Skip to content

Commit

Permalink
Update AusTraits tutorial (#10)
Browse files Browse the repository at this point in the history 
* Update AusTraits tutorials to run with the version 5.0.0 of austraits.build
  • Loading branch information
@ehwenk
ehwenk authored Nov 20, 2023
1 parent 692cae5 commit 2a281be
Show file tree
Hide file tree
Showing 12 changed files with 538 additions and 399 deletions.
689 changes: 401 additions & 288 deletions AusTraits_tutorial.qmd
View file

Large diffs are not rendered by default.

157 changes: 86 additions & 71 deletions data/extra_functions.R
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,15 +4,15 @@ austraits_weighted_means <- function(austraits, traits) {

# any data that is a mean, median or raw, create a site mean
data_means <-
(austraits %>% join_locations)$traits %>%
filter(trait_name %in% traits) %>%
filter(value_type %in% c("mean", "raw", "median")) %>%
mutate(
(austraits %>% austraits::join_locations())$traits %>%
dplyr::filter(trait_name %in% traits) %>%
dplyr::filter(value_type %in% c("mean", "raw", "median")) %>%
dplyr::mutate(
replicates = 1,
log_value = ifelse(!is.na(value) & value > 0, log10(as.numeric(value)), NA)
) %>%
group_by(taxon_name, trait_name, dataset_id, location_id) %>%
summarise(
dplyr::group_by(taxon_name, trait_name, dataset_id, location_id) %>%
dplyr::summarise(
mean = mean(as.numeric(value)),
min = min(as.numeric(value)),
max = max(as.numeric(value)),
Expand All @@ -21,11 +21,11 @@ austraits_weighted_means <- function(austraits, traits) {
`longitude (deg)` = first(`longitude (deg)`),
location_name = first(location_name),
all_replicates = sum(replicates),
geom_mean = mean(log_value)
geom_mean = mean(log_value)
) %>%
ungroup() %>%
distinct(taxon_name, trait_name, dataset_id, location_id, mean, min, max, median, geom_mean, all_replicates, location_name, `latitude (deg)`, `longitude (deg)`) %>%
mutate(
dplyr::ungroup() %>%
dplyr::distinct(taxon_name, trait_name, dataset_id, location_id, mean, min, max, median, geom_mean, all_replicates, location_name, `latitude (deg)`, `longitude (deg)`) %>%
dplyr::mutate(
location_replicates = 1,
flora_replicates = 0,
geom_mean = 10^(geom_mean)
Expand All @@ -34,21 +34,21 @@ austraits_weighted_means <- function(austraits, traits) {
# any data that is a max or a min (range) and basically from a flora, create a mean value
flora_means <-
austraits$traits %>%
filter(trait_name %in% traits) %>%
filter(value_type %in% c("minimum", "maximum"), basis_of_record %in% c("preserved_specimen", "literature"))
dplyr::filter(trait_name %in% traits) %>%
dplyr::filter(value_type %in% c("minimum", "maximum"), basis_of_record %in% c("preserved_specimen", "literature"))

if (nrow(flora_means > 0)) {
flora_means <-
flora_means %>%
group_by(taxon_name, trait_name, dataset_id, observation_id, original_name) %>%
summarise(
dplyr::group_by(taxon_name, trait_name, dataset_id, observation_id, original_name) %>%
dplyr::summarise(
mean = mean(as.numeric(value)),
min = min(as.numeric(value)),
max = max(as.numeric(value))
) %>%
ungroup() %>%
distinct(taxon_name, trait_name, dataset_id, observation_id, original_name, mean, min, max) %>%
mutate(
dplyr::ungroup() %>%
dplyr::distinct(taxon_name, trait_name, dataset_id, observation_id, original_name, mean, min, max) %>%
dplyr::mutate(
location_replicates = 0,
flora_replicates = 1,
all_replicates = 1,
Expand All @@ -58,20 +58,20 @@ austraits_weighted_means <- function(austraits, traits) {

means <-
if (nrow(flora_means > 0)) {
bind_rows(
data_means %>% mutate(across(any_of(c("mean", "min", "max")), ~as.numeric(.x))),
flora_means %>% mutate(across(any_of(c("mean", "min", "max")), ~as.numeric(.x)))
dplyr::bind_rows(
data_means %>% dplyr::mutate(across(dplyr::any_of(c("mean", "min", "max")), ~as.numeric(.x))),
flora_means %>% dplyr::mutate(across(dplyr::any_of(c("mean", "min", "max")), ~as.numeric(.x)))
)
} else {
data_means %>% mutate(across(any_of(c("mean", "min", "max")), ~as.numeric(.x)))
data_means %>% dplyr::mutate(across(dplyr::any_of(c("mean", "min", "max")), ~as.numeric(.x)))
}

means <- means %>%
mutate(
dplyr::mutate(
log_value = ifelse(!is.na(mean) & mean > 0, log10(as.numeric(mean)), NA)
) %>%
group_by(taxon_name, trait_name) %>%
summarise(
dplyr::group_by(taxon_name, trait_name) %>%
dplyr::summarise(
mean = mean(mean),
min = min(min),
max = max(max),
Expand All @@ -81,9 +81,9 @@ austraits_weighted_means <- function(austraits, traits) {
location_replicates = sum(location_replicates),
flora_replicates = sum(flora_replicates)
) %>%
ungroup() %>%
distinct() %>%
mutate(
dplyr::ungroup() %>%
dplyr::distinct() %>%
dplyr::mutate(
geom_mean = 10^(geom_mean)
)

Expand All @@ -94,33 +94,33 @@ categorical_summary <- function(austraits, trait_names) {

tmp <-
austraits$traits %>%
filter(trait_name %in% trait_names) %>%
select(dataset_id, taxon_name, trait_name, location_id, observation_id, value) %>%
mutate(value_tmp = stringr::str_split(value, " "))
dplyr::filter(trait_name %in% trait_names) %>%
dplyr::select(dplyr::all_of(c("dataset_id", "taxon_name", "trait_name", "location_id", "observation_id", "value"))) %>%
dplyr::mutate(value_tmp = stringr::str_split(value, " "))

tmp <-
tmp %>%
unnest_longer(value_tmp) %>%
mutate(
tidyr::unnest_longer(value_tmp) %>%
dplyr::mutate(
replicates = 1
) %>%
group_by(taxon_name, trait_name, value_tmp) %>%
dplyr::group_by(taxon_name, trait_name, value_tmp) %>%
summarise(
replicates = sum(replicates),
value_tmp = first(value_tmp)
) %>%
ungroup() %>%
distinct() %>%
mutate(
dplyr::ungroup() %>%
dplyr::distinct() %>%
dplyr::mutate(
tmp_summary = paste0(value_tmp, " (", replicates, ")")
) %>%
group_by(taxon_name, trait_name) %>%
mutate(
dplyr::group_by(taxon_name, trait_name) %>%
dplyr:: mutate(
value_summary = paste0(tmp_summary, collapse = "; ")
) %>%
ungroup() %>%
select(-value_tmp, - replicates, -tmp_summary) %>%
distinct()
dplyr::ungroup() %>%
dplyr::select(-dplyr::all_of(c("value_tmp", "replicates", "tmp_summary"))) %>%
dplyr::distinct()

tmp
}
Expand All @@ -129,23 +129,23 @@ categorical_summary_by_value <- function(austraits, trait_names) {

tmp <-
austraits$traits %>%
filter(trait_name %in% trait_names) %>%
select(dataset_id, taxon_name, trait_name, location_id, observation_id, value) %>%
mutate(value_tmp = stringr::str_split(value, " "))
dplyr::filter(trait_name %in% trait_names) %>%
dplyr::select(dplyr::all_of(c("dataset_id", "taxon_name", "trait_name", "location_id", "observation_id", "value"))) %>%
dplyr::mutate(value_tmp = stringr::str_split(value, " "))

tmp <-
tmp %>%
unnest_longer(value_tmp) %>%
mutate(
tidyr::unnest_longer(value_tmp) %>%
dplyr::mutate(
replicates = 1
) %>%
group_by(taxon_name, trait_name, value_tmp) %>%
summarise(
dplyr::group_by(taxon_name, trait_name, value_tmp) %>%
dplyr::summarise(
replicates = sum(replicates),
value_tmp = first(value_tmp)
) %>%
ungroup() %>%
distinct()
dplyr::ungroup() %>%
dplyr::distinct()

tmp
}
Expand All @@ -157,36 +157,48 @@ merge_entity_types <- function(dataset_id) {
austraits::extract_dataset(dataset_id))$traits

if(any(as.numeric(tmp$method_context_id) > 1) && !all(is.na(tmp$method_context_id))) {
tmp_columns <- tmp %>% filter(as.numeric(method_context_id) > 1) %>% distinct(trait_name)

tmp_columns <-
tmp %>%
dplyr::filter(as.numeric(method_context_id) > 1) %>%
dplyr::distinct(trait_name)

tmp <-
tmp %>%
mutate(trait_name = ifelse(trait_name %in% tmp_columns$trait_name,
paste(trait_name, "method_context_id", method_context_id, sep = "_"),
trait_name))
dplyr::mutate(trait_name = ifelse(trait_name %in% tmp_columns$trait_name,
paste(trait_name, "method_context_id", method_context_id, sep = "_"),
trait_name))
}

if(any(as.numeric(tmp$method_id) > 1) && !all(is.na(tmp$method_id))) {
tmp_columns <- tmp %>% filter(as.numeric(method_id) > 1) %>% distinct(trait_name)

tmp_columns <-
tmp %>%
dplyr::filter(as.numeric(method_id) > 1) %>%
dplyr::distinct(trait_name)

tmp <-
tmp %>%
mutate(trait_name = ifelse(trait_name %in% tmp_columns$trait_name,
paste(trait_name, "method_id", method_id, sep = "_"),
trait_name))
dplyr::mutate(trait_name = ifelse(trait_name %in% tmp_columns$trait_name,
paste(trait_name, "method_id", method_id, sep = "_"),
trait_name))
}

tmp <- tmp %>% dplyr::select(-dplyr::all_of(c("method_id", "method_context_id", "unit", "measurement_remarks", "replicates")))
tmp <-
tmp %>%
dplyr::select(-dplyr::all_of(c("method_id", "method_context_id", "unit", "measurement_remarks", "replicates")))

tmp <- tmp %>%
dplyr::mutate(counter = 1) %>%
group_by(trait_name, dataset_id, observation_id, repeat_measurements_id) %>%
dplyr::mutate(
counter = sum(counter),
value = ifelse(counter > 1, paste0(value, " (", value_type, ")"), value),
value = paste(value, collapse = "; ")
) %>%
ungroup() %>%
dplyr::select(-value_type) %>%
distinct()
tmp <-
tmp %>%
dplyr::group_by(trait_name, dataset_id, observation_id, repeat_measurements_id) %>%
dplyr::mutate(
count = length(trait_name),
value = ifelse(count > 1, paste0(value, " (", value_type, ")"), value),
value = paste(value, collapse = "; ")
) %>%
dplyr::ungroup() %>%
dplyr::select(-value_type) %>%
dplyr:: distinct()

for (i in c("source_id", "repeat_measurements_id", "plot_context_id", "treatment_context_id",
"temporal_context_id", "entity_context_id")) {
Expand All @@ -196,7 +208,10 @@ merge_entity_types <- function(dataset_id) {
}

if(length(unique(tmp$original_name)) == length(unique(tmp$taxon_name))) {
tmp <- tmp %>% dplyr::select(-original_name)

tmp <-
tmp %>%
dplyr::select(-dplyr::all_of(c("original_name")))
}

for (i in c("basis_of_record", "life_stage", "value_type")) {
Expand Down
28 changes: 15 additions & 13 deletions spatial_data_example.qmd
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,36 +4,38 @@
This tutorial is based on [this great tutorial](https://labs.ala.org.au/posts/2023-08-28_alternatives-to-box-plots/post.html) by Dax Kellie and Shandiya Balasubramaniam from the ALA team.


```{r, eval=FALSE}
# remotes::install_github("traitecoevo/austraits", ref = "changes-for-v5-austraits.build")
```{r, eval=TRUE, message = FALSE, warning = FALSE}
# remotes::install_github("traitecoevo/austraits")
library(tidyverse)
library(austraits)
```


Load austraits

```{r, eval=FALSE}
austraits <- read_rds("data/austraits.rds")
```
```{r, eval = TRUE}
most_recent <- austraits::get_versions()[[3]][1]
most_recent
austraits <- austraits::load_austraits(doi = most_recent)
```
Extract leaf mass per area (LMA) data

```{r, eval=FALSE}
```{r, eval = TRUE}
# You can use `lookup_trait()` to search for traits containing keywords
lookup_trait(austraits, "leaf_mass")
austraits::lookup_trait(austraits, "leaf_mass")
# Get trait data
leaf_mass <- austraits %>%
extract_trait("leaf_mass_per_area") %>%
pluck("traits") # Grab the traits table from the list of austraits tables
austraits::extract_trait("leaf_mass_per_area") %>%
purrr::pluck("traits") # Grab the traits table from the list of austraits tables
```


Filter to six species in the dataset

```{r, eval=FALSE}
```{r, eval = TRUE}
sample_names <- c("Cryptocarya rigida", "Pteridium esculentum",
"Eucalyptus baxteri", "Melaleuca armillaris",
"Eucalyptus wandoo", "Eucalyptus piperita")
Expand All @@ -44,7 +46,7 @@ leaf_mass_sample <- leaf_mass %>% filter(taxon_name %in% sample_names)

Plot raincloud plot of LMA for the six species

```{r, eval=FALSE}
```{r, eval = TRUE}
# install.packages(c("ggdist", "gghalves", "ggtext"))
# remotes::install_github("olihawkins/pilot")
ggplot(
Expand All @@ -65,7 +67,7 @@ ggplot(
size = 2.2) +
geom_boxplot(
aes(colour = taxon_name,
colour = after_scale(colorspace::darken(taxon_name, .7))),
colour = after_scale(colorspace::darken(colour, .7))),
width = .12, # Adjust box width
fill = NA,
size = 1.1, # Size of box line
Expand Down Expand Up @@ -93,7 +95,7 @@ ggplot(


Plot the species distributions of these six species with ALA occurrence data (using `galah`)
```{r, eval=FALSE}
```{r, eval = TRUE}
# install.packages(c("galah", "sf", "ozmaps"))
library(galah)
library(sf)
Expand Down
Loading

0 comments on commit 2a281be

Please sign in to comment.