search

petrelharp / treestats_ms

1 stars 1 forks source link

No tsinfer #24

Closed petrelharp closed 5 years ago

petrelharp commented 5 years ago

Ok, here's a version without time decomposition or tsinfer trees. Probably you should edit this version, @molpopgen ?

Also note the additional paragraph I put in the discussion.

molpopgen commented 5 years ago

Looks good. I like the bit re:ascertainment

molpopgen commented 5 years ago

Related, but so hard to think about: imputation! __ Kevin Thornton Associate Professor Ecology and Evolutionary Biology UC Irvine www.molpopgen.org github.com/molpopgen github.com/ThorntonLab

On Fri, Sep 13, 2019 at 9:43 AM Peter Ralph notifications@github.com wrote:

@petrelharp commented on this pull request.

In treestats_paper.tex https://github.com/petrelharp/treestats_ms/pull/24#discussion_r324275727 :

@@ -1458,6 +1354,15 @@ \section*{Discussion} calculation of branch statistics would provide a summary method that is not affected by this potentially important confounding factor.

+Many genomic datasets only contain genotype information at a subset of sites

yes!!! realized that on the bike ride in this morning, talking to andy.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/petrelharp/treestats_ms/pull/24?email_source=notifications&email_token=ABQ6OH7UCHFA6A66P3A6IRDQJO7F3A5CNFSM4IWR3Y42YY3PNVWWK3TUL52HS4DFWFIHK3DMKJSXC5LFON2FEZLWNFSXPKTDN5WW2ZLOORPWSZGOCEWIABA#discussion_r324275727, or mute the thread https://github.com/notifications/unsubscribe-auth/ABQ6OH7TXTB2IVJMIVIYK7TQJO7F3ANCNFSM4IWR3Y4Q .