Developer Blog: Accelerating Single Cell Genomic Analysis using RAPIDS

The human body is made up of nearly 40 trillion cells, of many different types. Recent advances in experimental biology have made it possible to explore the genetic material of single cells. With the birth of this new field of single-cell genomics, scientists can now probe the DNA and RNA of individual cells in the human body.

Single-cell genomic analysis has identified new types of cells in the human body, discovered what makes these cells different from each other, and how different types of cells respond to disease or drugs. Single-cell genomics has also proven key in the current COVID-19 pandemic, identifying cells susceptible to infection and revealing changes in the immune systems of infected patients.

Schematic showing a matrix of gene activity across single cells, which is analyzed to produce a 2-D visualization showing clusters of similar cells.
Figure 1. Workflow for a single-cell RNA sequencing experiment. Individual cells are isolated and gene activity is measured in each cell. Cells with similar gene activity are clustered together to identify the various types of cells in the population.

The availability of single-cell data is continuously increasing, as are dataset sizes, with recent experiments sequencing millions of cells. This analysis is often exploratory and further benefits from being interactive – to identify different types of cells at finer scales, to compare the cell types and to visualize the relationships between them. Current workflows are still very slow, making them prohibitive for the interactive analysis needed for research.

RAPIDS: Accelerating data science with GPUs

RAPIDS is a suite of open-source libraries that can speed up end-to-end data science workflows through the power of GPU acceleration. RAPIDS makes it possible to perform interactive data analysis on large datasets using Python APIs that closely resemble NumPy, Pandas, and scikit-learn.

Read the full tutorial on the NVIDIA Developer Blog.