Welcome

💦About

The Priciple of GSClassifier is a book for users of the R package GSClassifier who want to know the most details. If you’re looking for the PDF edition, you can find it here.

GSClassifier is an R-based comprehensive classification tool for subtypes modeling and personalized calling based on pure transcriptomics. It could be used for precision medicine, such as cancer diagnosis. The inspiration for GSClassifier comes from ImmuneSubtypeClassifier, an R package for classification of PanCancer immune subtypes based on the work of Gibbs et al [1,2].

Lots of surprising features in GSClassifier are as follows:

Optimized for just one sample
Available for modeling and calling of brand-new GEPs-based subtypes in any diseases (cancers)
No limitation of the number of gene signatures(≥1) or subtypes(≥2)
Insensitive normalization due to the use of the individual gene rank matrix
More ensemble and repeatable modeling process
More optimizations in the parallel computing
New useful functions as supplements

ATTENTION! In the future, there might be third-party contributors in GSClassifier platform, with some useful models for specific usages. If you use models provided by these people, you had better know more details as possible, including designs, data sources, destinations, training scripts, and limitations of models, especially those from studies under peer review.

💿License

GSClassifier is released under the Apache-2.0 license. See LICENSE for details.
The technical documentation, as a whole, is licensed under a Creative Commons Attribution- NonCommercial-ShareAlike 4.0 International License. The code contained in this book is simultaneously available under the MIT license; this means that you are free to use it in your packages, as long as you cite the source.

👍Installation

RStudio/Posit is one of the best Integrated Development Environments (IDE) in R programming. If you’re struggling in R-GUI, it is recommended to turn to RStudio/Posit.

For installation of GSClassifier, please run these commands in an R environment:

# Install "devtools" package
if (!requireNamespace("devtools", quietly = TRUE))
    install.packages("devtools")

# Install dependencies
if (!requireNamespace("luckyBase", quietly = TRUE))
    devtools::install_github("huangwb8/luckyBase")

# Install the "GSClassifier" package
if (!requireNamespace("GSClassifier", quietly = TRUE))
    devtools::install_github("huangwb8/GSClassifier")

In the future, a stable GSClassifier version might be sent to CRAN. Still beta.

👀Mirror

For some special countries or regions, users could also try:

# Install dependencies
install.packages("https://gitee.com/huangwb8/luckyBase/repository/archive/Primary?format=tar.gz", repos=NULL, method="libcurl")

# Install the "GSClassifier" package
install.packages("https://gitee.com/huangwb8/GSClassifier/repository/archive/Primary?format=tar.gz", repos=NULL, method="libcurl")

📚Change log

Version 0.1.27
- Enhaned geneMatch function
- Repair some bugs
Version 0.1.9
- Optimize function verbose
- Optimize for a routine scenario: one gene set and two subtypes
- Optimize the strategy of automatic parameters selection for modeling training with R package caret
- Interact with external models from the luckyModel package
Version 0.1.8
- Primary public version of GSClassifier
- Apache License, Version 2.0
- Friendly wiki-based tutorial
- Platform for developers

📆TODO

More medical fields included, such as in the pan-cancer utility
Advanced methods (such as artificial intelligence) for enhanced robustness
Unsupervised learning for de-novo classification based on intrinsic frames of omics instead of human knowledge
Multi-omics exploration and support
More friendly characteristics for developers and contributors
Web application for newbies to R programming

🌴Other Projects

References

1. Thorsson V, Gibbs DL, Brown SD, et al. The immune landscape of cancer. Immunity 2018; 48:812–830 e14

2. Gibbs DL. Robust classification of immune subtypes in cancer. 2020;

The Principle of R Package GSClassifier