Artificial intelligence
Signal Processing - Special issue: Genomic signal processing
Applying Two-Level Simulated Annealing on Bayesian Structure Learning to Infer Genetic Networks
CSB '04 Proceedings of the 2004 IEEE Computational Systems Bioinformatics Conference
Learning Bayesian Networks
Hi-index | 0.00 |
Understanding how genes are functionally related requires efficient algorithms to model networks from expression data. We report a heuristic search algorithm called Two-Level Simulated Annealing (TLSA) that is more likely to find the global optimal network structure compared to conventional simulated annealing and other searching schemes. We have applied this method to search for a global optimised network structure from a synthetic data set and an expression data set of S. cerevisiae mutants. We have achieved better precision and recall compared to other searching algorithms and are able to map relationships more accurately among functionally-linked genes.