Recurrent Neural Network Based Modeling of Gene Regulatory Network Using Bat Algorithm

摘要

Correct inference of genetic regulations inside a cell is one of the greatestchallenges in post genomic era for the biologist and researchers. Severalintelligent techniques and models were already proposed to identify theregulatory relations among genes from the biological database like time seriesmicroarray data. Recurrent Neural Network (RNN) is one of the most popular andsimple approach to model the dynamics as well as to infer correct dependenciesamong genes. In this paper, Bat Algorithm (BA) is applied to optimize the modelparameters of RNN model of Gene Regulatory Network (GRN). Initially theproposed method is tested against small artificial network without any noiseand the efficiency is observed in term of number of iteration, number ofpopulation and BA optimization parameters. The model is also validated inpresence of different level of random noise for the small artificial networkand that proved its ability to infer the correct inferences in presence ofnoise like real world dataset. In the next phase of this research, BA based RNNis applied to real world benchmark time series microarray dataset of E. coli.The results prove that it can able to identify the maximum number of truepositive regulation but also include some false positive regulations.Therefore, BA is very suitable for identifying biological plausible GRN withthe help RNN model.