NASCUP: Nucleic Acid Sequence Classification by Universal Probability

Nucleic acid sequence classification is a fundamental task in the field of bioinformatics. Due to the increasing amount of unlabeled nucleotide sequences, fast and accurate classification of them on a large scale has become crucial. In this work, we developed NASCUP, a new classification method that captures statistical structures of nucleotide sequences by compact context-tree models and universal probability from information theory. A comprehensive experimental study involving nine public databases for functional non-coding RNA, microbial taxonomy and coding/non-coding RNA classification demonstrates the advantages of NASCUP over widely-used alternatives in efficiency, accuracy, and scalability across all datasets considered. NASCUP achieved BLAST-like classification accuracy consistently for several large-scale databases in orders-of-magnitude reduced runtime, and was applied to other bioinformatics tasks such as outlier detection and synthetic sequence generation.