Skip to content

Research at St Andrews

A spatially heterogeneous network-based metapopulation software model applied to the simulation of a pulmonary tuberculosis infection

Research output: Contribution to journalArticlepeer-review


Tuberculosis (TB) is an ancient disease that, although curable, still accounts for over 1 million deaths worldwide. Shortening treatment time is an important area of research but is hampered by the lack of models that mimic the full range of human pathology. TB shows distinct localisations during different stages of infection, the reasons for which are poorly understood. Greater understanding of how heterogeneity within the human lung influences disease progression may hold the key to improving treatment efficiency and reducing treatment times.

In this work, we present a novel in silico software model which uses a networked metapopulation incorporating both spatial heterogeneity and dissemination possibilities to simulate a TB infection over the whole lung and associated lymphatics. The entire population of bacteria and immune cells is split into a network of patches: members interact within patches and are able to move between them. Patches and edges of the lung network include their own environmental attributes which influence the dynamics of interactions between the members of the subpopulations of the patches and the translocation of members along edges.

In this work, we detail the initial findings of a whole-organ model that incorporates distinct spatial heterogeneity features which are not present in standard differential equation approaches to tuberculosis modelling. We show that the inclusion of heterogeneity within the lung landscape when modelling TB disease progression has significant outcomes on the bacterial load present: a greater differential of oxygen, perfusion and ventilation between the apices and the basal regions of the lungs creates micro-environments at the apex that are more preferential for bacteria, due to increased oxygen availability and reduced immune activity, leading to a greater overall bacterial load present once latency is established.

These findings suggest that further whole-organ modelling incorporating more sophisticated heterogeneities within the environment and complex lung topologies will provide more insight into the environments in which TB bacteria persist and thus help develop new treatments which are factored towards these environmental conditions.


Original languageEnglish
Article number33
Number of pages21
JournalApplied Network Science
Publication statusPublished - 23 Aug 2018

    Research areas

  • Complex networks, Metapopulation, Spatial heterogeneity, Tuberculosis, In-host modelling, Computational biology

Discover related content
Find related publications, people, projects and more using interactive charts.

View graph of relations

Related by author

  1. Modelling the effects of environmental heterogeneity within the lung on the tuberculosis life-cycle

    Pitcher, M. J., Bowness, R., Dobson, S. A., Eftimie, R. & Gillespie, S. H., 7 Dec 2020, In: Journal of Theoretical Biology. 506, 18 p., 110381.

    Research output: Contribution to journalArticlepeer-review

  2. Tuberculosis bacillary load, an early marker of disease severity and treatment response: the utility of tuberculosis Molecular Bacterial Load Assay

    Sabiiti, W., Azam, K., Farmer, E., Kuchaka, D., Mtafya, B., Bowness, R., Oravcova, K., Honeyborne, I., Evangelopoulos, D., McHugh, T. D., Khosa, C., Rachow, A., Heinrich, N., Kampira, E., Davies, G., Bhatt, N., Ntinginya, E. N., Viegas, S., Jani, I., Kamdolozi, M. & 8 others, Mdolo, A., Khonga, M., Boeree, M. J., Phillips, P. PJ., Sloan, D. J., Hoelscher, M., Kibiki, G. & Gillespie, S. H., Jul 2020, In: Thorax. 75, 7, p. 606-608 3 p.

    Research output: Contribution to journalArticlepeer-review

ID: 255191074