TMRS: an algorithm for computing the time to the most recent substitution event from a multiple alignment column

Table 2 Trained rate matrix and equilibrium distribution

Substitution type	Parameter	Rate
\(\text {A}\leftarrow \text {C}\), \(\text {T}\leftarrow \text {G}\)	\(\alpha\)	0.16
\(\text {A}\leftarrow \text {G}\), \(\text {T}\leftarrow \text {C}\)	\(\beta\)	0.57
\(\text {A}\leftarrow \text {T}\), \(\text {T}\leftarrow \text {A}\)	\(\gamma\)	0.20
\(\text {C}\leftarrow \text {G}\), \(\text {G}\leftarrow \text {C}\)	\(\delta\)	0.24
\(\text {C}\leftarrow \text {T}\), \(\text {G}\leftarrow \text {A}\)	\(\epsilon\)	0.59
\(\text {G}\leftarrow \text {T}\), \(\text {C}\leftarrow \text {A}\)	\(\eta\)	0.25

Nucleotide	Equilibrium frequency \(\pi\)
A, T	0.23
C, G	0.27

The elements of rate matrix \(R_{\text {Symmetric}}\) and its equilibrium frequency \(\pi\) are shown. Parameter variables correspond to matrix \(R_{\text {Symmetric}}\) in Table 1. We averaged the parameters optimized using 100,000 sampled alignment columns in the 4d sites. Due to the symmetry of rate matrix, complementary nucleotides have the same equilibrium frequency

ISSN: 1748-7188