Table 1. Notations.

Notation	Description
N	total # of genes
T	set of transcripts; T ik is the k th transcript of the i th gene; T i denotes the transcripts of the i th gene
l ik	length of transcript T ik
r	set of reads; r ij is the j th read aligned to the i th gene; r i is the read set aligned to the i th gene
p ik	the probability of a read generated by transcript T ik in the i th gene
P i	the probability of a read generated by transcript T i in the i th gene, specifically, [p i1, …, p i,|Ti|]
P	concatenate of all P i, specifically, [P 1,P 2, …,P N]
ρ ik	relative abundance of the transcript T ik in the i th gene
π	transcript expression; π ik is the expression of the k th transcript of the i th gene
ϕ ik	average expressions (normalized) of transcript T ik’s neighbors in the transcript network
α	parameters of Dirichlet distribution; α ik = λϕ ik + 1 is the parameter of the Dirichlet distribution of p ik
q ijk	read sampling probability, $q_{ijk}=\frac{1}{l_{ik}-l_r+1}$ if read r ij is aligned to transcript T ik, otherwise q ijk = 0
S	binary matrix for transcript interaction network