---
title: "gDRtestData"
author: "gDR team"
output: BiocStyle::html_document
vignette: >
  %\VignetteIndexEntry{gDRtestData}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

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

```{r, echo=FALSE}
library(gDRtestData)
```

# Overview

The `gDRtestData` package is intended to store and generate example data that can be used through the `gDR` suite.

# Use Cases

## Synthetic data generation

Since gDR is a computational suite for drug response data from any experiment, a synthetic dataset is also needed for testing and exploration.

The basis of this package are two functions to generate the synthetic sets of cell lines and drugs.

```{r}
cell_lines <- create_synthetic_cell_lines()
drugs <- create_synthetic_drugs()
```

These base objects can be extended with additional information.

1. Replicates

```{r eval=FALSE}
cl_rep <- add_data_replicates(cell_lines)
head(cl_rep)
```

2. Concentration

```{r eval=FALSE}
cl_conc <- add_concentration(cell_lines)
head(cl_conc)
```

Or the user can do both with one function:

```{r eval=FALSE}
df_layout <- prepareData(cell_lines, drugs)
head(df_layout)
```

Additionally, the user may fill in the full response data with the day 0 part.

```{r eval=FALSE}
df_layout_small <- prepareData(cell_lines[seq_len(2), ], drugs[seq_len(4), ])
df_layout_small$Duration <- 72
df_layout_small$ReadoutValue <- 0
df_layout_small_with_Day0 <- add_day0_data(df_layout_small)
head(df_layout_small_with_Day0)
```

In a further step, the user may generate a set of synthetic results:

1. Hill coefficient
```{r}
hill <- generate_hill_coef(cell_lines, drugs) 
```

2. EC50 metric
```{r}
ec50_met <- generate_ec50(cell_lines, drugs) 
```

3. E inf metric
```{r}
einf_met <- generate_e_inf(cell_lines, drugs)
```

Or the user can create full response data with one function (for single-agent):

```{r eval=FALSE}
response_data <- prepareMergedData(cell_lines, drugs)
head(response_data)
```


**SUMMARY**

| Step | Function     | Output (`data.table`) |
| :--: | :----------- | :-------------------- |
|  0   | create_synthetic_cell_lines() | synthetic cell lines |
|  0   | create_synthetic_drugs()      | synthetic drugs |
|  1   | prepareData()                 | cell lines and drug merged with replicates and concentration |
|  2   | prepareMergedData()           | full response data for single-agent |
|  2   | prepareComboMergedData()      | full response data for combo |
|  2   | prepareCodilutionData ()      | full response data for co-dilution |

## Synthetic object of gDR data model

The gDR data model is built on the MultiAssayExperiments (MAE) structure. A detailed description of the gDR data model can be found in `gDRcore` package vignette.  

In `inst/testdata` the user may find a set of `qs` files that are examples of gDR data model for different data types. In the file `synthetic_list.yml` one can find a list of these datasets. Currently available are: 
```{r echo=FALSE}
yml_path <- system.file(package = "gDRtestData", "testdata", "synthetic_list.yml")
cat(paste0("* ", names(yaml::read_yaml(yml_path)), collapse = ", \n"), ".")
```

The script `generate_example_data.R` which shows how to generate and process above-mentioned datasets is in `inst/scripts` dir.
All key functions can be found in package `gDRcore` in script `generate_wrappers.R`. 

## Annotation data

In `inst/annotation_data` the user can find `CSV` files used in `gDRcore` for testing annotation functions.

## Other

Other files which were not mentioned above are used for testing gDR suite functionality.

# SessionInfo {-}

```{r sessionInfo}
sessionInfo()
```