Eigenvalue sensitivity-based analysis for evaluation of biological network stability versus disturbances

Area of Science:

• Systems Biology
• Network Science
• Computational Biology

Background:

• Complex biological systems are often represented as networks.
• Understanding network stability is crucial for biological insights.
• Identifying influential network components is a key challenge.

Purpose of the Study:

• To develop a novel stochastic strategy for identifying influential edges in biological networks.
• To propose a new criterion, the random sensitivity index matrix (RSIM), for evaluating edge influence on network stability.
• To compare the contribution of edges to network instability under varying disturbance levels.

Main Methods:

• Utilized network principles and control-theory basics, including Jacobian and eigenvalue sensitivity analysis.
• Developed the random sensitivity index matrix (RSIM) based on Monte Carlo algorithms.
• Applied RSIM to evaluate eigenvalue sensitivity of edges under stochastic disturbances.

Main Results:

• RSIM effectively identifies sensitive edges within biological networks.
• The method's results remained consistent across different percentages of stochastic disturbances.
• Simulations on the lactose operon and MAPK pathways validated the proposed method's performance.

Conclusions:

• The developed stochastic strategy and RSIM provide a robust approach for analyzing biological network stability.
• RSIM is a valuable tool for pinpointing critical edges that impact network dynamics.
• The findings offer new insights into the stability of complex biological systems.