Additional file 3: of QAPA: a new method for the systematic analysis of alternative polyadenylation from RNA-seq data

Poly(A) site feature matrix for modeling the APA regulatory code. This tab-delimited file contains feature matrix used to build a regression model of poly(A) site strength. Each row is a training/test example of a poly(A) site based on proximal and constitutive 3′ UTRs. Except for the first six columns, each column is a feature vector describing a particular property of the example’s poly(A) site. The response vector (“y”) is the average PAU of all samples from the Hubbard et al. RNA-seq dataset. Polyadenylation signal features begin with the prefix “PAS”. Poly(A) site dinucleotide features have the suffix “_pasite”. Overall upstream and downstream dinucleotide contents have the suffices “_upcontent” and “_downcontent”, respectively. Conservation features have the prefix “phyloP60way” followed by the bin location. Cis RBP motif features have the prefix “cis”. (TXT 18071 kb)