institutions.qmd

---
title: "Institutions"
format: html
params:
  refresh_investigator_data: true
  refresh_volumes_sharing_data: true
  save_volumes_sharing_status_info: true
  csv_path: "private"
---

## About

This page provides information about Databrary's institutional partners.

For more information about institutional roles and responsibilities, see the [Databrary Guide](https://databrary.github.io/guide).

```{r}
#| label: set-up
#| echo: FALSE
#| message: FALSE
#| warning: FALSE
knitr::opts_chunk$set(echo = FALSE,
                      message = FALSE,
                      warning = FALSE)

suppressPackageStartupMessages(library(databraryr))
suppressPackageStartupMessages(library(tidyverse))
library(ggplot2)

# Source functions
fl <- list.files(file.path(here::here(), "R"), "*.R", 
                 full.names = TRUE)
p <- purrr::map(fl, suppressPackageStartupMessages(source)) |>
  purrr::quietly()

# Login to Databrary
x <- databraryr::login_db(email = Sys.getenv("USERNAME"),
         password = Sys.getenv("PASSWORD"),
         client_id = Sys.getenv("CLIENT_ID"),
         client_secret = Sys.getenv("CLIENT_SECRET"),
         store = FALSE,
         overwrite = FALSE)
```

## Summary

```{r}
#| label: create-institution-affiliates-list

inst_affils_df <- make_inst_affil_df()

n_institutions <- length(unique(inst_affils_df$id))
```

Databrary has n=`{r} n_institutions` active institutional partners.

## Recent additions

```{r}
#| label: tbl-institutions-recent-additions
#| tbl-cap: "List of the 10 most recent additions to the network of Databrary institutions."
inst_affils_df |>
  dplyr::arrange(desc(id)) |>
  dplyr::select(name, institution_url) |>
  dplyr::mutate(name = stringr::str_sub(name, start = 1, end = 25)) |>
  head(n = 10) |>
  kableExtra::kbl() |>
  kableExtra::kable_material(full_width = F) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "hover")) |>
  kableExtra::scroll_box(width = "700px")
```
## Full list

::: {.callout-warning}
This is a work-in-progress.
:::

Here is a full list in alphabetical order.

```{r}
#| label: tbl-databrary-institutions
#| tbl-cap: "List of all Databrary institutions in alphabetical order."
inst_affils_df |>
  dplyr::select(name, institution_url) |>
  dplyr::arrange(name) |>
  dplyr::mutate(name = stringr::str_sub(name, start = 1, end = 25)) |>
  kableExtra::kbl() |>
  kableExtra::kable_material(full_width = F) |>
  kableExtra:: kable_styling(bootstrap_options = c("striped", "hover")) |>
  kableExtra:: scroll_box(width = "700px", height = "500px")
```

## Institutional participation

This section summarizes data about institutional participation in Databrary.

```{r}
#| label: import-inst-affils-data
inst_affils_df <- inst_affils_df |>
  dplyr::group_by(id, name) |>
  dplyr::summarise(n_ais = n())
```

```{r}
#| label: import-inst-shared-vols-data

inst_invest_shared_path <- file.path(params$csv_path, "inst_invest_shared")
inst_invest_shared_fl <- list.files(inst_invest_shared_path,
                                    pattern = ".csv$",
                                    full.names = TRUE)
inst_invest_shared <- purrr::map(inst_invest_shared_fl, readr::read_csv, show_col_types = FALSE) |>
  purrr::list_rbind()
```

```{r}
#| label: clean-summarize-inst-affils-data

inst_invest_shared <- inst_invest_shared |>
  dplyr::group_by(inst_id) |>
  dplyr::rename(id = inst_id) |>
  dplyr::summarise(
    n_ais_w_vols = n(),
    tot_n_vols = sum(n_vols),
    tot_n_public = sum(n_public),
    tot_n_overview = sum(n_overview)
  ) |>
  dplyr::arrange(id)

inst_summ_df <- dplyr::left_join(inst_affils_df, inst_invest_shared, by = c("id"))

inst_summ_df <- inst_summ_df |>
  dplyr::mutate(n_ais_w_vols = if_else(is.na(n_ais_w_vols), 0, n_ais_w_vols),
                tot_n_vols = if_else(is.na(tot_n_vols), 0, tot_n_vols),
                tot_n_public = if_else(is.na(tot_n_public), 0, tot_n_public),
                tot_n_overview = if_else(is.na(tot_n_overview), 0, tot_n_overview))
```

### Tabular summaries

::: {.panel-tabset}

## Authorized Investigators

```{r}
#| label: tbl-institutional-footprint-ais
#| tbl-cap: "Institutions with the largest number of Authorized Investigators (n_ais). Also reported are the percent of authorized investigators who own publicly visible volumes (p_vols)."
#| eval: true
#| message: false
#| warning: false

inst_invest <- inst_summ_df[,c(2:4)]
inst_invest |>
  # dplyr::select(name, n_ais, n_ais_w_vols) |>
  dplyr::filter(!is.na(n_ais), n_ais > 0) |>
  arrange(desc(n_ais)) |>
  dplyr::mutate(p_vols = round(100*n_ais_w_vols/n_ais, 1)) |>
  dplyr::select(name, n_ais, p_vols) |>
  kableExtra::kbl() |>
  kableExtra::kable_material(full_width = F) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "hover")) |>
  kableExtra::scroll_box(width = "700px")
```

## Shared volumes

```{r}
#| label: tbl-institutional-footprint-vols
#| tbl-cap: "Institutions with the largest number of shared volumes (n_vols). Also shown is the percent of volumes shared by investigators at an institution that are shared publicly."
#| eval: true
#| message: false
#| warning: false

inst_vols <- inst_summ_df[,c(2,5:7)]
inst_vols |>
  dplyr::rename(n_vols = tot_n_vols,
                n_public = tot_n_public,
                n_overview = tot_n_overview) |>
  dplyr::filter(!is.na(n_vols), n_vols > 0) |>
  arrange(desc(n_vols)) |>
  dplyr::mutate(p_public = round(100*n_public/n_vols, 1)) |>
  dplyr::select(name, n_vols, p_public) |>
  kableExtra::kbl() |>
  kableExtra::kable_material(full_width = F) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "hover")) |>
  kableExtra::scroll_box(width = "700px")
```

:::

### Histograms

::: {.panel-tabset}
## Authorized Investigators

```{r}
#| label: fig-n-ais-per-institution-histogram
#| fig-cap: "Histogram of the number of Authorized Investigators per institution."
inst_summ_df |>
  dplyr::select(n_ais) |>
  dplyr::filter(n_ais > 0) |>
  ggplot() +
  geom_histogram(mapping = aes(n_ais), bins = 15) +
  xlab("Authorized Investigators at an institution") +
  ylab("Number of institutions") +
  theme_classic()
```

## Investigators with volumes

```{r}
#| label: fig-p-ais-per-institution-histogram
#| fig-cap: "Histogram of percent of Investigators at an Institution who have created publicly visible volumes."
inst_invest |>
  dplyr::mutate(p_vols = round(100*n_ais_w_vols/n_ais, 1)) |>
  dplyr::select(p_vols) |>
  ggplot() +
  geom_histogram(mapping = aes(p_vols), bins = 15) +
  xlab("Percent of Authorized Investigators") +
  ylab("Number of institutions") +
  theme_classic()
```

## Shared volumes

```{r}
#| label: fig-n-vols-per-institution-histogram
#| fig-cap: "Histogram of the number of shared volumes (datasets) per institution."

institutions_w_vols <- inst_summ_df |>
  dplyr::filter(tot_n_vols > 0)

institutions_w_vols |>
  dplyr::select(tot_n_vols) |>
  ggplot() +
  geom_histogram(mapping = aes(tot_n_vols), bins = 15) +
  xlab("Total volumes shared by investigators at an institution") +
  ylab("Number of institutions") +
  theme_classic()
```

```{r}
n_institutions_w_vols <- length(unique(institutions_w_vols$id))
p_institutions_shared_vols <- n_institutions_w_vols/n_institutions
```

N=`{r} n_institutions_w_vols` institutions have Authorized Investigators who have shared volumes. This is `{r} round(100*p_institutions_shared_vols, 2)`% of the total number of institutions.

:::

## Clean up

Logout from Databrary.

```{r}
#| label: clean-up-logout
x <- logout_db()
```