Original title: Comparison and benchmark of deep learning methods for non-coding RNA classification
Authors: Constance Creux,Farida Zehraoui,Francois Radvanyi,Fariza Tahi
This article explores the classification of non-coding RNAs, which are RNA molecules that do not directly code for proteins but play important roles in biological processes and diseases. The study traces the history of classifying non-coding RNAs, starting in the 1950s with the identification of housekeeping RNAs. Over time, additional classes have been discovered, leading to the need for computational methods that can effectively classify large sets of non-coding RNAs.
The article highlights the recent success of deep learning in various fields and how it has been applied to non-coding RNA classification. Despite the development of numerous novel deep learning architectures, there is a lack of comprehensive literature reviews on this topic. To address this gap, the authors propose a comparison of different approaches and datasets used in state-of-the-art research.
The authors then conduct experiments to evaluate the performance of various tools for non-coding RNA classification. Based on the results, they analyze the relevance of different architectural choices and provide recommendations for future methods. The article concludes with a statement of no competing interests from the authors.
Original article: https://www.biorxiv.org/content/10.1101/2023.11.24.568536v1