Introduction

This package can be used to query the newest version (e.g. 5.1.3) of DrugBank database in R. It can transform the original DrugBank database in xml file format, which is available here, into a dataframe and store the dataframe into a SQLite database. The generated DrugBank SQLite database in specific version can be queried by using queryDB function. We can

• Get the entire DrugBank dataframe

• Get all the DrugBank ids

• Determine whether the query drugs are FDA approved

• Get targets of the query drugs.

Installation and load

The drugbankR package can be directly installed from github

# install.packages("devtools")
devtools::install_github("yduan004/drugbankR")

load the package into R

library(drugbankR)

Generate your own DrugBank SQLite database

Download and unzip the full DrugBank database (xml file) from here to your present working directory of R session, rename the unziped xml file as drugbank.xml. The dbxml2df function will read in the xml file and transform it into a data.frame object in R. This process may take about 20 minutes. Argument version is a character indicating the version of the downloaded DrugBank database. Since now, the newest version is 5.1.3. Note, you need to creat a DrugBank account and log in to access the data.

drugbank_dataframe <- dbxml2df(xmlfile="drugbank.xml", version="5.1.3") 

The df2SQLite function stores the drugbank dataframe into a SQLite database. The created SQLite database (drugbank_versionNumber.db) is under your present working direcotry of R, you could check your current R working directory by running getwd() in R.

df2SQLite(dbdf=drugbank_dataframe, version="5.1.3")

Query the DrugBank SQLite database generated above

# get the entire drugbank data.frame
all <- queryDB(type = "getAll", db_path="drugbank_5.1.3.db") 
dim(all)

# retrieve all the valid drugbank ids
ids <- queryDB(type = "getIDs", db_path="drugbank_5.1.3.db") 
ids[1:4]

# given drugbank ids, determine whether they are FDA approved
queryDB(ids = c("DB00001","DB00002"),type = "whichFDA", db_path="drugbank_5.1.3.db") 

# given drugbank ids, get their target gene/protein IDs 
queryDB(ids = c("DB00001","DB00002"),type = "getTargets", db_path="drugbank_5.1.3.db")