smc++ split #272
Description
Hi terhorst,
I get two very different results in this step, how should I choose.
#189
①smc++ split -o split/ pop1/model.final.json pop2/model.final.json data/pop1_pop2*.smc.gz
②smc++ split -o split/ pop2/model.final.json pop1/model.final.json data/pop2_pop1.*.smc.gz
the result of ①
the result of ②
And the divergence times inferred from .model.iter0.json are older than those from model.final.json, and appear to better reflect the evolutionary scenario we anticipate. Would it be justified to adopt .model.iter0.json for downstream analysis?