A Simplified Complex Network-Based Approach to mRNA and ncRNA Transcript Classification

Bioinformatics is an interdisciplinary area that presents several important computational challenges. These challenges are usually related to the large volume of biological data generated and that needs to be analyzed for information discovery. An important challenge is the need to distinguish mRNAs and ncRNAs in an efficient and assertive way. The correct identification of these transcripts is due to the existence of thousands of non-coding transcripts, whose function and meaning are not known, as well as the challenge to understand the expression and regulation of genetic information. On the other hand, the complex network theory has been successfully applied in many real-world problems in different contexts. Therefore, this work presents a simplified and efficient complex network-based approach for the classification of mRNA and ncRNA sequences. Experiments were performed to evaluate the proposed approach considering a dataset with six different species and with important methods in the literature such as CPC, CPC2 and PLEK. The results indicated the assertiveness of the proposed approach achieving average accuracy rates higher than 98% in the classification of mRNA and ncRNA considering all compared species. Besides, the proposed approach presents fewer variations on its results when compared to competitor methods, indicating its robustness and suitability for the classification of transcripts.