Hi everyone.
I am trying to date the divergence of Cuscuta from the rest of Convolvulaceae using the BEAST software package. Cuscuta is a parasitic lineage and therefore many of the generally well conserved plastid genes are highly mutated, resulting in very long branch lengths compared to the other taxa in the analysis.
I am using a lognormal uncorrelated relaxed clock model and have a target tree topology from a previous paper of my supervisor.
My problem is this: when I summarize the results of my analysis using TreeAnnotator (constraining the tree to my desired topology), I get many negative branches (see "Summary_StefanovicTopology_RelaxedClock.trees.txt"). I suspect this has something to do with the long branch lengths of Cuscuta because when I do an identical analysis but with Cuscuta removed I get a very well behaved tree (see "Summary_StefanovicTopology_NoCuscuta.trees.txt").
Obviously I cannot remove Cuscuta from my dataset as it is the genus in question, so any ideas as to why this problem occurs / how I can fix it would be much appreciated!
Sian
Hi Sian,
Try using the option "-heights ca" when you run TreeAnnotator. This stands for "common ancestor trees" and aims at summarizing clade ages across all posterior trees and not only the values for subset of trees that have that clade. If that subset of trees is low in frequency the average node height might end up being older than the direct ancestor.
Here is the explanation for negative branches from http://beast.community/faq.html:
"MCC trees produced by TreeAnnotator can have a descendent node that is older than its direct ancestor (a negative branch length). This may seem like an error but is actually the correct behaviour. The MCC tree is, by default, generated with average node heights across all trees in the sample which contain that clade. The negative branch lengths result when a clade is at low frequency and tends not to occur in those trees that have the MCC tree's ancestral clade (or vice versa). This means the average heights are for the adjacent nodes are derived from different sets of trees and may not have any direct ancestor-descendent relationship."
In this paper you will find more information about the "ca" option:
Article Looking for trees in the forest: Summary tree from posterior samples
I hope this helps,
Santiago
Hi Sian,
Try using the option "-heights ca" when you run TreeAnnotator. This stands for "common ancestor trees" and aims at summarizing clade ages across all posterior trees and not only the values for subset of trees that have that clade. If that subset of trees is low in frequency the average node height might end up being older than the direct ancestor.
Here is the explanation for negative branches from http://beast.community/faq.html:
"MCC trees produced by TreeAnnotator can have a descendent node that is older than its direct ancestor (a negative branch length). This may seem like an error but is actually the correct behaviour. The MCC tree is, by default, generated with average node heights across all trees in the sample which contain that clade. The negative branch lengths result when a clade is at low frequency and tends not to occur in those trees that have the MCC tree's ancestral clade (or vice versa). This means the average heights are for the adjacent nodes are derived from different sets of trees and may not have any direct ancestor-descendent relationship."
In this paper you will find more information about the "ca" option:
Article Looking for trees in the forest: Summary tree from posterior samples
I hope this helps,
Santiago
Santiago Sanchez-Ramirez
Thank you for your response! This fixed my problem. I had actually seen that quote before but did not think it would apply to target-topology trees as well as MCC trees!
Hi Sian,
I'm glad your problem has been fixed, and I thought I might throw in an additional comment. Any method that estimates branch lengths to achieve a max/min or optimum of some sort imposes compensation across branch length estimates. The error in each estimate propagates across all the other estimates, and these methods approximate a configuration of errors that is locally best across some sampling of the space of possibilities. I think your insight about correlations in branch length estimates was a good one.
Cheers,
Guy
- Badges
- Science method
- Similar topics
- Biological Science
- Evolutionary Biology
- Phylogenetics
- Phylogenetic Analysis