Hardly the most pressing issue, but the current version of the SLiM manual (5.1, last updated in September 2025) states on p. 849 that the mmJukesCantor() and mmKimura() functions return 2-by-2 matrices. Surely this is wrong – it's easy to verify in the Eidos console that they produce 4-by-4 matrices, as one would expect if the goal is to model nucleotide substitution.
Hardly the most pressing issue, but the current version of the SLiM manual (5.1, last updated in September 2025) states on p. 849 that the
mmJukesCantor()andmmKimura()functions return 2-by-2 matrices. Surely this is wrong – it's easy to verify in the Eidos console that they produce 4-by-4 matrices, as one would expect if the goal is to model nucleotide substitution.