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

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• MicroRNAs (miRNAs) are crucial regulators of gene expression in living organisms.
• Accurate identification of miRNA target genes is essential for understanding gene regulation.
• Existing computational methods for miRNA target prediction often yield suboptimal results.

Purpose of the Study:

• To develop and evaluate novel computational algorithms for more accurate identification of miRNA target genes.
• To enhance the prediction of microRNA-target interactions using integrated scores and hybrid fuzzy algorithms.

Main Methods:

• Development of hybrid fuzzy algorithms: Fuzzy, Fuzzy + Support Vector Machine (SVM), and Fuzzy + SVM + Random Forest (RF).
• Utilizing majority voting for final predictions with the proposed algorithms.
• Comparison of proposed algorithms against existing techniques like SVM, RF, and Neural Network (NN).
• Estimation of ranking features using Arabidopsis thaliana microRNA sequences.

Main Results:

• The proposed Fuzzy + SVM + RF algorithm demonstrated superior performance compared to existing methods.
• Hybrid fuzzy algorithms showed improved accuracy in predicting miRNA target interactions.
• Experimental validation using Arabidopsis thaliana microRNA sequences confirmed the efficacy of the new approach.

Conclusions:

• The integrated prediction scores combined with hybrid fuzzy algorithms offer a significant advancement in miRNA target gene identification.
• The Fuzzy + SVM + RF algorithm represents a powerful tool for researchers studying gene regulation by microRNAs.
• This research provides a more reliable method for predicting crucial microRNA-target interactions.