Main content area

Fuzzy based algorithms to predict MicroRNA regulated protein interaction pathways and ranking estimation in Arabidopsis thaliana

Manikandan, P., Ramyachitra, D., Nandhini, R.
Gene 2019 v.692 pp. 170-175
Arabidopsis thaliana, databases, genes, microRNA, prediction, support vector machines
In living organisms, the MicroRNAs act as an important role by controlling regulatory mechanisms, and likely manipulating the output of numerous protein-coding genes. Several computational databases, algorithms and tools have been developed to discover the miRNA target genes. But, the existing methods obtain poorer results in identification of miRNA target gene. Hence in this research work, integrated prediction scores is used to identify the microRNA target interactions and hybrid fuzzy algorithms are used to make final predictions. The proposed algorithms such as Fuzzy, Fuzzy + Support Vector Machine (SVM) and Fuzzy + SVM + Random Forest (RF) are used to conduct prediction by majority voting and it is compared with the existing techniques such as SVM, RF and Neural Network (NN) to demonstrate the performance of the proposed algorithms. Furthermore, the ranking features are estimated using the Arabidopsis thaliana microRNA sequence. From the experimental results, it is inferred that the proposed Fuzzy + SVM + RF algorithm performs superior than the existing ones.