Relative model selection of evolutionary substitution models can be sensitive to multiple sequence alignment uncertainty

Table 2 SP and TC score comparisons among variant MSA groups for model selection with AIC

Datatype	Score type	Comparison	Estimate (99% CI)
AA	SP	Differs-matches	− 0.002 (− 0.011, 0.007)
AA	SP	Stable-matches	0.014 (0.006, 0.022)
AA	SP	Stable-differs	0.016 (0.008, 0.024)
AA	TC	Differs-matches	− 0.007 (− 0.033, 0.018)
AA	TC	Stable-matches	0.022 (− 0.001, 0.045)
AA	TC	Stable-differs	0.029 (0.006, 0.052)
NT	SP	Differs-matches	− 0.003 (− 0.015, 0.01)
NT	SP	Stable-matches	0.043 (0.031, 0.055)
NT	SP	Stable-differs	0.046 (0.034, 0.058)
NT	TC	Differs-matches	− 0.006 (− 0.032, 0.021)
NT	TC	Stable-matches	0.093 (0.066, 0.119)
NT	TC	Stable-differs	0.098 (0.071, 0.125)

The “Comparison” column indicates the difference in effect sizes between the given groupings, which were compared in a post-hoc Tukey test with corrected P-values. Rows with adjusted \(P\le 0.01\) are shown in bold. Corresponding results for model selection with BIC and AICc are in Additional file 1: Tables S1 and S2, respectively

ISSN: 2730-7182