norla_ME3_yell_convencional_2024.qmd

---
title: "me3_yellow_high_tier"
format:
  html:
    toc: true
    toc_float: true
    embed-resources: true
    self-contained-math: true
editor_options: 
  chunk_output_type: console
---
```{r packs, include=FALSE}
knitr::opts_chunk$set(echo = FALSE, include=TRUE, warning = FALSE, message = FALSE, fig.width = 5)
pacman::p_load(tidyverse, rio, knitr, ggrepel, pheatmap, RColorBrewer)
theme_set(theme_bw(base_family = 14)) 
ReScale <- function(x,first,last){(last-first)/(max(x)-min(x))*(x-min(x))+first}
```

("LDSR_CERCZM", "LDSR_CURVSP", "LDSR_DIPDMA", "LDSRP_PHYRM", "LDSR_PUCCSO", "LDSR_SETOTU", "LDSR_TARSC", "YLD")

```{r}
ye_conv <- import("data/me3_yellow_high_tier.csv")

ye_conv %>% count(commercialName)

ye_conv %>% 
  count(FIELD_plantingSeason, trait) %>%  
  ggplot() + 
  facet_wrap("trait") + 
  aes(round(FIELD_plantingSeason), n, group=1) + 
  geom_point() + geom_line() + 
  scale_x_continuous(breaks=scales::pretty_breaks(3)) +
  theme(axis.text.x = element_text(angle=30,hjust=1,vjust=1.0)) +
  guides(col="none")   + 
  labs(title="M3 - yellow high tier")
```

### Evol press

```{r}
ye_conv %>%  
  filter(!trait == "YLD") %>% 
  ggplot() + 
  facet_wrap("trait", scales="free") +
  aes(x=FIELD_plantingSeason, y=OBS_numValue, col=FIELD_Country) +
  geom_jitter(width=.2, alpha=.2) + 
  geom_boxplot(aes(group=interaction(FIELD_Country, FIELD_plantingSeason)), alpha=.2, width=.2) + 
  geom_smooth(alpha=.1) + 
  scale_x_continuous(breaks=scales::pretty_breaks(3)) + 
  scale_y_continuous(breaks=scales::pretty_breaks(6)) +
  # stat_summary(fun = "mean", colour = "red", size = 5,
  #              geom = "text", fontface= "bold", hjust=1,
  #              aes(label = round(after_stat(y),1))) +
  labs(title="M3 - yellow high tier")
```


### LDSR_TARSC

```{r}
ye_conv %>% 
  filter(trait == "LDSR_TARSC") %>% 
  group_by(FIELD_name) %>% 
  summarise(LDSR_TARSC=quantile(OBS_numValue, 0.95)) %>% 
  filter(LDSR_TARSC > 3) %>% 
  pull(FIELD_name) -> TARSC_keep

ye_conv %>% 
  filter(FIELD_name %in% TARSC_keep) %>% 
  filter(trait == "LDSR_TARSC") %>% 
  select(FIELD_name, commercialName, OBS_numValue) -> TARSC_full 
  # pivot_wider(names_from = commercialName, 
  #             values_from = OBS_numValue) %>% 
  # drop_na(-FIELD_name) %>% 
  # pivot_longer(-FIELD_name) -> TARSC_full

TARSC_full %>% 
  ggplot(aes(x = FIELD_name, y = commercialName)) + 
  geom_tile(colour="black", aes(fill = OBS_numValue)) +
  scale_fill_viridis_c() +
  theme(axis.text.x = element_text(angle=30,hjust=1,vjust=1.0))
```

### LDSR_SETOTU

```{r}
ye_conv %>%  
  filter(trait == "LDSR_SETOTU") %>% 
  ggplot() + 
  facet_wrap("commercialName") + 
  aes(x=FIELD_plantingSeason, y=OBS_numValue) +
  geom_jitter(width=.2, alpha=.2) + 
  stat_summary(fun.y=median, colour="red", geom="line", aes(group = 1)) + 
  scale_x_continuous(breaks=scales::pretty_breaks(3)) + 
  scale_y_continuous(breaks=scales::pretty_breaks(6)) +
  labs(title = "LDSR_SETOTU - M3 / yellow biotech")
```

```{r}
wht %>% 
  filter(trait == "LDSR_PUCCSO") %>% 
  ggplot(aes(x = FIELD_name, y = commercialName)) + 
  geom_tile(colour="black", aes(fill = OBS_numValue)) +
  scale_fill_viridis_c() +
  theme(axis.text.x = element_text(angle=30,hjust=1,vjust=1.0))

wht %>% 
  filter(trait == "LDSR_PUCCSO") %>% 
  group_by(year_loc) %>% 
  summarise(SDP_FUSASP=quantile(OBS_numValue, 0.9)) %>% 
  filter(SDP_FUSASP > 2) %>% 
  pull(year_loc) -> PUCCSO_keep

wht %>% 
  filter(year_loc %in% PUCCSO_keep) %>% 
  filter(trait == "LDSR_PUCCSO") %>% 
  select(year_loc, commercialName, OBS_numValue) %>% 
  pivot_wider(names_from = commercialName,
              values_from = OBS_numValue, 
              values_fn = mean) %>%
  drop_na(-year_loc) %>%
  pivot_longer(-year_loc, names_to = "commercialName", values_to ="OBS_numValue") -> PUCCSO

PUCCSO %>% 
  ggplot(aes(x = year_loc , y = commercialName )) + 
  geom_tile(colour="black", aes(fill = OBS_numValue)) +
  scale_fill_viridis_c() +
  theme(axis.text.x = element_text(angle=30,hjust=1,vjust=1.0))
```



```{r}
wht %>% 
  filter(!trait == "YLD") %>% 
  group_by(commercialName, trait)%>%
  summarise(score= mean(OBS_numValue, na.rm = TRUE)) %>%
  pivot_wider(names_from=trait, values_from = score) %>% 
  column_to_rownames("commercialName") %>% 
  mutate_if(is.numeric, ~ReScale(., 1, 5), na.rm = TRUE) %>% 
  pheatmap(
    cutree_rows = 3, cluster_col = FALSE, 
    # display_numbers = TRUE,  number_format = "%.1f", 
    cellwidth = 20, cellheight = 12, 
    angle_col = "45" ,
    col = rev(brewer.pal(6, 'RdYlGn')),
    legend_breaks = c(0, 1, 3, 5), # legend customisation
    legend_labels = c("Ausente", "Baja sev", "Media", "Alta sev"), 
    main = "ME1 / White high-tier" 
  ) 
```

### LDSR_PUCCSO

```{r}
dat %>%  
  filter(trait == "LDSR_PUCCSO") %>% 
  # filter(commercialName %in% c("CALAMAR", "HIPOPOTAMO", "DK4050", "MX8639", "P3270W")) %>%
  ggplot() + 
  facet_wrap("commercialName") + 
  aes(x=FIELD_plantingSeason, y=OBS_numValue) +
  # aes(x=FIELD_plantingSeason, y=OBS_numValue) + 
  geom_jitter(width=.2, alpha=.2) + 
  stat_summary(fun.y=median, colour="red", geom="line", aes(group = 1)) + 
  scale_x_continuous(breaks=scales::pretty_breaks(3)) + 
  scale_y_continuous(breaks=scales::pretty_breaks(6)) +
  # stat_summary(fun = "median", colour = "white", size = 3) + 
  # stat_summary(fun = "median", colour = "red", size = 5,
  #              geom = "text", 
  #              aes(label = paste("~", after_stat(y)))) + 
  labs(title = "LDSR_PUCCSO - M3 / yellow biotech")
```