TileDB Launches Cross-Language Access to Single-Cell Data

TileDB has launched TileDB-SOMA, a collection of software libraries implementing the open-source SOMA API specification, developed in collaboration with the Chan Zuckerberg Initiative to enhance single-cell research. This initiative addresses the challenge of massive single-cell RNA sequencing (scRNA-seq) data growth by enabling large-scale computations on commodity hardware, thus eliminating data silos. TileDB-SOMA supports R and Python and allows interoperability with tools like Seurat, Anndata/Scanpy, and soon Bioconductor, facilitating data sharing and analysis across programming languages without requiring data conversion. The SOMA API is designed to make cloud-based single-cell data easily accessible for analysis by modeling datasets as groups of annotated sparse 2D matrices, suitable for various scientific data applications. The TileDB-SOMA platform, based on TileDB's open-source storage engine, accelerates research by streamlining access to large cloud-based datasets, thus enhancing the efficiency of complex queries and tackling data accessibility challenges faced by the single-cell community.