PolyA_DB version 4 catalogs polyA sites (PAS) in several mammalian genomes using both 3' end short-read RNA-seq (3'READS+) data and long-read RNA-seq (PacBio and Oxford Nanopore) data. Gene annotations are based on NCBI Gene and RefSeq databases. LR-RNA-seq data are used to connect PAS identified by 3'READS+ to genes and determine PAS types in the context of gene and splicing patterns. PolyA_DB v4 also contains PAS conservation and strength information. The data are linked to UCSC genome browser for visualization.
The version 4.1 contains ~1.4 million PAS for human and mouse genomes each, based on ~360 human and ~450 mouse 3'READS+ data sets corresponding to more than 2.3 billion PAS-supporting reads for each genome. About 20% of the full PAS set (polyA_DB v4.Max) match current, public LR-RNA-seq data. The matched PAS set is named polyA_DB v4.Main.