Lambda values too high

[Nitin R]

Hello,

I am running CAFE5 with 18 proteomes whose orthogroups were identified using Orthofinder. After filtering the counts file ( gene counts >100 and high variance), I ran CAFE5 in the base model as well as the gamma model (k=2). While all my runs have converged, I am getting a lambda value in the range of 1.42. Here is the output from my base run with 500 iterations:

Model Base Final Likelihood (-lnL): 401424
Lambda: 1.4208669703278
Epsilon: 0.0700954
Maximum possible lambda for this topology: 6.18024
128 values were attempted (0% rejected)

I am also attaching my gene counts, ultrametric tree as well as branched tree for your reference.

Are these high lambda values and what could be causing this error?

[Hahn, Matthew]

Hi Nitin,

Since your tree is expressed in such small units (which I assume are not millions of years), lambda is expressed in those units. I think the value is not necessarily that high compared to ones estimated in millions of years.

matt

--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hahnlabcafe/c47b4ec3-27cb-4e56-bb81-b2521e56b492n%40googlegroups.com.
<SpeciesTree_rooted.txt><SpeciesTree_ultrametric.tre><hog_gene_counts.tsv>

[Nitin R]

Hi Matt,

Thank you so much for such a quick response !

Nitin