RPINBASE: An online toolbox to extract features for predicting RNA-protein interactions

Area of Science:

• Bioinformatics
• Computational Biology
• Molecular Biology

Background:

• Accurate feature extraction is crucial for machine learning models predicting RNA-protein interactions.
• Existing tools lack comprehensive structural feature extraction capabilities for this purpose.

Purpose of the Study:

• Introduce RPINBASE, an online toolbox for RNA-protein interaction prediction.
• Provide efficient structural feature extraction for diverse biological applications.

Main Methods:

• Developed an online toolbox, RPINBASE, accessible at http://rpinbase.com.
• Implemented efficient nested queries to accelerate data retrieval.
• Generated features in positive and negative sample formats.

Main Results:

• RPINBASE demonstrated capability in structural feature extraction for RNA-protein interactions.
• The toolbox was successfully applied to the aptamer design problem, yielding discussed results.

Conclusions:

• RPINBASE offers a novel solution for structural feature extraction in RNA-protein interaction studies.
• The toolbox enhances the speed and efficiency of preprocessing for machine learning applications in biology.