Apache 2.0 License.


🟡 Project page
🟡 Documentation
🖥 GitHub

#AI #ML #Framework #NVIDIA