Question

MrBayes chains converged, how to stop early?

0

Entering edit mode

9.5 years ago

Adrian Pelin ★ 2.6k

Hello,

My run is programmed for 5000000 MCMC steps, but it looks like it's converging early

Average standard deviation of split frequencies: 0.008884

1726000 -- [-112794.860] [...15 remote chains...] -- 32:50:30
1727000 -- [-112792.714] [...15 remote chains...] -- 32:50:20
1728000 -- (-112794.589) [...15 remote chains...] -- 32:50:08
1729000 -- [-112808.611] [...15 remote chains...] -- 32:49:56

Any way to stop this early? Or is it best to let it converge more?

Adrian

phylogeny bayesian mrbayes • 7.3k views

ADD COMMENT • link updated 2.1 years ago by dongzhang0725 • 0 • written 9.5 years ago by Adrian Pelin ★ 2.6k

score 1 · Answer 1 · 2015-04-22

1

Entering edit mode

9.5 years ago

qiyunzhu ▴ 130

Ctrl + C will do the job.

Then you need to manually add footers to the output files, *.t and *.p, to make them intact.

Then you can go back to MrBayes and do sump.

ADD COMMENT • link 9.5 years ago by qiyunzhu ▴ 130

0

Entering edit mode

seems like just adding end; at end of *.t files worked. Thanks! I wish there was an easier way to do this.

ADD REPLY • link 9.5 years ago by Adrian Pelin ★ 2.6k

0

Entering edit mode

You can try PhyloSuite, see step 8 in the example: http://phylosuite.jushengwu.com/dongzhang0725.github.io/documentation/#5-12-1-Brief-example

ADD REPLY • link 2.1 years ago by dongzhang0725 • 0