Bug in sumt of MrBayes v3.2.6 ?

[carmelagissi]

Hi,

I am using MrBayes v3.2.6 x64. After completion of mcmc run and after correct run of sump, I got the following error with sumt: " A matrix must be specified before sumt can be used Error when setting parameter "Filename" (1)

As you can see below, the problem seems to be that the "filename" option of sumt does not work. So, I changed the names of my ".run1.t" and ".run2.t" files to "temp" but then I got this error:

" Summarizing trees in files "temp.run1.t" and "temp.run2.t" Using relative burnin ('relburnin=yes'), discarding the first 25 % of sampled trees Writing statistics to files temp.<parts|tstat|vstat|trprobs|con> Examining first file ... Found one tree block in file "temp.run1.t" with 10001 trees in last block Expecting the same number of trees in the last tree block of all files

Tree reading status:

0 10 20 30 40 50 60 70 80 90 100 v-------v-------v-------v-------v-------v-------v-------v-------v-------v-------v

• Data matrix should be defined before translation table could be set. Error in command "Translate"

Someone can help me?

thanks Carmela

---

Current settings:

Parameter Options Current Setting

---

Relburnin Yes/No Yes
Burnin 0
Burninfrac 0.25
Nruns 2
Ntrees 1
Filename temp<.run.t> Minpartfreq 0.10
Contype Halfcompat/Allcompat Halfcompat Conformat Figtree/Simple Figtree
Outputname temp<.parts etc> Calctreeprobs Yes/No Yes
Showtreeprobs Yes/No No
Hpd Yes/No Yes

---

MrBayes > sumt filename=55botryllidae_2out.nex

A matrix must be specified before sumt can be used Error when setting parameter "Filename" (1)

[nylander]

Dear carmelagissi,

I could not reproduce a faulty behavior of the sumt filename option, neither with the old 3.2.6, nor the current version of the code (3.2.7a). In the case where you wish to summarize tree files with other names than the data file, then you can do (example):

$ mb.3.2.6 -i primates.nex

MrBayes > set autoclose=yes nowarn=yes
MrBayes > mcmc nruns=2 nchains=1 ngen=10000
MrBayes > sump
MrBayes > sumt
MrBayes > quit

$ mv primates.nex.run1.t temp.run1.t
$ mv primates.nex.run2.t temp.run2.t
$ mb.3.2.6 -i primates.nex

MrBayes > sumt filename=temp

From the output of help sumt (note the base of the file name, i.e. temp in your example):

Filename  --  The name of the file(s) to be summarized. This is the base of
              the file name, to which endings are added according to the   
              current settings of the 'Nruns' and 'Ntrees' options.     

[carmelagissi]

Dear Sir,

thanks a lot for the fast answer. I used the command the are listed: the mcmc and the sump were OK, just sumt gave problems

However, if you don’t have this error, may be it is because I did something wrong during the installation of the program? Could it be? Should I try to install again? thanks

Carmela

Il giorno 20 giu 2019, alle ore 10:35, Johan Nylander notifications@github.com ha scritto:

Dear carmelagissi,

I could not reproduce a faulty behavior of the sumt filename option, neither with the old 3.2.6, nor the current version of the code (3.2.7a). In the case where you wish to summarize tree files with other names than the data file, then you can do (example):

$ mb.3.2.6 -i primates.nex

MrBayes > set autoclose=yes nowarn=yes MrBayes > mcmc nruns=2 nchains=1 ngen=10000 MrBayes > sump MrBayes > sumt MrBayes > quit

$ mv primates.nex.run1.t temp.run1.t $ mv primates.nex.run2.t temp.run2.t $ mb.3.2.6 -i primates.nex

MrBayes > sumt filename=temp From the output of help sumt (note the base of the file name, i.e. temp in your example):

Filename -- The name of the file(s) to be summarized. This is the base of the file name, to which endings are added according to the
current settings of the 'Nruns' and 'Ntrees' options.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/NBISweden/MrBayes/issues/111?email_source=notifications&email_token=AMMZ5YLKZMFGLWDLMZDL5ULP3M6NDA5CNFSM4HZMLDKKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYEWTJI#issuecomment-503933349, or mute the thread https://github.com/notifications/unsubscribe-auth/AMMZ5YKYQBRSPIDKTZFCFGTP3M6NDANCNFSM4HZMLDKA.

---

Prof. Carmela Gissi Università degli Studi di Bari "A. Moro" Dipartimento di Bioscienze, Biotecnologie e Biofarmaceutica Campus "E. Quagliariello" Via E. Orabona, 4 Nuovo Palazzo di Biologia, Primo piano, Studio 51 70125 Bari

tel: + 39 080 5443308 e-mail: carmela.gissi@uniba.it

Life is what happens to you while you're busy making other plans

[nylander]

Closing