Deep learning predicts non-coding RNA functions from only raw sequence data

摘要: Abstract Non-coding RNAs (ncRNAs) are small non-coding sequences involved in gene regulation many biological processes and diseases. The lack of a complete comprehension their functionality, especially genome-wide scenario, has demanded new computational approaches to annotate roles. It is widely known that secondary structure determinant know RNA function machine learning based have been successfully proven predict from information. Here we show can be predicted with good accuracy raw sequence information without the necessity computing features which computationally expensive. This finding appears go against dogma being key RNA. Compared recent methods, proposed solution more robust boundary noise reduces drastically cost allowing for large data volume annotations. Scripts datasets reproduce results experiments this study available at: https://github.com/bioinformatics-sannio/ncrna-deep