- Title
- Binary-organoid particle swarm optimisation for inferring genetic networks
- Creator
- Chanthaphavong, Santi; Chetty, Madhu
- Date
- 2010
- Type
- Text; Conference paper
- Identifier
- http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/46063
- Identifier
- vital:5943
- Identifier
-
https://doi.org/10.1109/CEC.2010.5586339
- Identifier
- ISBN:978-1-4244-6909-3
- Abstract
- A holistic understanding of genetic interactions is crucial in the analysis of complex biological systems. However, due to the dimensionality problem (less samples and large number of genes) of microarray data, obtaining an optimal gene regulatory network is not only difficult but also computationally expensive. In this paper, a Bayesian model for the genetic interactions using the Minimum Description Length as a scoring metric is proposed. For fast optimisation of the network structure, we propose a novel Swarm Intelligence algorithm called Binary-Organoid Particle Swarm (BORG-Swarm). In BORG-Swarm we introduce the concepts of probability threshold vector and particle drift to update particle positions. Experimental studies are carried out using real-life yeast cell cycle dataset. Results indicate that existing binary swarms fail to converge and suffer from long runtimes. In constrast, BORG-Swarm's fast convergence towards the global optimum becomes apparent from results of extensive simulations.
- Publisher
- Barcelona IEEE
- Relation
- Evolutionary Computation (CEC), 2010 IEEE Congress
- Rights
- Copyright IEEE
- Rights
- This metadata is freely available under a CCO license
- Subject
- 0801 Artificial Intelligence and Image Processing
- Reviewed
- Hits: 572
- Visitors: 561
- Downloads: 1
Thumbnail | File | Description | Size | Format |
---|