Fig. 1
Principles and overall flow of EMSAR. An illustration of the key elements of EMSAR for single-end RNA-seq. Gene 1 has three splice isoforms and gene 2 has one. The two genes share some sequences, indicated by the yellow reads that are mapped to two locations. The RNA-seq reads are colored according to which transcripts share the read sequence. The read count (X1, …, X6) is the number of RNA-seq reads in the same ‘segments’ or that are shared by the same combination of transcripts. The length of the segments (L1, …, L6) correspond to the number of possible distinct virtual reads in each group. The read counts depend on the total expression level of the isoforms associated with each segment times the length of the segment. Transcripts are grouped into a sequence-sharing such that transcripts in the same segment belong to a set. A sequence-sharing set may contain transcripts from one or more genes. In this illustration, the top set (set 1) represents the transcripts shown on the left side. The four expression quantities (eA, …, eD) are estimated by considering the six segments simultaneously within this set