Skip to content

NVIDIA Digital Biology Research

Welcome to the NVIDIA Digital Biology Research Github Organization.

The repositories hosted here often accompany our publications and are released open source to the community to accelerate scientific innovation. Unless otherwise noted, the code is provided as-is and is not actively maintained.

Find out more about our research labs at research.nvidia.com/labs/dbr and research.nvidia.com and about our healthcare products at nvidia.com/clara/biopharma and nvidia.com/clara/genomics.

Explore our pinned repositories below to see some of our featured projects!

Pinned Loading

  1. ReaSyn ReaSyn Public

    ReaSyn is a model for predicting a molecule's synthesis pathway, reaction steps from building blocks to final product(s), using an encoder-decoder Transformer and a Chain-of-Reaction (CoR) notation.

    Python 44 8

  2. la-proteina la-proteina Public

    A partially latent flow matching model for the joint generation of a protein’s amino acid sequence and full atomistic structure, including both the backbone and side chain.

    Python 209 23

  3. nvMolKit nvMolKit Public

    A high-performance, GPU-accelerated library for key computational chemistry tasks, such as molecular similarity, conformer generation, and geometry relaxation.

    Cuda 187 14

  4. genmol genmol Public

    GenMol is a generative AI model for creating novel molecules. It utilizes masked discrete diffusion and fragment-based generation to create valid molecules, which are encoded in the SAFE molecular …

    Python 103 13

  5. proteina proteina Public

    Proteina is a new large-scale flow-based protein backbone generator that utilizes hierarchical fold class labels for conditioning and relies on a tailored scalable transformer architecture.

    Python 213 29

  6. cuik-molmaker cuik-molmaker Public

    cuik-molmaker is a specialized package designed for molecular featurization, converting chemical structures into formats that can be effectively used as inputs for deep learning models, particularl…

    C++ 19 1

Repositories

Showing 10 of 11 repositories
  • KERMT Public

    KERMT is a pretrained graph neural network model for molecular property prediction.

    NVIDIA-Digital-Bio/KERMT’s past year of commit activity
    Python 18 Apache-2.0 4 0 0 Updated Oct 15, 2025
  • avgflow Public

    Official code repository for the paper titled "Efficient Molecular Conformer Generation with SO(3) Averaged Flow-Matching and Reflow" (ICML 2025)

    NVIDIA-Digital-Bio/avgflow’s past year of commit activity
    Python 2 0 0 0 Updated Oct 14, 2025
  • nvMolKit Public

    A high-performance, GPU-accelerated library for key computational chemistry tasks, such as molecular similarity, conformer generation, and geometry relaxation.

    NVIDIA-Digital-Bio/nvMolKit’s past year of commit activity
    Cuda 187 14 7 2 Updated Oct 10, 2025
  • cuik-molmaker Public

    cuik-molmaker is a specialized package designed for molecular featurization, converting chemical structures into formats that can be effectively used as inputs for deep learning models, particularly graph neural networks (GNNs).

    NVIDIA-Digital-Bio/cuik-molmaker’s past year of commit activity
    C++ 19 1 0 0 Updated Oct 9, 2025
  • .github Public
    NVIDIA-Digital-Bio/.github’s past year of commit activity
    0 0 0 0 Updated Sep 29, 2025
  • la-proteina Public

    A partially latent flow matching model for the joint generation of a protein’s amino acid sequence and full atomistic structure, including both the backbone and side chain.

    NVIDIA-Digital-Bio/la-proteina’s past year of commit activity
    Python 209 23 2 0 Updated Sep 23, 2025
  • ReaSyn Public

    ReaSyn is a model for predicting a molecule's synthesis pathway, reaction steps from building blocks to final product(s), using an encoder-decoder Transformer and a Chain-of-Reaction (CoR) notation.

    NVIDIA-Digital-Bio/ReaSyn’s past year of commit activity
    Python 44 Apache-2.0 8 0 0 Updated Sep 23, 2025
  • genmol Public

    GenMol is a generative AI model for creating novel molecules. It utilizes masked discrete diffusion and fragment-based generation to create valid molecules, which are encoded in the SAFE molecular representation.

    NVIDIA-Digital-Bio/genmol’s past year of commit activity
    Python 103 13 4 0 Updated Aug 29, 2025
  • dualbind Public

    DualBind is a 3D structure-based deep learning model with a dual-loss framework for accurate and fast protein-ligand binding affinity prediction.

    NVIDIA-Digital-Bio/dualbind’s past year of commit activity
    Python 8 1 0 0 Updated Jul 29, 2025
  • proteina Public

    Proteina is a new large-scale flow-based protein backbone generator that utilizes hierarchical fold class labels for conditioning and relies on a tailored scalable transformer architecture.

    NVIDIA-Digital-Bio/proteina’s past year of commit activity
    Python 213 29 6 0 Updated Jul 24, 2025

Most used topics

Loading…