ABSTRACT: Known as the "fifth RNA nucleotide", pseudouridine (? or psi) is the first-discovered and most abundant RNA modification occurring at the Uridine site, and it plays a prominent role in a number of biological processes. Thousands of ? sites have been identified within different biological contexts thanks to the advancement in high-throughput sequencing technology; nevertheless, the transcriptome-wide distribution, biomolecular functions, regulatory mechanisms, and disease relevance of pseudouridylation are largely elusive. We report here a web server-PIANO-for pseudouridine site (?) identification and functional annotation. PIANO was built upon a high-accuracy predictor that takes advantage of both conventional sequence features and 42 additional genomic features. When tested on six independent datasets generated from four independent ?-profiling technologies (?-seq, RBS-seq, Pseudo-seq, and CeU-seq) as benchmarks, PIANO achieved an average AUC of 0.955 and 0.838 under the full transcript and mature mRNA models, respectively, marking a substantial improvement in accuracy compared to the existing in silico ?-site prediction methods, i.e., PPUS (0.713 and 0.707), iRNA-PseU (0.713 and 0.712), and PseUI (0.634 and 0.652). Besides, PIANO web server systematically annotates the predicted ? sites with post-transcriptional regulatory mechanisms (miRNA-targets, RBP-binding regions, and splicing sites) in its prediction report to help the users explore potential machinery of ?. Moreover, a concise query interface was also built for 4,303 known ? sites, which is currently the largest collection of experimentally validated human ? sites. The PIANO website is freely accessible at: http://piano.rnamd.com.