Locate-R: Subcellular localization of long non-coding RNAs using nucleotide compositions

Area of Science:

• Molecular Biology
• Bioinformatics
• Computational Biology

Background:

• Long non-coding RNAs (lncRNAs) are crucial for cellular functions and implicated in various diseases, including cancers.
• Understanding lncRNA sub-cellular localization is key to deciphering their roles in health and disease.
• The rapid increase in known lncRNA data necessitates efficient computational methods for localization prediction.

Purpose of the Study:

• To develop a novel, accurate, and efficient computational method for predicting the sub-cellular localization of long non-coding RNAs (lncRNAs).
• To provide a user-friendly web application for researchers to predict lncRNA locations.

Main Methods:

• Feature extraction using n-gapped l-mer composition and l-mer composition.
• Selection of the top 655 features for model training.
• Development of a prediction model based on locally deep support vector machines.

Main Results:

• The proposed Locate-R method demonstrates significantly enhanced prediction accuracy compared to existing state-of-the-art approaches.
• The model effectively utilizes sequence composition features for accurate lncRNA localization prediction.

Conclusions:

• Locate-R offers a powerful and efficient tool for predicting lncRNA sub-cellular localization.
• Accurate lncRNA localization prediction can accelerate the identification of disease biomarkers and therapeutic targets.