Pertpy: an end-to-end framework for perturbation analysis

Area of Science:

• Genomics
• Computational Biology
• Molecular Biology

Background:

• Single-cell technologies allow molecular state measurement across diverse perturbations.
• Current analysis methods are limited in scalability and biological context integration for complex studies.

Purpose of the Study:

• To introduce pertpy, a scalable Python framework for analyzing large-scale single-cell perturbation experiments.
• To provide harmonized datasets, metadata, and efficient analysis tools for perturbation data.

Main Methods:

• Development of a modular Python framework (pertpy).
• Integration of harmonized perturbation datasets and metadata databases.
• Implementation of established and novel analysis methods, including metadata annotation and perturbation distances.

Main Results:

• Pertpy offers fast and user-friendly implementations for analyzing perturbation data.
• The framework facilitates efficient analysis by incorporating biological context.
• It interoperates with the scverse ecosystem and is designed for extensibility.

Conclusions:

• Pertpy addresses the need for scalable analysis of complex single-cell perturbation studies.
• The framework enhances the efficiency and accessibility of perturbation data analysis.
• It supports the integration of biological context for deeper insights.