search

matsengrp / ecgtheow

Ancestral lineage reconstruction using BEAST or RevBayes
2 stars 2 forks source link

Amrit dev #18

Closed dunleavy005 closed 6 years ago

dunleavy005 commented 6 years ago

I've created the scons interface for ecgtheow based on our chalkboard diagram (https://matsengrp.slack.com/files/U031DHRSB/F9XB1H2CE/img_20180326_180054.jpg). It divides the pipeline into 3 steps: 1) input data creation, 2) inference, 3) post-processing.

1) incorporates both Kristian's simulator and available CFT data. 2) incorporates BEAST/RevBayes, Amrit/Not-Amrit corrected inferences. 3) always makes graphviz summary plots of the posterior trees, but also computes Chris's posterior accuracy statistics in process_beast.py in the case of simulated data.

metasoarous commented 6 years ago

@dunleavy005 Great! Excited to look through this! For starters though, would you please merge in master so that the ultimate/full changeset will be easier to see?

matsen commented 6 years ago

@dunleavy005 I'm pretty sure he means rebase on master!

metasoarous commented 6 years ago

Oy. Looks like we bungled this. I realize now that I didn't phrase that clearly. I meant merge master into amrit_dev, not merge amritdev into master --

metasoarous commented 6 years ago

@dunleavy005 I just reverted the merge. Would you please try to merge master into amrit_dev and create a new PR so that I can review? Sorry again for the confusion...