Information-theoretic analysis of multivariate single-cell signaling responses

Area of Science:

• Systems biology
• Information theory
• Computational biology

Background:

• Information theory offers a framework for understanding biological signaling.
• Current computational tools are limited to simple systems with few inputs/outputs.
• Robust and accessible implementations for complex systems are lacking.

Purpose of the Study:

• To introduce a novel algorithm, SLEMI, for analyzing high-dimensional signaling systems.
• To provide an efficient computational method for estimating mutual information in biological pathways.
• To enable broader application of information-theoretic approaches in systems biology.

Main Methods:

• Developed SLEMI (statistical learning-based estimation of mutual information) algorithm.
• Designed for high-dimensional outputs and numerous input values.
• Optimized for computational efficiency and reduced sample size requirements.

Main Results:

• SLEMI efficiently analyzes complex signaling systems.
• NF-κB signaling dynamics were analyzed in response to TNF-α.
• NF-κB signaling enhances discrimination of high TNF-α concentrations.

Conclusions:

• SLEMI provides an efficient and accessible tool for information-theoretic analysis of biological signaling.
• The approach facilitates understanding of how signaling dynamics influence stimulus discrimination.
• An R-package is available for computational biologists with basic information theory knowledge.