CPPVec: an accurate coding potential predictor based on a distributed representation of protein sequence.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are vital in biological processes.
• High-throughput RNA sequencing generates vast amounts of data, necessitating efficient analysis tools.
• Existing computational methods for coding potential prediction have limitations in capturing local RNA sequence context.

Purpose of the Study:

• To develop a novel, fast, and accurate computational method for predicting RNA coding potential.
• To address the limitations of current methods by incorporating RNA sequence contextual information.
• To improve the analysis of large-scale RNA sequencing datasets.

Main Methods:

• Developed CPPVec, a novel alignment-free method for coding potential prediction.
• CPPVec utilizes the contextual information of RNA sequences.
• Implemented using distributed representation (doc2vec) of protein sequences translated from the longest open reading frame (ORF).

Main Results:

• CPPVec demonstrates high accuracy in predicting RNA coding potential.
• The method significantly outperforms existing state-of-the-art prediction tools.
• Experimental findings validate the effectiveness of exploiting RNA sequence contextual information.

Conclusions:

• CPPVec represents a significant advancement in computational prediction of coding potential.
• The method's ability to capture contextual information offers improved accuracy.
• CPPVec facilitates more effective analysis of lncRNAs and large RNA sequencing datasets.