SC-framework: A robust and FAIR semi-interactive environment for single-cell resolution datasets

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Single-cell (SC) sequencing technologies enable detailed cellular heterogeneity analysis.
• Current SC data analysis methods are complex, lack standardization, and pose reproducibility challenges.

Purpose of the Study:

• To introduce the SC-Framework, a novel analysis environment for reproducible and flexible SC data analysis.
• To provide a guided, semi-interactive workflow for multi-modal SC data analysis.

Main Methods:

• Developed a FAIR-compliant, layered analysis environment combining a Python package and Jupyter Notebooks.
• Implemented self-documenting data objects and configuration-defined structures for traceability.
• Utilized containerized releases for long-term reproducibility on diverse computing platforms.

Main Results:

• Successfully retraced published single-cell RNA sequencing (scRNA-seq) and single-nucleus ATAC sequencing (snATAC-seq) results.
• Demonstrated scalability of the framework to analyze datasets with up to 1,000,000 cells.
• Validated the framework's ability to balance automation with interactivity for robust SC analysis.

Conclusions:

• The SC-Framework bridges the gap between automated pipelines and unstructured toolkits, enhancing SC analysis accessibility.
• It offers a standardized, reproducible, and flexible solution for multi-modal single-cell data analysis.
• This framework empowers a wider range of users to conduct robust SC data analysis.