Network optimal retrieval of sparse perturbations for steady-state control.

Area of Science:

• Systems Biology
• Computational Biology
• Network Science

Background:

• Identifying optimal targets for biological network manipulation is challenging due to experimental constraints.
• Systems biology requires efficient methods to predict and achieve desired system states.

Purpose of the Study:

• Introduce NORSP (Network Optimal Retrieval of Sparse Perturbations), a novel computational framework.
• Enable prediction and control of biological system steady states using minimal perturbations.
• Provide a generalizable solution for systems-level experimental design.

Main Methods:

• Integrate network propagation with supervised subset selection.
• Utilize a sensitivity matrix derived from network topology for control prediction.
• Apply to undirected, directed, and signed biological networks.

Main Results:

• NORSP identifies minimal perturbation sets to shift biological systems to desired steady states.
• The framework demonstrates robustness, scalability, and experimental relevance in validations.
• NORSP reliably infers effective alternative targets even with obscured true perturbations.

Conclusions:

• NORSP offers a practical and generalizable approach for steady-state control in complex biological systems.
• The framework supports multi-omics hypothesis generation and systems-level experimental design.
• NORSP facilitates targeted perturbation experiments under realistic constraints.