scverse @ Helmholtz Zentrum München
About
scverse is a consortium of foundational, mostly Python-based, tools for omics data in life sciences. It has been founded to ensure the long-term maintenance of these core tools and is a community project currently governed by the developers of the core packages.
The Theislab at Helmholtz Munich, as a founding member of the consortium, hosts and employs many of the core members.
Services
Reproducible Single-Cell Analysis with scverse
Single-cell omics technologies have revolutionized our understanding of biological systems by enabling us to profile cellular constituents precisely at scale to uncover mechanisms underlying development, homeostasis, and disease. With the unprecedented growth in the data collected, there has been a similar dedicated effort to develop novel, scalable and effective computational and statistical tools to process and analyze the resulting data. Helmholtz Munich together with the scverse consortium address the challenges posed by the rapid expansion of single-cell computational tools, where incompatibilities in data formats, Application Programming Interfaces (APIs), and user interfaces have created obstacles for users and developers alike. In doing so, we provide tools for both established data, such as single-modality dissociated single-cell data, as well as for complex multi-omics spatial single-cell data. As a multi-institutional, open-source consortium, scverse provides core functionality that is both robust and maintainable, focusing on interoperable data formats and a supportive community structure. At its core, scverse includes widely used frameworks that support data analytic functions. The field has widely adopted scverse tools for various biological applications, from neuroscience, immuno-oncology and in constructing of single-cell reference atlases across a range of biological systems10. This community-driven approach ensures that these tools remain accessible and interoperable for a diverse user base, advancing an ecosystem in Python that meets the growing demands of single-cell profiling data analysis. However, to make the tools accessible to the broad range of its users, there is an acute need to organize training workshop programs that would provide an opportunity for users to interact directly with the tool developers and discuss usability and applicability for specific biological contexts.
Currently, the core scverse packages include:
Data formats:
Anndata: Manages annotated data matrices
MuData: Manages multimodal annotated data matrices
SpatialData: Manages spatially resolved annotated data matrices
Analysis tools:
Scanpy: Scalable single-cell RNA-seq data analysis
Muon: Scalable multimodal single-cell data analysis
Squidpy: Scalable spatially resolved transcriptomics data analysis
Scvi-tools: Deep learning with single-cell data
Scirpy: Adaptive immune receptor analysis
Community resources:
Single-cell best practices book: Interactive online book for best practice analysis of single-cell and spatial data
Training activities will comprise a modular training structure in:
● Beginner to advanced workflows in single-cell RNA-seq analysis following best practices
● Introductory sessions to Anndata’s data handling and integration
● Multimodal single-cell analysis using Muon and MuData
● Spatial data analysis with Squidpy and SpatialData
● Adaptive immune receptor analysis with scirpy
● Probabilistic modeling and deep learning of single-cell data with scvi-tools
Project Management
Helmholtz Zentrum München
Computational Health Center
Institute of Computational Biology
Theis Lab
Ingolstädter Landstraße 1
85764 Neuherberg
Contact:
Contact:
Contact:
Contact: