kwdoyle / phylo-compare Goto Github PK

There's a convoluted problem wherein the script is failing to successfully create this file, which I guess is also causing the reverse DMC tree from being fully made?

Two trees need to have the same taxon namespace in order to run robinson_foulds_distance() on them. We're going to have to find some way to use the "same" taxon when writing the tree files.

This binary was compiled on GNU/Linux 2.6.24 and thus won't run on Mac OS X nor CentOS 7.1.1503, which are the only two Unix environments available to us.

iSG states a closing bracket is missing in ph1.nre when 50 nodes and 1000 edges are specified to make it, given by the example in the readme. It can successfully read the file if these numbers are scaled down proportionally to their smallest amount that is allowed (2 and 40).

It doesn't make any sense why adding more nodes/edges would break the nre file.

In the first step, if an attempt to make a tree with 7 or more nodes the following error is given:

Traceback (most recent call last):
File "framework_v2_Hirak.py", line 685, in
main()
File "framework_v2_Hirak.py", line 672, in main
G = forward_dmc(nodes,edges,outfile,qmod,qcon,treefile)
File "framework_v2_Hirak.py", line 276, in forward_dmc
G.add_edge(u,anchor_node_dic[u])
KeyError: 3

For the second dmc step in framework_v2.py, the tree file must have all object IDs removed in order to run. While these can be deleted manually, it would be nice if this could be automated.