Tracking Equilibrium and Nonequilibrium Shifts in Data with TREND

Area of Science:

• Multivariate data analysis
• Biophysical measurements
• Computational biology

Background:

• Principal Component Analysis (PCA) is effective for pattern discovery in complex datasets.
• Recent advancements allow direct PCA application to spectra, images, and movies for identifying changes.
• A need exists for accessible tools to apply these methods across diverse biophysical measurements.

Purpose of the Study:

• To develop user-friendly, multiplatform software for analyzing population shifts in equilibrium and nonequilibrium states.
• To enable the application of PCA and Independent Component Analysis to various biophysical data types.
• To provide tools for resolving concurrent processes within complex datasets.

Main Methods:

• Development of TREND (Track Equilibrium and Nonequilibrium population shifts among two-dimensional Data frames) software.
• Implementation of Principal Component Analysis (PCA) and Independent Component Analysis (ICA) algorithms.
• Testing TREND with Nuclear Magnetic Resonance (NMR) titration spectra and cardiac Magnetic Resonance Imaging (MRI) movies.

Main Results:

• TREND successfully extracts dual binding phases from NMR titration spectra.
• TREND resolves principal components representing breathing and cardiac cycles in MRI movies.
• Reconstruction of measurement series from selected PCs is demonstrated, with higher fidelity requiring more PCs.

Conclusions:

• TREND provides convenient tools for resolving processes in diverse biophysical measurements.
• The software facilitates the analysis of equilibrium and nonequilibrium population shifts.
• TREND enhances the utility of PCA and ICA for complex data analysis in scientific research.