Accelerate drug discovery with NVIDIA Clara™ for Biopharma, a collection of frameworks, applications, generative AI solutions, and pretrained models.
Accelerate breakthrough drug identification and improve the accuracy of target and compound selection.
Keep pace with AI innovation and drive outcomes within your organization.
Improve developer productivity and accelerate time to outcome.
Drug discovery spans many workflows, from exploring the chemical universe and predicting protein structures to scanning drug candidates and simulating molecules. Drive breakthroughs in these critical research areas with the powerful cloud APIs and tools available in the NVIDIA NGC™ catalog.
Foundation models understand the underlying data that they’re trained on, like the grammar and syntax of protein sequences. They can leverage these rules to represent input sequences as numerical matrices that can be used for predictions. The foundation models can then be fine-tuned to focus on particular tasks, like predicting protein structures or functions.
Molecular generative models encode chemical space, allowing researchers to optimize their search for molecules with desirable properties. Paired with a molecular docking model and leveraging 3D information from protein structure prediction models accelerates the end-to-end virtual screening of small molecules.
Single-particle cryo-EM is a popular experimental technique used to determine large molecular structures at near-atomic resolution. Cryo-EM is popular in structural biology because macromolecules can be imaged in near-native conditions. Cryo-EM has revolutionized drug discovery by providing insights into molecular structure and disease mechanisms. Machine learning methods for object detection, for example, accelerate the cryo-EM structure elucidation process.
NVIDIA BioNeMo™ is a supercomputing platform, built on the NVIDIA NeMo™ framework, for training and inferencing biomolecular large language models (LLMs) and to help scientists quickly identify candidate therapeutics. It contains AI models for predicting protein and small-molecule properties (ESM-1, ESM-2, MegaMolBART, MoFlow), protein generation (ProtGPT2), pose prediction (DiffDock), and 3D protein structure prediction (OpenFold, AphaFold2, ESMFold).
GROMACS is an open-source software package designed for molecular dynamics simulations of biomolecules, such as proteins, nucleic acids, and lipids. It plays a critical role in advancing our understanding of biological systems at the molecular level.
AutoDock is a growing collection of methods for computational docking and virtual screening for use in structure-based drug discovery and exploration of the basic mechanisms of biomolecular structure.
Image by Veronica Falconieri and Sriram Subramaniam, licensed from the National Cancer Institute under public domain
Deep learning-based approaches like RELION are powering high-throughput automation of cryo-EM for protein structure determination. RELION implements an empirical Bayesian approach for analysis of cryo-EM to refine singular or multiple 3D reconstructions as well as 2D class averages.
To understand protein structures with atomistic detail, tools like MELD can be used to infer structures from sparse, ambiguous, or noisy data. MELD harnesses data in a physics-based, Bayesian framework for improved protein structure determination.
Image courtesy of Evozyne
Learn more about NVIDIA BioNeMo, a platform composed of managed services, software application frameworks, and reference AI workflows that simplify, accelerate, and scale generative AI for drug discovery.
Using Generative AI to Enhance Biologics Discovery and Development
Drug Discovery Platform Explores Novel Chemical Space with Higher Accuracy
Accelerating Protein Structure Discovery
Fuel Faster Insights for Healthcare and Life Sciences
NVIDIA Privacy Policy