AUTHOR=Gardes Joël , Tchatchueng-Mbougua Jules Brice , Maldivi Christophe , Jelassi Mariem , ben Khalfallah Houssem , Demongeot Jacques 

TITLE=Maxwell® an AAAA classifier well-suited to biomedical data clustering

JOURNAL=Frontiers in Applied Mathematics and Statistics

VOLUME=Volume 11 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/applied-mathematics-and-statistics/articles/10.3389/fams.2025.1634300

DOI=10.3389/fams.2025.1634300

ISSN=2297-4687

ABSTRACT=IntroductionA new classifier called Maxwell®, Adiabatic, Agnostic and Almost Autonomous, is presented and used to classify species according to their early occurrence in evolution.MethodsAfter a precise description of all the steps of the clustering process, two examples of application are given: first, the classification of simulated genomic data, whose simulation mode is processed by an algorithm allowing the successive application of known operators having acted during the evolution of species. The clustering thus obtained makes it possible to identify correctly the genomes of species having evolved in the same ecosystem. Then, mitochondrial genomes of mammals and giant viruses associated with their bacterial or fungal targets they infect, are classified according to the same criteria.ResultsThe results show a good adequacy of the obtained classifications to the evolutionary reality and a high consistency with the known knowledge on the evolution of the oldest species.DiscussionThe Maxwell® classifier presents a unique set of properties, adiabatic, agnostic and almost autonomous, making it particularly suitable for biomedical applications.