In Dainanahan/drugverse: Drug Verse R Ecosystem

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

drugverse

Introduction

drugverse provides pharmaceutical researchers with different tools that mimic the data science life cycle and helps them to achieve their missions in solving challenges pertaining to drug development.

Following is what drugverse would be capable of doing upon its completion:

• wrangle data obtained from different drugs-related data sources and formats into forms that are more analysis-ready under R (i.e. dbparser),
• explore and mine drugs-related data in Shiny applications containing user-friendly interfaces and interactive visualizations (i.e. dvminer),
• run prediction algorithms (e.g. similarity based algorithms, feature based algorithms, semantic based algorithms, etc) for the prediction of different associations (e.g. drug-target interactions, drug-disease associations, etc.) (i.e. dvsem)
• validate prediction results both quantitatively (via latest known statistical validation methods) and qualitatively (e.g. viewing highest-ranked predictions and highlighting those that are already known in the literature) in intuitive and easy-to-understand reports (i.e. dvpred).

Packages & Tools

The following are the packages and tools that are working under drugverse:

dbparser

The main purpose of dbparser, DrugBank Parser, is to parse the DrugBank database which is downloadable in XML format into more than 75 R dataframes. These dataframes can then be explored and analyzed as desired by the user. This package further provides the facility of saving the parsed data into a given database (e.g. MySQL).

Shiny application for that package is under development

dvminer (under development)

dvminer is a complete R Shiny application package that can be used to do the following:

• Displaying an interactive drug-target network.
• Augmenting the visualized network with drug-enzyme, drug-transporter and drug-carrier relations.
• Selecting certain drugs to perform specific tasks on them using one of the other tools that are available in drugverse such as the Drug-Target Interaction Predictor.

All services behind that shiny application can be used separately without calling the shiny app.

dvpred (under development)

dvpred is a package that provides implementations of different algorithms for predicting drug-target interactions. These algorithms can be categorized as follows:

• Similarity-based methods
• Bipartite local models
• Matrix factorization methods
• Feature-based methods
• Network-based methods

The package will also contain a shiny application that can be used separately without calling the shiny app.

dvsem (under development)

dvsem is a package that:

• search and extract pharmacological information of drugs and biomedical documents using NLP and text mining techniques from different online resources to find implicit co-occurrent compound–protein relations,
• build its own knowledge base from the extracted information,
• answer queries about drugs and targets based on the knowledge parsed from the different data sources (e.g. article abstracts, online biological databases, RDF databases, etc.)

A shiny application, called Byakugan, that wrap the above functionalities in addition to interactive visualization.

Installation

You can install the released version of drugverse from CRAN with:

install.packages("drugverse")

And the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("Dainanahan/drugverse")

Example

This is a basic example which shows you how to solve a common problem:

library(drugverse)
## basic example code

Code of Conduct

Please note that the 'drugverse' project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

Contributing Guide

👍🎉 First off, thanks for taking the time to contribute! 🎉👍 Please review our Contributing Guide.

Share the love ❤️

Think drugverse is useful? Let others discover it, by telling them in person, via Twitter or a blog post.

Using drugverse for a paper you are writing? Consider citing it

citation("drugverse")

Dainanahan/drugverse documentation built on Feb. 19, 2020, 5:24 p.m.