Output files MSci - Pas-Kapli/bpp-tutorial GitHub Wiki
Running BPP
The following command executes BPP with the A00.bpp.ctl control file.
$ bpp --cfile A00.bpp.ctl
Outputs files:
1. Posterior sample of parameters.
In this analysis, the species relationships are fixed therefore the posterior samples stored in the "mcmc.txt" file contain only the theta, tau and phi parameters for the yeast phylogeny.
Below is how the first few lines of the file would look like:
Gen theta_6R theta_7H theta_8C theta_9B theta_10A theta_11D tau_6R tau_7H tau_8C tau_9B tau_10A tau_11D phi_H lnL
2 0.025386 0.039072 0.025316 0.014104 0.007243 0.008033 0.089481 0.068105 0.079800 0.060613 0.038491 0.068105 0.345577 -393529.027
4 0.025482 0.033927 0.032226 0.015595 0.007271 0.008063 0.089820 0.069352 0.080102 0.060842 0.038636 0.069352 0.345577 -393538.033
6 0.025482 0.024968 0.022767 0.015595 0.007455 0.008063 0.089820 0.069352 0.079859 0.060842 0.037679 0.069352 0.379608 -393532.595
8 0.017848 0.014851 0.021358 0.015007 0.007506 0.008117 0.090425 0.068810 0.080397 0.062329 0.037933 0.068810 0.379608 -393536.098
10 0.022353 0.016898 0.028849 0.015007 0.007506 0.008117 0.090425 0.068810 0.080397 0.062329 0.037933 0.068810 0.379608 -393554.914
.
.
.
2. General output and summary file.
This file is often named "out.txt" and it contains the information printed in the screen while executing the program.
Before the MCMC sampling the program prints in the "out.txt" some information for the input files, i.e.,
The first part of the output contains information regarding the data read from the input files as we show in the species tree inference example.
During the MCMC sampling
Performance traits and current estimates of some parameters are printed that can help in evaluating the efficiency of the run
At the end of the MCMC sampling
A summary of the posterior sample of the estimated parameters is provided at the end of the output.txt
The most information is the mean and the standard deviation for each of the sampled parameters.
The ESS values help us assess whether the analysis suffers from mixing problems. The larger this number is for each of the parameters the more efficient the MCMC sampling has been (i.e., covering a wider range of possible parameter combinations).
theta_6R theta_7H theta_8C theta_9B theta_10A theta_11D tau_6R tau_7H tau_8C tau_9B tau_10A tau_11D phi_H lnL
mean 0.016326 0.019114 0.027967 0.016614 0.009500 0.012635 0.093424 0.066804 0.079418 0.061169 0.037566 0.066804 0.302642 -393524.547195
median 0.016136 0.014790 0.027404 0.016392 0.009413 0.011993 0.093396 0.066851 0.079430 0.061180 0.037570 0.066851 0.302498 -393524.089000
S.D 0.003385 0.014772 0.005271 0.002719 0.001535 0.004188 0.002626 0.001440 0.001064 0.001016 0.000735 0.001440 0.069993 20.239520
min 0.005629 0.002527 0.010902 0.008135 0.004881 0.003198 0.083993 0.059519 0.074746 0.056280 0.033970 0.059519 0.056362 -393608.277000
max 0.036216 0.175607 0.081364 0.033302 0.018394 0.041609 0.104087 0.072244 0.083916 0.065164 0.040459 0.072244 0.615157 -393441.049000
2.5% 0.010326 0.005514 0.019366 0.011879 0.006785 0.006281 0.088339 0.063888 0.077285 0.059142 0.036090 0.063888 0.167412 -393565.186000
97.5% 0.023465 0.060793 0.039880 0.022565 0.012794 0.022560 0.098652 0.069507 0.081472 0.063127 0.038989 0.069507 0.440236 -393486.102000
2.5%HPD 0.009901 0.003558 0.018545 0.011579 0.006613 0.005402 0.088452 0.063972 0.077350 0.059157 0.036139 0.063972 0.167631 -393565.279000
97.5%HPD 0.022904 0.047034 0.038498 0.022074 0.012568 0.021014 0.098744 0.069578 0.081525 0.063137 0.039029 0.069578 0.440419 -393486.210000
ESS* 1436.575 1640.948 5787.539 9752.434 4751.187 1835.736 682.2608 5128.242 4361.526 7605.001 6151.933 5128.242 1580.856 5115.941100
Eff* 0.014366 0.016409 0.057875 0.097524 0.047512 0.018357 0.006823 0.051282 0.043615 0.076050 0.061519 0.051282 0.015809 0.051159
To verify the reliability of the results it would be necessary to repeat the analysis at least one more time.