System Requirements

CALISTA toolbox is originally written for MATLAB. The subroutines in CALISTA (version 1.2.2) have been successfully tested on MATLAB 2016b, 2017a, 2018a and 2018b.

CALISTA-R toolbox is also provided for R users. The subroutines in CALISTA (version 1.1.2) have been successfully tested on R version 3.6.0 and RStudio Version 1.2.1335 

Last Update


Current version: CALISTA-MATLAB v1.2.2 (20.02.2019) CALISTA-R v1.1.2 (16.05.2019)




By default CALISTA employs Principal Component Analysis (PCA) for dimensional reduction and visualization of the expression data. Additionally, the user can also generate 2D t-SNE plots by setting INPUTS.plot_tsne = 1; in the main script as shown in this Example 7. CALISTA adopts FIt-SNE subroutines (please visit here for more details). The only prerequisite for FItt-SNE is FFTW, which can be downloaded and installed from the website.


CALISTA R package is built using Fortran subroutines. Before running CALISTA-R, please install the latest version of GFortran compiler.

Download and Installation


Download and unzip the CALISTA-MATLAB for codes and data.

CALISTA-R (R version):

To install CALISTA-R directly from github repository please run the following commands in R:




For the old version of CALISTA-R (version 1.0.2): download and unzip the CALISTA-R for codes and data



Redistribution and use in source and binary forms, with or without modification, are permitted provided agreeing to the BSD 3-Clause Style License.