Figure 4
From: A jumping profile Hidden Markov Model and applications to recombination sites in HIV and HCV genomes
A toy example of a jumping profile Hidden Markov Model. This is build from a multiple sequence alignment of nucleotides with two subtypes. The first subtype consists of three sequences with four consensus columns, the second subtype consists of two sequences with five consensus columns. With each match and insert state a vector is associated for the emission probability values corresponding to the nucleotides A, C, G and T. For clarity, some transitions are omitted. Also, the figure does not show the delete states immediately after B from which each match state can be reached nor the delete state immediately before E that can be reached from each match state. Fat lines indicate high transition probabilities, thin lines correspond to low probabilities.