search

xavierdidelot / TransPhylo

Reconstruction of transmission trees using genomic data
http://xavierdidelot.github.io/TransPhylo/
GNU General Public License v2.0
60 stars 22 forks source link

Multi-tree capability #34

Closed MarliesJFrancine closed 5 months ago

MarliesJFrancine commented 5 months ago

Hi, I read that TransPhylo can be used to infer transmission trees over multiple input trees (here: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7725332/). The article mentions analysis of multiple trees while using single parameters, but I am struggling to implement this in the following: res<-inferTTree(ptree,mcmcIterations=1000,w.shape=w.shape,w.scale=w.scale,dateT=dateT)

Is there an example of this available somewhere? Or can you please provide some guidance?

For background information: I have ran a BEAST analysis and so I have multiple phylogenetic trees. Instead of selecting a single MCC tree I wish to capture as much of the signal as possible and use this "multi-tree capability".

Thank you in advance!

Kind regards, Marlies

xavierdidelot commented 5 months ago

Hi Marlies,

The functionality that you mentioned is implemented in the command infer_multittree_shar_param and there is an example of how to use it here: https://xavierdidelot.github.io/TransPhylo/articles/multitree.html

Note however that this is NOT to be used on several tree sampled by BEAST in the way you described. Instead, the command infer_multittree_shar_param is designed to analyse several separate transmission clusters thought to share the same parameters.

To do what you described, you would need to run the standard function inferTTree separately on several trees sampled by BEAST and compare/combine results.

Best wishes, Xavier

MarliesJFrancine commented 5 months ago

Hi Xavier,

Great! Thank you for the clarification :)

Kind regards, Marlies