find genotypes that are repeatead in selected environments

get_genotypes.R

pacman::p_load(tidyverse)

# Fake data
full_grid <- expand_grid(
  gen = paste0("G", 1:3),
  loc = paste0("L", 1:3),
  season = paste0("S", 1:4)
) %>% 
  mutate(env = paste(loc, season))
replications(~ gen * env, full_grid)

# Generate inbalance
red_grid <- sample_frac(full_grid, size = 0.8)
replications(~ gen * env, red_grid)

# Get available seasons
seasons <- red_grid %>% 
  distinct(season) %>% 
  pull %>% 
  sort
seasons

# Get last Nth seasons
last_nseasons <- tail(seasons, 1)
last_nseasons

# Get distinct environments within selected seasons
envs <- red_grid %>% 
    filter(season %in% last_nseasons) %>% 
    distinct(env) %>% 
    pull
envs 

# Get genotypes that were present in all of those the environments
red_grid %>% 
  filter(env %in% envs) %>% 
  count(gen) %>% 
  filter(n == n_distinct(envs)) %>% 
  distinct(gen) %>% 
  pull