Detecting Information Flow in Regulatory Networks
Author Information
Author(s): Ispolatov Iaroslav, Maslov Sergei
Primary Institution: Ariadne Inc.
Hypothesis
Can a probabilistic algorithm effectively identify the dominant direction of information flow in regulatory networks?
Conclusion
The proposed annealing algorithm effectively produces optimal layouts of protein networks, revealing biologically meaningful feedback links.
Supporting Evidence
- The annealing algorithm outperformed the greedy algorithm in speed and memory requirements.
- The algorithm was tested on various regulatory and signaling networks in human cells.
- The greedy algorithm is limited to networks with around 100-200 vertices.
Takeaway
This study created a smart way to organize complex networks of proteins, helping scientists understand how signals flow and where feedback happens.
Methodology
The study compared a probabilistic simulated annealing algorithm with a deterministic greedy algorithm to minimize feedback links in regulatory networks.
Limitations
The greedy algorithm is limited to smaller networks and may miss optimal solutions due to its local optimization approach.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website