Curate and standardize black cottonwood phenophases

[jessicaguo]

Follow up to #50

• [x] Meet with Wellington on Aug 2, document data description
• [x] Provide Wellington with list of files with missing dates and duplicate genotypes from site-year-block combinations
• [x] Decide how to standardize file names across 76 files
• [x] Update 01_explore.R to accomodate new naming conventions
• [ ] create figures comparing bud set and bud break between genotypes, plots, years,
• [x] Hillary Cooper brainstorming
• [ ] Provide Hillary with summary of data available, including efforts to model budburst or budset 3.5

[jessicaguo]

I explored the Placerville 2010 Bud Set data, which have date and 3 sets of sampling dates. However, even restricting to the most data-rich genotypes (with 9 points or more), only some span the range of the bud set levels (1-6).

I am not sure how to derive a response variable from these data. >1000 genotypes were planted at this site, but genotypes were only replicated 3 times (blocks). 474 genotypes were sampled 3 times, which yields a maximum of 9 data points from which to derive a phenotype. @dlebauer, any thoughts on how to proceed?

[KristinaRiemer]

We did talk the other day about one option, of binning all the responses into two categories, e.g., bud set values below 3 vs bud set values 3 and above. We just need some sort of justification for choosing what those categories are.

[jessicaguo]

For phenophase data, I feel we need some kind of date as a response variable. It's also not great that the plants were surveyed at most 3 times, because they clearly missed the transition period for some genotypes.

At our meeting yesterday, I asked Kevin to look into grouping the genotypes into haplotypes or populations. I also think I need to schedule a meeting with Anne about the data sooner rather than later.