Merge changes from Rose's branch - Githubissues

rosemckeon / ploidy

How does disturbance on a landscape affect the establishment of new polyploid plant species?

3 stars 0 forks source link

Merge changes from Rose's branch #64

Closed rosemckeon closed 5 years ago

rosemckeon commented 5 years ago

Everything since our meeting with Mario:

Growth rate
Reduced genome size
Changes to data output so we see all life stages before survival and selection
Massive reproduction changes
- simplified pollen/ovule matching
- various mating rules that affect zygote survival
- variable pollen dispersal range
Improved plots in tests.R

rosemckeon commented 5 years ago

Hi @bradduthie I'm going through your replies this morning. I've pushed my output enhancements this morning too, so you will need to do another pull, but can set a file to save to now.

The main functions I have changed, that you might want to look over, are:

reproduce
get_pollen_donors (how I've decided which dads pollinate which mums)
create_zygotes (where all the mating rules are set by modifying fertilisation probabilities).
sample_genomes (with better control over the ploidy level output and proper numbering of alleles)

If I can, I might try and get a few of the bugs fixed and pushed so they can be included too.

bradduthie commented 5 years ago

@rozeykex Sounds good! I'll look over the functions that you mentioned and let you know if there are any edits to consider.

bradduthie commented 5 years ago

@rozeykex I think that the line defining pop$size in the grow function is causing plants of infinite population size? Did this with some quick debugging

bradduthie commented 5 years ago

Functions reproduce and get_pollen_donors look good.

bradduthie commented 5 years ago

The function create_zygotes works. Surprised to learn that:

t1 <- c(1, 0, 1, 0, 1, 1)
t2 <- c(TRUE, FALSE, TRUE, FALSE, TRUE, TRUE)

Are not evaluated by which in the same way.

bradduthie commented 5 years ago

In sample_genomes, is the explicit dplyr::select(mum, dad) necessary? dplyr should be read in, and you can require it as a dependency. Otherwise, this also looks good. Happy to approve merge if you're ready @rozeykex .

rosemckeon commented 5 years ago

In sample_genomes, is the explicit dplyr::select(mum, dad) necessary? dplyr should be read in, and you can require it as a dependency. Otherwise, this also looks good. Happy to approve merge if you're ready @rozeykex .

@bradduthie It is at the moment, because I made my natural selection function called select (d'oh!) - I'll make an issue to fix this once the important stuff is done.

rosemckeon commented 5 years ago

@rozeykex I think that the line defining pop$size in the grow function is causing plants of infinite population size? Did this with some quick debugging

Hmm, @bradduthie I think maybe the Inf growth rates were causing Inf sizes? Can you see if this is still a problem for you with the new fixes I've pushed? I'll have a go at running the sim with your stop edits merged in too.

rosemckeon commented 5 years ago

Ooops, didn't mean to dismiss the review! I'm going to merge anyway @bradduthie.