vlanore
commented
5 years ago Hello,
There is a bug in tracecomp where it outputs wrong results when no burn-in is specified via command line.
To work around this, just manually specify your burn-in with -x <burn-in>.
For example in your case, this should do the trick:
tracecomp -x 240 samhdi_run_1200_1 samhdi_run_1200_2Let me know if this fixes your problem.
I'm sorry about this bug. tracecomp is an old piece of code that we haven't really looked into for a long time.
kanglizhu
According to the updated diffsel README, I have used samhd1_short.ali and samhd1.tree labelled with three "conditions" to conduct MCMC convergence checking and post-analysis. But I met some problems with MCMC convergence checking.
According to the paper “Detecting adaptive convergent amino acid evolution”, I know we can use two different approaches to establish MCMC convergence: the original method compares two MCMC chains using the tracecomp program, and the new method uses the Raftery and Lewis’s Diagnostic implemented in the R package coda after 200 iterations to estimate the number of necessary iterations, then run as many iterations as 120% of the estimated number and finally perform the same diagnostic to check convergence (as written in ‘diffselMCMCConvergenceAnalysis.Rmd’ script). For the original “ tracecomp” method, the chain has converged well if all lines have a large effsize (>300) and a small rel_diff (<0.1). For the “Raftery and Lewis” method, no error message printed means the chain has converged (is it right? ).
Using the samhd1_short.ali and samhd1.tree with 3 conditions, I used both methods to establish MCMC convergence. When the iterations in “Raftery and Lewis” method works well, the effsize of all lines in “ tracecomp” method is only about 5. I’m confused about this result, and I don’t know which step is wrong. So I have attached my result, sequence file and tree file, can you please help me check my results? check mcmc convergece .pdf samhd1.zip