BioGeoBears GenSA test in RASP (Reconstruct Ancestral State Phylogeny)

[Caio Geroto]

Hello, thanks for accept me. I try run a analisys in RASP 4.1.2. and the manual recommend first run a test in BioGeoBears. RASP package come with GenSA and I start the test. I left de computer run for 72 hours and the test make little (if none progress). That is normal? Or I'm doing some thing wrong?

[Nick Matzke]

Hi -- I've never tried running BioGeoBEARS via RASP so I don't know what they've done. I would recommend just running the example script:

http://phylo.wikidot.com/biogeobears#script

...that should make the problems more obvious. Most likely issue: you have too many areas, so the state space is huge, so the likelihood calculation fills up the memory on the computer, so you never get through 1 likelihood calculation, let alone the ~1000 that GenSA requires. But that's just a guess.

Cheers!

Nick

On Sun, Jun 23, 2019 at 5:37 AM Caio Geroto <cge...@gmail.com> wrote:

Hello, thanks for accept me. I try run a analisys in RASP 4.1.2. and the manual recommend first run a test in BioGeoBears. RASP package come with GenSA and I start the test. I left de computer run for 72 hours and the test make little (if none progress). That is normal? Or I'm doing some thing wrong?

--
You received this message because you are subscribed to the Google Groups "BioGeoBEARS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biogeobears...@googlegroups.com.
To post to this group, send email to bioge...@googlegroups.com.
Visit this group at https://groups.google.com/group/biogeobears.
To view this discussion on the web visit https://groups.google.com/d/msgid/biogeobears/941cab6a-3687-48cd-95df-ae5ef9337701%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Nick Matzke]

PS: Where do you download RASP 4.1.2, quick googling suggests only older versions as far as I can see...

[plant.bio...@gmail.com]

The RASP website is here with the most current versions for all platforms. You can also directly email Yan Yu for specific questions on the implementation of algorithms introduced in BioGeoBEARS within RASP. 

To view this discussion on the web visit https://groups.google.com/d/msgid/biogeobears/CAJdu7BBs7whaVxkOrmmWL_owt1mSK1zmCWK5Wsd0DHUNjmVL4Q%40mail.gmail.com.

[David Zimmermann]

Had the same problem and informed Yan Yu. He promptly answered and also uploaded a new version of RASP, which works fine!

You can now choose between GenSA and optimx for the implemented BGB analyses.

[Caio Geroto]

Now it's working! Thank you all for the answers!!

On Tue, Jul 9, 2019 at 5:55 AM David Zimmermann <davidzimm...@gmail.com> wrote:

Had the same problem and informed Yan Yu. He promptly answered and also uploaded a new version of RASP, which works fine!

You can now choose between GenSA and optimx for the implemented BGB analyses.

--

You received this message because you are subscribed to the Google Groups "BioGeoBEARS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biogeobears...@googlegroups.com.
To post to this group, send email to bioge...@googlegroups.com.
Visit this group at https://groups.google.com/group/biogeobears.

To view this discussion on the web visit https://groups.google.com/d/msgid/biogeobears/ca649841-ab4d-46d6-b9d4-95cc1e753d01%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Amor pelo efeito deificador da organização, ó materialistas!... Por que o pensamento, sendo uma secreção do cérebro, é mais maravilhoso que a gravidade, uma propriedade da matéria? Por nossa arrogância, por nossa auto-admiração. - Charles Darwin