Issues when searching trees from ram

[Adam Cossette]

All,

I taught myself how to use TNT so I may be using it in a different way than yourself – I do not use commands in the command line, I use the mouse and a computer with Windows. I would like to explain how I am getting my MPTs and if you may, please let me know what you are doing differently. The point of this activity is determining the relationships of the ingroup taxa via the creation of a strict consensus. I am using a morphological dataset and my matrix includes 112 taxa, 171 characters, 16 states. Below are the steps I am using to get the MPTs from TNT.

I load the .ss file into TNT, select 1000 megabytes of RAM and set the max trees as 99,999 which is as many as the interface allows when not using the hold function in the command line. I do not use collapsing rules.

Next I select a traditional search. 1000 replicates for this example, saving 10 trees per replicate and using the TBR swapping algorithm.

Then I run the search and get 620 trees, but some have overflowed.

Because there was an overflow of replications I re-run the analysis on the trees stored in RAM.

During the second round of TBR I get the max number of trees I selected in the beginning, if I set max trees to 10,000 then it returns that many, if max trees set to 99,999 then I get that many. This is where my confusion lies – I’ve looked at many references and cannot solve this issue. A third round of TBR produces the max number of trees selected in the beginning (99,999). 

Any help is appreciated! --- Adam

[Martín Ramírez]

Hi Adam, 

If the dataset produces more than 99,999 trees, you will fill the buffer with as many trees again and again in every independent search.

 

If you start a TBR with the buffer already filled by 99,999 trees, TNT will examine every one of them trying to find a better (shorter) tree. If it finds one, it will discard all the rest of the buffer and start from that better tree. If no better tree is found, then it will keep the original 99,999 trees.

I hope this helps.

Martin

--
Has recibido este mensaje porque estás suscrito al grupo "TNT-Tree Analysis using New Technology" de Grupos de Google.
Para cancelar la suscripción a este grupo y dejar de recibir sus mensajes, envía un correo electrónico a tnt-tree-analysis-using-n...@googlegroups.com.
Para ver esta conversación en el sitio web, visita https://groups.google.com/d/msgid/tnt-tree-analysis-using-new-technology/569f2762-863f-497c-8111-ef43e2f3743fn%40googlegroups.com.

[Adam Cossette]

Martin,

Thank you for the response! I used the hold1000000 function and had the same results. I don't think a traditional search is appropriate for this dataset but had luck with it in the past. I have now begun trying the New Technology search to see if there is an improvement.

Best,

Adam

[Mark]

One of the powers of the new technology search, over the traditional search is that in the traditional search, each next tree is one tbr (or spr) swap from a tree that preceded it, thus they are all related in a chain. 

For example

tree (A(B(CD))) —> tree (A(C(BD)) which is a branch-adjacent swap of B and C

In the new technology search, supposing you use the ratchet, instead of swapping and recalculating the tree score, the characters in the matrix are reweighed randomly, a new tree is found on the basis of the weighted data, so it may have only some topological similarity to the tree(s) already found, then the data are weighted back to equal and the new tree from weighting is used as a starting point for swapping on the unweighted data. 

Kevin Nixon and Pablo’s genius in this was understanding that this was a way to teleport around tree space and escape local minima. 

Tree fusing on the other hand takes chunks of various trees with the same subset of taxa but different sub-topologies and and grafts those reciprocally across trees. 

Be aware though, that the new technology search also does not necessarily on its own flesh out the full scope of equally parsimonious trees. 

Thus you would not be poorly served by following up a new technology search with the same branch breaking on the "trees from RAM” if you have say only 620 optimal trees in memory after but have room for 99999. It’s unlikely to find you shorter trees but it is a way to find more trees of the same score. It usually matters little to the strict consensus if the new tech search has well-covered tree space. May matter more for the resolution of the tips. 

Para ver esta conversación en el sitio web, visita https://groups.google.com/d/msgid/tnt-tree-analysis-using-new-technology/af74ace3-c01e-4429-bbaf-daddc5cd00ban%40googlegroups.com.