Process & Systems Modelling

Maintaining phenotypic diversity in mechanistic models using system feedback: gut bacteria

We have developed a detailed mechanistic model describing how the interactions of many functionally-distinct groups of gut bacteria interact and adapt to create a diverse, stable microbial community. To simulate adaptation to a changing environment, we have taken a complex systems style approach and modelled the interactions of hundreds of bacterial strains with stochastically-generated traits. In a resource competition model with a constant environment, each functional group will eventually become completely dominated by the strain with the characteristics most suited to that environment. However, in reality phenotypic diversity is high in the colon with many functionally similar strains co-existing. We have found that diversity can be maintained in our models by including a feedback from the microbial system to the environment through changes in pH caused by the acidic metabolites produced during bacterial growth. In conjunction with a trade-off between acid tolerance and maximum bacterial growth rate, this feedback allows many strains to co-exist. Our methods are transferable to many analogous ecosystems featuring resource competition.

time evolution graphs Time evolution of the concentration of 30 bacterial strains in the Bacteroides group with a sinusoidal resource input representing food consumption during the day. Left panels show results when there is no feedback; right panels show results when there is a pH feedback from bacterial growth.

Further details from: Helen Kettle

Article date 2011


Statistical Genomics and Bioinformatics

Process and Systems Modelling

Statistical Methodology

PhD Opportunities

Meetings & Seminars