Title :
Constructing signaling pathways from RNAI data using genetic algorithms
Author :
Ayaz, E.S. ; Can, Tolga
Author_Institution :
Dept. of Bioinf., Middle East Tech. Univ. Inf. Inst., Ankara, Turkey
Abstract :
RNAi system allows us to see the phenotypes when some genes are removed from living cells. By observing these phenotypes, we can build signaling pathways without dealing with the chemistry inside the cell. However it is costly in terms of time and space complexity. Furthermore, there are some interactions RNAi data cannot distinguish that results in many different signaling pathways all of which are consistent with the RNAi data. In this paper, we combine genetic algorithms with some greedy approaches to find most of the networks that fits the RNAi experiments. Our algorithm works much faster than previous algorithms and finds many results in a small amount of time. The resulting topologies have equal priority which would be used as inputs of classification algorithms.
Keywords :
RNA; biological techniques; biology computing; genetic algorithms; molecular biophysics; RNA interference technology; RNAi data; RNAi system; classification algorithm input; genetic algorithms; greedy approaches; phenotypes; signaling pathway building; signaling pathways; Biological cells; Equations; Feedforward neural networks; Genetic algorithms; Genetics; Network topology; Topology;
Conference_Titel :
Health Informatics and Bioinformatics (HIBIT), 2011 6th International Symposium on
Conference_Location :
Izmir
Print_ISBN :
978-2-4673-4394-4
DOI :
10.1109/HIBIT.2011.6450816