Project Completion
SIAM ALA plenary talk on Coupled Matrix and Tensor Factorizations – Improving our Understanding of Complex Systems Through the Analysis of Temporal and Multimodal Data (May 17, 2024 - Paris, France)
We have officially completed the TrACEr (Time-aware Constrained Multimodal Data Fusion) project funded by the Research Council of Norway and Novo Nordisk Foundation.
The goal of the project has been to develop novel data mining methods that can jointly analyze static and dynamic data from multiple sources, and capture underlying (evolving) patterns. Our method development efforts have been motivated by a challenging system: the human metabolome. We have used the developed methods to analyze measurements of blood samples collected during a meal challenge test from the COPSAC2000 cohort to capture differences among subject stratifications in terms of their metabolic response to the challenge test. In addition, we have aimed to show the broader impact of the developed methods in other domains, in particular, in neuroscience.
The project has brought together researchers from computer science, applied math, chemometrics, food science, medicine, nutrition, computational biology, neuroscience, and signal processing. Here are some highlights from the project:
Data Science:
Two completed PhD theses:
Marie Roald, Understanding the Dynamics of Complex Systems Through Time-Evolving Data Mining, 2023
Carla Schenker, A Flexible Framework for Data Fusion Based on Coupled Matrix and Tensor Factorizations for Interpretable Pattern Discovery, 2023
Algorithmic and methodological advances:
C. Schenker, J. E. Cohen, E. Acar. A Flexible Optimization Framework for Regularized Matrix-Tensor Factorizations with Linear Couplings, IEEE Journal of Selected Topics in Signal Processing, 2021
M. Roald, C. Schenker, V. Calhoun, T. Adali, R. Bro, J. E. Cohen, E. Acar. An AO-ADMM approach to constraining PARAFAC2 on all modes, SIAM Journal on Mathematics of Data Science, 2022
Metabolomics:
L. Li, S. Yan, B. M. Bakker, H. Hoefsloot, B. Chawes, D. Horner, M. A. Rasmussen, A. K. Smilde, E. Acar. Analyzing postprandial metabolomics data using multiway models: A simulation study, BMC Bioinformatics, 2024
S. Yan, L. Li, D. Horner, P. Ebrahimi, B. Chawes, L. O. Dragsted, M. A. Rasmussen, A. K. Smilde, E Acar. Characterizing human postprandial metabolic response using multiway data analysis, Metabolomics, 2024
L. Li, S. Yan, D. Horner, M. A. Rasmussen, A. K. Smilde, E. Acar. Revealing static and dynamic biomarkers from postprandial metabolomics data through coupled matrix and tensor factorizations, Metabolomics, 2024
Neuroscience:
E. Acar, M. Roald, K. M. Hossain, V. D. Calhoun, T. Adali. Tracing Evolving Networks using Tensor Factorizations vs. ICA-based Approaches. Frontiers in Neuroscience, 2022
We have given over 35 talks in diverse venues including signal processing, data science conferences as well as metabolomics, nutrition conferences, and organized several workshops/minisymposia. As the last talk of the project, I had the privilege to present our results in a plenary talk at SIAM Conference on Applied Linear Algebra in May 2024.
For papers and repositories, and further updates in future, please check out the webpage: https://tracer.simulamet.no/
Thank you to everyone who has contributed to TrACEr!
Carla Schenker, Lu Li, Marie Roald, Christos Chatzis, Shi Yan, Xuilin Wang, Rasmus Bro, Age Smilde, David Horner, Parvaneh Ebrahimi, Morten A. Rasmussen, B. Chawes, Jeremy E. Cohen, Tulay Adali, Huub Hoefsloot, Barbara M. Bakker, TrACEr scientific advisory board: Nikos Sidiropoulos, Tammy Kolda, Nathan Price, Suzan Wopereis
