scverse/rapids-singlecell

rapids-singlecell: GPU-accelerated framework for scRNA analysis

What it solves

It addresses the performance bottlenecks associated with analyzing large single-cell datasets, which can be computationally expensive and slow on traditional CPUs.

How it works

The library uses CuPy and NVIDIA CUDA-X Data Science (RAPIDS) to move single-cell analysis workflows onto the GPU. It maintains an AnnData-first API that is largely compatible with Scanpy, allowing users to leverage GPU acceleration while staying within the scverse ecosystem, including integrated functionality from Squidpy, decoupler, and pertpy.

Who it’s for

Researchers and data scientists working with single-cell analysis who need to process large-scale biological datasets more efficiently.

Highlights

  • GPU Acceleration: Runs common single-cell workflows on the GPU for significantly faster processing.
  • Ecosystem Compatibility: Integrates seamlessly with Scanpy APIs and other scverse tools like Squidpy and pertpy.
  • AnnData-first API: Designed specifically for the AnnData data structure.
  • Agent Skills: Includes model-neutral analysis skills that can be installed for AI agents like Claude and Codex.

Related

  • Project
    NVIDIA/CUDALibrarySamples

    A collection of GPU-accelerated examples demonstrating the use of NVIDIA CUDA libraries for high-performance math, image processing, and linear algebra.

  • Project
    rapidsai/cugraph

    A collection of GPU-accelerated graph analytics packages that enable the creation and manipulation of graphs and the execution of scalable graph algorithms.

  • Project
    lattice/quda

    A GPU-accelerated library for lattice QCD calculations, providing optimized Dirac operators and solvers for various fermion actions on NVIDIA hardware.

  • Project
    NVIDIA/cuvs

    A GPU-accelerated library for vector search and clustering, providing high-performance approximate nearest neighbors implementations for AI and data mining.

  • Project
    scverse/scanpy

    A scalable Python toolkit for analyzing single-cell gene expression data, supporting preprocessing, clustering, and trajectory inference for datasets with over a million cells.