Hi,
I’m using the function “make.simmap” from Phytools to reconstruct ancestral
states in 10 trees with 248 tips obtained from a BEAST analysis. I believe
rates might be different between states (four states) and I think this
would probably be the best way to run it:
simmap.trees.bs <- make.simmap(trees.bs, x = hab.bs, nsim = 1, "ARD", Q =
"empirical", pi = "estimated")
However after a while it returns me an error:
Error in rmultinom(1, 1, y/sum(y)) : non-positive probability
Do you have any suggestions of why is this happening?.
Any help is greatly appreciated.
Laura Alencar.
R output:
simmap.trees.bs <- make.simmap(trees.bs, x = hab.bs, nsim = 1, "ARD", Q =
"empirical", pi = "estimated")
Warning: some elements of Q not numerically distinct from 0; setting to
1e-08
make.simmap is sampling character histories conditioned on the transition
matrix
Q =
1 2 3
4
1 -0.0968244043 0.06758894 0.029235453 0.0000000100
2 0.0225745893 -0.02675244 0.004024555 0.0001532927
3 0.0212390324 0.01319612 -0.053640721 0.0192055634
4 0.0003248535 0.00000001 0.006457870 -0.0067827337
(estimated using likelihood);
and (mean) root node prior probabilities
pi =
1 2 3 4
0.1141217 0.3555619 0.1363109 0.3940054
Done.
Warning: some elements of Q not numerically distinct from 0; setting to
1e-08
make.simmap is sampling character histories conditioned on the transition
matrix
Q =
1 2 3 4
1 -0.08346002 0.0724513259 0.01100869 0.000000010
2 0.01457896 -0.0317484605 0.00000001 0.017169490
3 0.01022154 0.0066068752 -0.02122808 0.004399662
4 0.01546898 0.0001664635 0.01762501 -0.033260454
(estimated using likelihood);
and (mean) root node prior probabilities
pi =
1 2 3 4
0.1391800 0.3730572 0.2607005 0.2270623
*Error in rmultinom(1, 1, y/sum(y)) : non-positive probability*
Laura Rodrigues Vieira de Alencar
PhD student
Department of Ecology
University of São Paulo
Phone 55-11-96190222
Website: alencarlaura.weebly.com
Lattes: *http://lattes.cnpq.br/3112833291166033
<http://lattes.cnpq.br/3112833291166033>*
Lab Meme: *http://ecologia.ib.usp.br/labmeme/Site/Inicio.html*
<http://ecologia.ib.usp.br/labmeme/Site/Inicio.html>
Labvert: *http://eco.ib.usp.br/labvert/* <http://eco.ib.usp.br/labvert/>
[[alternative HTML version deleted]]
_______________________________________________
R-sig-phylo mailing list - [email protected]
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at http://www.mail-archive.com/[email protected]/
