Rethinking Performance Measures of RNA Secondary Structure Problems
This work tackles the problem of improving evaluation metrics for RNA structure prediction, which is crucial for researchers in computational biology and bioinformatics, though it appears incremental as it focuses on refining existing measurement approaches.
The paper addresses the limitations of traditional evaluation measures for RNA secondary structure prediction by proposing the Weisfeiler-Lehman graph kernel as an alternative metric, which enables fairer and more accurate assessment and provides guidance in RNA design experiments.
Accurate RNA secondary structure prediction is vital for understanding cellular regulation and disease mechanisms. Deep learning (DL) methods have surpassed traditional algorithms by predicting complex features like pseudoknots and multi-interacting base pairs. However, traditional distance measures can hardly deal with such tertiary interactions and the currently used evaluation measures (F1 score, MCC) have limitations. We propose the Weisfeiler-Lehman graph kernel (WL) as an alternative metric. Embracing graph-based metrics like WL enables fair and accurate evaluation of RNA structure prediction algorithms. Further, WL provides informative guidance, as demonstrated in an RNA design experiment.