IBC signature (Inflammatory Breast Cancer signature)

IBC signature is a set of genes which were determined to have high accuracy in discriminating IBC from non-IBC samples (pre-treatment core biopsy samples) in a random forest based model (see citation below for details).

Citing IBC signature

Zare, A., Postovit, LM. & Githaka, J.M. Robust inflammatory breast cancer gene signature using nonparametric random forest analysis. Breast Cancer Res 23, 92 (2021). https://doi.org/10.1186/s13058-021-01467-y

OPTION 1; IBC signature standalone GUI (no need to install MATLAB)

Reproduces the analysis reported in Zare et al. To score novel IBC/nonIBC dataset, see option 2 (scoring novel datasets in this GUI will be incorporated in the near future).

Supported Operating Systems

Windows

Has been successfully tested on Windows 7 and Windows 10 (see installation process below).

Mac

See option2 below.

Linux

See option2 below.

Installation

-Download 'IBCsignatureGUI.exe' and 'Data.tar.gz' file ('Data.tar.gz' contains data files used in Zare et al). If you don't have a software to uncompress '.tar' and '.gz' files, download 'unzip_untar.exe'.
-Install 'IBCsignatureGUI.exe' file as you would any other windows executable file. During the installation process, there will be a one time request to install MATLAB Compiler if not found in your system (MATLAB Runtime Version 9.5 for APPs compiled in MATLAB R2018b). After succesfully installing 'IBCsignatureGUI.exe', if need be, install 'unzip_untar.exe'.

Analysis

-Start the installed 'IBCsignatureGUI.exe' software (the graphical user interface (GUI) is shown below. Note, the cyan 'Initiate G59 analysis & random forest' button will be invisible and will only appear when data to be analyzed is selected using the knob).
-To uncompress 'Data.tar.gz', start 'unzip_untar.exe' GUI, click 'Uncompress .tar.gz file' and follow instructions.
-Click 'Data directory' button and select uncompressed Data directory/folder.
-Turn the knob to select data to analyze. Click 'Initiate G59 analysis & random forest' button to initiate analysis.

OPTION 2; Run directly in MATLAB

-Technically, should work on any Operating System with MATLAB installed (successfully tested in MATLAB R2018a/b on Windows OS).

Dependency

-Statistics and Machine Learning Toolbox, version 11.4 or higher.
-Bioinformatics Toolbox, version 4.11 or higher.
-MatSurv (Optional) https://github.com/aebergl/MatSurv (download and add to MATLAB path).

Installation and analysis

-Download 'RunG59.m', 'IBCsignature.m' and 'Data.tar.gz' file ('Data.tar.gz' contains data files used in Zare et al).
-Add 'IBCsignature.m' to MATLAB path.
-Open 'RunG59.m' and follow instructions in this file.
-'IBCsignature.m' can easily be customized to score novel datasets.