pharmOncoX.Rmd

---
title: "Getting started"
output: 
  rmarkdown::html_vignette:
    toc: true
vignette: >
  %\VignetteIndexEntry{Getting started}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
bibliography: '`r system.file("bibliography.bib", package = "pharmOncoX")`'
nocite: |
  @Kim2021-ye, @Freshour2021-bh, @Sioutos2007-nl, 
  @Ochoa2021-jc, @Nakken2023-ab, @Griffith2017-do,
  @Tamborero2018-aj
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
```

<br><br>

# Installation

```{r install, echo = T, eval = T}

if (!("remotes" %in% installed.packages())) {
 install.packages("remotes")
}

remotes::install_github('sigven/pharmOncoX')

library(pharmOncoX)
cache_dir <- tempdir()
#cache_dir <- 
#  "/Users/sigven/project_data/packages/package__pharmOncoX/pharmOncoX/data-raw/gd_local"

```

<br><br>

# Cancer drug classifications

## Targeted agents

Plotted below are key statistics with respect to drug classification
numbers for targeted and chemotherapy agents found in _pharmOncoX_. Existing drug classifications have been retrieved from [ATC](https://www.whocc.no/), and these have been extended with manual addition/curation, and also through the establishment of multiple novel levels in the ATC tree, particularly for targeted therapies. Note that only drugs that are indicated for cancer conditions (as harvested from the Open Targets platform) are considered in the numbers plotted below.

<br>

```{r targeted_agents, echo = F, eval = T, results="hide"}

drugs <- get_drugs(
  cache_dir = cache_dir,
  treatment_category =  c("targeted_therapy_classified",
                          "hormone_therapy_classified"),
  output_resolution = "drug",
	drug_action_inhibition = T, drug_source_opentargets = T)

drug_classifications <- 
  drugs$records |>
  dplyr::select(drug_name, atc_level3) |>
  dplyr::mutate(
    atc_level3 = stringr::str_replace_all(
      atc_level3, 
      "\\|(Antineovascularisation agents|Selective immunosuppressants)",
      ""
    )
  ) |>
  dplyr::group_by(atc_level3) |>
  dplyr::summarise(n = dplyr::n()) |>
  dplyr::mutate(pct = round(n/sum(n) * 100, digits = 2)) |>
  dplyr::arrange(dplyr::desc(pct)) |>
  dplyr::filter(n >= 2)

p_targeted_classifications <- ggplot2::ggplot(
  data = drug_classifications, ggplot2::aes(x = reorder(atc_level3, pct), y = pct)) + 
  ggplot2::geom_bar(stat = "identity",fill="steelblue") +
  ggplot2::theme_classic() + 
  ggplot2::coord_flip() + 
  ggplot2::ylab("Percent of targeted agents") + 
  ggplot2::xlab("") +
  ggplot2::scale_y_continuous(breaks=seq(0,12,2)) +
  ggplot2::geom_text(
    ggplot2::aes(label=n), hjust=1.4, vjust=0.50, color="white", size=4) +
  ggplot2::theme(
    legend.title = ggplot2::element_blank(),
    axis.text.x = ggplot2::element_text(size = 14, vjust = 0.5),
    legend.text = ggplot2::element_text(family = "Helvetica", size = 14),
    axis.text.y = ggplot2::element_text(family = "Helvetica", size = 13),
    axis.title.x = ggplot2::element_text(family = "Helvetica", size = 14),
    axis.title.y = ggplot2::element_text(family = "Helvetica", size = 14)
  )


```

```{r p_targeted, fig.width=9, fig.height=8, results="asis"}
p_targeted_classifications
```

## Chemotherapy agents

```{r chemo_agents, echo = F, eval = T, results="hide"}

drugs <- get_drugs(
  cache_dir = cache_dir,
  treatment_category =  c("chemo_therapy_classified"),
  output_resolution = "drug",drug_classified_cancer = T,
	drug_source_opentargets = T)

drug_classifications_chemo <- 
  drugs$records |>
  dplyr::select(drug_name, atc_level3) |>
  dplyr::group_by(atc_level3) |>
  dplyr::summarise(n = dplyr::n()) |>
  dplyr::mutate(pct = round(n/sum(n) * 100, digits = 2)) |>
  dplyr::arrange(dplyr::desc(pct))

p_chemo_classifications <- ggplot2::ggplot(
  data = drug_classifications_chemo, ggplot2::aes(
    x = reorder(atc_level3, pct), y = pct)) + 
  ggplot2::geom_bar(stat = "identity",fill="#E69F00") +
  ggplot2::theme_classic() + 
  ggplot2::coord_flip() + 
  ggplot2::ylab("Percent of chemotherapy agents") + 
  ggplot2::xlab("") +
  ggplot2::scale_y_continuous(breaks=seq(0,30,5)) +
  ggplot2::geom_text(ggplot2::aes(label=n), 
                     hjust=1.4, vjust=0.50, color="white", size=4) +
  ggplot2::theme(
    legend.title = ggplot2::element_blank(),
    axis.text.x = ggplot2::element_text(size = 14, vjust = 0.5),
    legend.text = ggplot2::element_text(family = "Helvetica", size = 14),
    axis.text.y = ggplot2::element_text(family = "Helvetica", size = 13),
    axis.title.x = ggplot2::element_text(family = "Helvetica", size = 14),
    axis.title.y = ggplot2::element_text(family = "Helvetica", size = 14)
  )


```

```{r chemo_plot, fig.width=9, fig.height=6, results="asis"}
p_chemo_classifications
```



# Retrieval of drugs - examples 

## Get BRAF-targeted drugs, list records per indication


```{r braf_drugs_full, echo = T}

drugs <- get_drugs(
  cache_dir = cache_dir,
  treatment_category =  c("targeted_therapy_classified",
                          "targeted_therapy_unclassified"),
	drug_target = c('BRAF'))

## Number of drug records
nrow(drugs$records)

## Column names of drug records
colnames(drugs$records)


```

<br><br>

## Get RAS-targeted drugs, list per drug target

```{r ras_inhibitors, echo = T, eval = T, results = "hide"}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  treatment_category =  c("targeted_therapy_classified"),
  drug_action_inhibition = T,
  output_resolution = "drug2target")$records |>
  dplyr::filter(atc_level3 == "RAS inhibitors")

drugs <- drugs |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    )

dt_drugtable_ras_inhibitors <- DT::datatable(
  drugs,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)

```

<br>

```{r dt_ras_inhibitors, echo = F, eval = T}
dt_drugtable_ras_inhibitors
```

<br><br>

## Get MEK inhibitors, list per drug only

```{r mek_inhibitors, echo = T, eval = T, results = "hide"}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  treatment_category =  c("targeted_therapy_classified"),
  drug_action_inhibition = T,
  drug_source_opentargets = T, 
  output_resolution = "drug" )$records |>
  dplyr::filter(atc_level3 == "MEK inhibitors")

drugs <- drugs |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    )

dt_drugtable_mek_inhibitors <- DT::datatable(
  drugs,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)

```

<br>

```{r dt_mek_inhibitors, echo = F, eval = T}
dt_drugtable_mek_inhibitors
```

<br><br>

## Get immune checkpoint inhibitors, list per drug target

```{r ici, echo = T, eval = T, results = "hide"}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  treatment_category =  c("targeted_therapy_classified"),
  drug_source_opentargets = F,
  drug_classified_cancer = T,
   output_resolution = "drug2target")

drugs$records <- drugs$records |>
  dplyr::filter(
    (!is.na(atc_level3) & 
      (atc_level3 == "PD-1/PDL-1 inhibitors" |
     atc_level3 == "Other immune checkpoint inhibitors")
  )) |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    ) |>
  dplyr::select(
    drug_id,
    drug_name,
    drug_type,
    target_symbol,
    target_genename,
    dplyr::everything()
  )

dt_drugtable_ici <- DT::datatable(
  drugs$records,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)
```

<br>

```{r dt_ici, echo = F, eval = T}
dt_drugtable_ici
```

<br><br>

## Get immune checkpoint inhibitors indicated for lung cancer conditions, list per drug-target entry

```{r ici_lung, echo = T, eval = T, results = "hide"}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  output_resolution = "drug2target", 
  treatment_category = c("targeted_therapy_classified"),
  drug_source_opentargets = T,
  drug_indication_main = "Lung")

drugs$records <- drugs$records |>
  dplyr::filter(
    (!is.na(atc_level3) & 
      (atc_level3 == "PD-1/PDL-1 inhibitors" |
     atc_level3 == "Other immune checkpoint inhibitors")
  )) |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    ) |>
  dplyr::select(
    drug_id,
    drug_name,
    drug_type,
    target_symbol,
    target_genename,
    dplyr::everything()
  )

dt_drugtable_ici_lung <- DT::datatable(
  drugs$records,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)

```

<br>

```{r dt_ici_lung, echo = F, eval = T}
dt_drugtable_ici_lung
```

<br><br>

## Get antimetabolite drugs

```{r antimetabolites, echo = T}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  treatment_category = c("chemo_therapy_classified"),
  output_resolution = "drug")

drugs$records <- drugs$records |>
  dplyr::filter(
    !is.na(atc_level2) &
    stringr::str_detect(
      atc_level2, "ANTIMETABOLITES"
    )
  ) |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    )

dt_drugtable_metabolites <- DT::datatable(
  drugs$records,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)

```

<br>

```{r dt_metabolites, echo = F, eval = T}
dt_drugtable_metabolites
```

<br><br>

## Get taxanes

```{r taxanes, echo = T}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  treatment_category = "chemo_therapy_classified",
  output_resolution = "drug")

drugs$records <- drugs$records |>
  dplyr::filter(
    stringr::str_detect(
      atc_level3, "Taxanes"
    )
  ) |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    )

dt_drugtable_taxanes <- DT::datatable(
  drugs$records,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)

```

<br>

```{r dt_taxanes, echo = F, eval = T}
dt_drugtable_taxanes
```

<br><br>

## Get platinum compounds

```{r platins, echo = T}

drugs <- get_drugs(
  cache_dir = cache_dir, 
  treatment_category = "chemo_therapy_classified",
  output_resolution = "drug")

drugs$records <- drugs$records |>
  dplyr::filter(
    stringr::str_detect(
      atc_level3, "Platinum compounds"
    )
  ) |>
  dplyr::select(
    -c("drug_alias",
       "disease_main_group",
       "drug_clinical_id")) |>
  dplyr::mutate(
    disease_indication = stringr::str_replace_all(
      disease_indication, "\\|",", ")
    )

dt_drugtable_platins <- DT::datatable(
  drugs$records,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)

```

<br>

```{r dt_platins, echo = F, eval = T}
dt_drugtable_platins
```

<br><br>

# Retrieval of biomarkers

## Reported associations between BRCA1/2 alterations and drug sensitivity

* Get evidence from [CIViC](https://civicdb.org) and [CGI](https://www.cancergenomeinterpreter.org/biomarkers) for cancer drug sensitivity of BRCA1/2 alterations (somatically (tumor) or inherited/germline)

```{r biomarkers, echo = T}

biomarkers <- get_biomarkers(
  cache_dir = cache_dir)

brca1_biomarkers <- list()
for(source in c('civic','cgi')){
  brca1_biomarkers[[source]] <- 
    biomarkers$data[[source]]$variant |> 
    dplyr::filter(
      !is.na(symbol) & (symbol == "BRCA1" | symbol == "BRCA2")) |>
    dplyr::group_by(variant_id, variant_name_primary, variant_consequence) |>
    dplyr::summarise(
      variant_alias = paste(variant_alias, collapse=", "), 
      .groups = "drop") |> 
    dplyr::inner_join(
      biomarkers$data[[source]]$clinical, by = "variant_id") |>
    dplyr::select(
      variant_id, variant_name_primary, therapeutic_context,
      evidence_type, evidence_level, 
      biomarker_source, biomarker_source_datestamp,
      molecular_profile_name, evidence_id, variant_origin,
      primary_site, evidence_id, source_id,
      evidence_url, 
      evidence_description, 
      clinical_significance) |> 
    dplyr::distinct() |> 
    dplyr::rename(literature_id = source_id,
                  variant_name = variant_name_primary) |>
    dplyr::filter(evidence_type == "Predictive") |>
    dplyr::select(
      variant_name, 
      primary_site,
      therapeutic_context,
      molecular_profile_name, 
      evidence_level,
      dplyr::everything()
    )
}

brca1_biomarkers_all <- 
  dplyr::bind_rows(brca1_biomarkers[['civic']],
                   brca1_biomarkers[['cgi']]) |>
    dplyr::arrange(evidence_level)
  

dt_brca1_biomarkers <- DT::datatable(
  brca1_biomarkers_all,
  escape = FALSE,
  extensions = c("Buttons", "Responsive"), 
  width = "100%",
  options = list(
    buttons = c("csv", "excel"), 
    dom = "Bfrtip")
)
```
<br>

```{r dt_brca_biomarkers, echo = F, eval = T}
dt_brca1_biomarkers
```
<br><br>

# Session Info

\vspace{5pt}
\footnotesize
```{r sessioninfo, eval = TRUE}
# set eval = FALSE if you don't want this info (useful for reproducibility) 
# to appear
sessionInfo()
```
\normalsize

<br><br>

# References