Open liamxg opened 9 months ago
Dear @liamxg, I believe you are trying to evaluate empirical frequencies only (with the -f e argument) in a protein MSA with missing states. Since several states, like C, F, and W, are not present in any sequence, empirical frequencies are disabled to avoid numerical issues. Therefore, there are no models to evaluate. Consider using the model-specific frequencies.
Dear @ddarriba,
For this situation, I have two choices: one is use -f f
, which is the default setting; another is use model-specific frequencies, right?
BTW, how to set the model-specific frequencies? Thanks.
Dear @ddarriba, I set the
-m MTZOA,PMB
, but it still reported asmodeltest: Candidate models set is empty
:modeltest x.y.z Copyright (C) 2017 Diego Darriba, David Posada, Alexandros Stamatakis License GPLv3+: GNU GPL version 3 or later http://gnu.org/licenses/gpl.html. This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law.
Written by Diego Darriba.
Physical cores: 4 Logical cores: 8 Memory: 8GB Extensions: AVX
WARNING: State C is missing in the alignment WARNING: State F is missing in the alignment WARNING: State W is missing in the alignment WARNING: Empirical frequencies will be disabled WARNING: ... There are 236 more warnings. Check the log file for details modeltest: Candidate models set is empty