Use additional or different tree statistics to fit speciation models (yule, bd)

PoonLab / Kaphi

Kernel-embedded ABC-SMC for phylodynamic inference

GNU Affero General Public License v3.0

4 stars 2 forks source link

Use additional or different tree statistics to fit speciation models (yule, bd) #100

Closed MathiasRenaud closed 6 years ago

MathiasRenaud commented 7 years ago

First, determine which stats to use.

gtng92 commented 6 years ago

Some stats other than Kaphi::kernel attempted to fit Yule model

[x] Align
[x] Trip
[x] MAST
[x] TripL
[x] Sim
[x] Node.dist
[x] sackin
[x] colless
[x] phangorn::RF.dist
[x] phangorn::KF.dist

And for birth-death model

[x] Align
[x] Trip
[x] MAST
[x] TripL
[x] Sim
[x] Node.dist
[x] sackin
[x] colless
[x] phangorn::RF.dist
[x] phangorn::KF.dist

PDF results are stored on Nagisa in the Documents folder.

gtng92 commented 6 years ago

MAST metric freezes in parameter space early on for both models.

Trip and TripL don't work for Birth-death model. The error message for Trip occurs when trying to retrieve the MRCA of a triplet for each tree. The error message for TripL occurs because the two trees are apparently not of the same size, or do not retain the same labels. I will be investigating this further, but it's not a high priority at the moment.

I will close this issue when I forward the PDFs to @ArtPoon