HS MV Analysis Documentation

Welcome to the documentation for HS MV Analysis, a software tool for analyzing hyperspectral microscopy data using multivariate analysis techniques. This documentation is designed to guide users through the installation, usage, and troubleshooting of the software, as well as provide detailed tutorials and reference materials.

Screenshot placeholder: annotated full GUI overview with the data loader, image viewer, spectral-axis widget, ROI Manager, analysis panel, and result/export areas labeled.

GIF placeholder: first-use overview from launching the GUI to loading a stack, adding a seed, running analysis, and opening the result viewer.

Start Here

Tutorials

The tutorials explain the app from the basic workflow upward. They are intentionally modality-independent first; dataset-specific examples are listed separately below.

GUI basics

Here you learn everything about the GUI and particularly the pyqtgraph plotting library, which is used for all interactive plots in the app. This is important to understand how to interact with the plots and how to interpret the visualizations.

01 Data loading

Here you learn how to load different data formats to the GUI. An important prerequisite is the tiff file format. The app can load both 3D TIFFs (x/y/spectral) and 4D TIFFs (x/y/z/spectral or x/y/time/spectral). The app also supports loading tile folders containing multiple TIFFs that together form a larger image.

02 Analysis

To understand the concepts of multivariate analysis, it is essential to understand the different analysis modes and how they work. This section explains the different modes and how to use them. The overview can help in particular to find better seed spectra for the NNMF and NNLS modes (see Seeds, spectra, and W maps), which can be crucial for getting good results.

The GPU acceleration page outlines how the optional GPU acceleration paths are set up. It is not essential to understand this for using the app, but it can be helpful for troubleshooting and for understanding the performance of the app.

03 Seeds

Remaining workflow

The rest of the tutorial is dedicated to handling results, exporting data, and ensuring reproducibility. This is important for getting the most out of the app and for sharing results with others.

Examples

The example section is reserved for data-specific workflows and for reproducing figure panels from a paper.

Methods

  • NNMF and NNLS modes — algorithms, math, convergence criteria, and literature references behind the analysis modes.

Reference