search

xavierdidelot / ClonalFrameML

ClonalFrameML: Efficient Inference of Recombination in Whole Bacterial Genomes
GNU General Public License v3.0
109 stars 27 forks source link

Can I get branch-specific values for R/Theta, 1/delta and nu from a standard CFML run? #115

Closed jodsikorski closed 4 years ago

jodsikorski commented 4 years ago

Dear Xavier,

When running CFML in standard mode (default settings) the em.txt output file provides me on the first 4 lines the recombination parameter for the entire data set. However, from line 5 on, the em.txt output file provides me per branch with additional 4 values which I do not yet have understood to what they refer to. What are these values? (I am not sure if they present the branch values.)

Please find below the first six lines of the *em.txt output file. Would it be possible to calculate from this output (standard model, default settings) the branch-specific values of the recombination parameter, without necessarily running CFML in the branchspecific mode (-embranch true)? If yes, then again by multiplying the values with the mean (which I assume are in lines 2-4)?

Thanks for your help, best, Johannes

Parameter Posterior Mean Posterior Variance a_post b_post R/theta 0.129686 3.21547e-06 5230.45 40331.8 1/delta 0.013032 3.24761e-08 5229.45 401279 nu 0.108303 2.70561e-07 43352.6 400290 Psal182 4.01294e-05 2.71643e-11 59.2827 1.47729e+06 AY6492A 1.98804e-06 1.34551e-12 2.9374 1.47754e+06

xavierdidelot commented 4 years ago

Dear Johannes,

When running in the default setting, the values in the em.txt from line 5 concern the lengths of the branches in the tree. You cannot deduce from this the branch-specific values of recombination parameters. To get this, you need to run with the -embranch true option.

Best wishes, Xavier

jodsikorski commented 4 years ago

Dear Xavier, thanks a lot for the information. best, Johannes