lncRNA_Mdeep: An Alignment-Free Predictor for Distinguishing Long Non-Coding RNAs from Protein-Coding Transcripts by...

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are vital in biological processes and diseases.
• Accurate identification of lncRNAs is essential for understanding their functions.
• Experimental methods for lncRNA identification are costly and time-consuming.

Purpose of the Study:

• To develop an efficient computational method for distinguishing lncRNAs from protein-coding transcripts.
• To present an alignment-free multimodal deep learning framework, lncRNA_Mdeep.

Main Methods:

• Developed lncRNA_Mdeep, a multimodal deep learning framework.
• Incorporated three distinct input modalities for analysis.
• Utilized deep learning for high-level representation learning and prediction.

Main Results:

• Achieved 98.73% prediction accuracy in human 10-fold cross-validation.
• Demonstrated 93.12% accuracy on an independent human test set, outperforming eight other methods.
• Showed strong predictive performance across 11 cross-species datasets.

Conclusions:

• lncRNA_Mdeep is a highly accurate and efficient tool for lncRNA identification.
• The framework offers a cost-effective and rapid alternative to experimental approaches.
• lncRNA_Mdeep shows potential for broad application in cross-species lncRNA prediction.