AbstractMotivationTo capture structural homology in RNAs, predicting RNA structural alignments has been a fundamental framework around RNA science. Learning simultaneous RNA structural alignments in their rich scoring parameterization is an undeveloped subject because evaluating them is computationally expensive in nature.ResultsWe developed ConsTrain—a gradient-based machine learning method for rich structural alignment scoring. We also implemented ConsAlign—a simultaneous RNA structural

more »

... r composed of ConsTrain's learned scoring parameters. To aim for better structural alignment quality, ConsAlign employs (1) transfer learning from well-defined scoring models and (2) the ensemble model between the ConsTrain model and a mature thermodynamic scoring model. Keeping comparable running time, ConsAlign demonstrated the best alignment prediction quality among current RNA structural aligners.Availability and implementationOur code and our data are freely available at https://github.com/heartsh/consalign.Contactheartsh@heartsh.ioSupplementary informationSupplementary data are available at Bioinformatics