A tissue time machine to forward-predict single-timepoint data applied to tuberculosis granulomas
- 1Department of Microbiology and Immunology, University of Michigan-Michigan Medicine, Ann Arbor, MI, USA
Abstract
Recent advances have improved understanding of infection; however many detailed datasets are based on single timepoints post-infection and cannot generate longitudinal insights needed to predict infection outcomes. For example, pulmonary infection with Mycobacterium tuberculosis (Mtb) initiates a complex system of immune and bacterial interactions that leads to granuloma formation, yet single snapshots in time need elaboration long-term. We develop a new approach, the tissue time machine, applied to predict outcomes of non-human primate granulomas from initiation to endpoint using single timepoint tissue datasets including high-dimensional spatial features. This involves identifying granuloma digital partners, simulations most closely resembling real granulomas from a virtual library of over 50,000. By analyzing digital partners, we predict infection outcomes of 22/28 granulomas, and that Mtb-specific, IFN-{gamma}-producing T cell recruitment rates influence uncertainty for the 6/28. This method is applicable to any system that allows for mechanistic predictions over time using heterogeneous, incomplete, and high-dimensional datasets.
Related
- Response to comment on “Using genomic data and machine learning to predict antibiotic resistance: A tutorial paper”
- Serum proteins predict incident inflammatory bowel disease 14 years before diagnosis: a machine-learning and Cox modelling in the UK Biobank
- Engineered probiotics target granulomas to treat mycobacterial infection
- ARMD-Predict: multi-site machine learning to predict antibiotic resistance before culture results
- A Machine Learning Approach to Predict Listeria monocytogenes Growth by Integrating FSSP and PMP Simulations: A Focus on t₁₀₀ as a Possible Endpoint Parameters for Predictive Models.
BibTeX
@misc{nanda2026tissue,
title = {{A tissue time machine to forward-predict single-timepoint data applied to tuberculosis granulomas}},
author = {P. Nanda and E. F. McCaffrey and C. T. Michael and A. C. Delmastro and G. Michael and J. T. Mattila and J. L. Flynn and R. M. Angelo and D. Kirschner},
howpublished = {bioRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.09.758083},
url = {https://www.biorxiv.org/content/10.64898/2026.10.09.758083v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - A tissue time machine to forward-predict single-timepoint data applied to tuberculosis granulomas
AU - P. Nanda
AU - E. F. McCaffrey
AU - C. T. Michael
AU - A. C. Delmastro
AU - G. Michael
AU - J. T. Mattila
AU - J. L. Flynn
AU - R. M. Angelo
AU - D. Kirschner
PY - 2026
DA - 2026/10/09
PB - bioRxiv
DO - 10.64898/2026.10.09.758083
UR - https://www.biorxiv.org/content/10.64898/2026.10.09.758083v1
AB - Recent advances have improved understanding of infection; however many detailed datasets are based on single timepoints post-infection and cannot generate longitudinal insights needed to predict infection outcomes. For example, pulmonary infection with Mycobacterium tuberculosis (Mtb) initiates a complex system of immune and bacterial interactions that leads to granuloma formation, yet single snapshots in time need elaboration long-term. We develop a new approach, the tissue time machine, applied to predict outcomes of non-human primate granulomas from initiation to endpoint using single timepoint tissue datasets including high-dimensional spatial features. This involves identifying granuloma digital partners, simulations most closely resembling real granulomas from a virtual library of over 50,000. By analyzing digital partners, we predict infection outcomes of 22/28 granulomas, and that Mtb-specific, IFN-{gamma}-producing T cell recruitment rates influence uncertainty for the 6/28. This method is applicable to any system that allows for mechanistic predictions over time using heterogeneous, incomplete, and high-dimensional datasets.
ER -