Classification of multiple segmented profiles, application to the study of the neuronal protein Tau

TL;DR

This paper introduces a Gaussian mixture model with shared means for classifying neuronal protein Tau profiles, aiding understanding of its functioning and aggregation states in neuroscience.

Contribution

It proposes a novel constrained Gaussian mixture model with EM inference for classifying intensity profiles of Tau protein, tailored for neuroscience applications.

Findings

Model performs well in simulation studies

Effective classification of Tau protein profiles in real data

Provides insights into Tau protein aggregation states

Abstract

This work is motivated by an application in neuroscience, in particular by the study of the (dys)functioning of a protein called Tau. The objective is to establish a classification of intensity profiles, according to the presence or absence of the protein and its monomer or dimer proportion. For this, we propose a Gaussian mixture model with a fixed number of clusters whose mean parameters are constrained and shared by the clusters. The inference of this model is done via the classical EM algorithm. The performance of the method is evaluated via simulation studies and an application on real data is done.