enmelvan

Chapter-5

The human microbiome is an essential factor to human health and disruption of the gut microbiota can lead to diseases such as obesity, diabetes, and irritable bowel syndrome. Understanding the interactions of these bacteria can lead to a better understanding of how they might be managed to achieve certain goals. Even though our current knowledge of both commensal and pathogenic microbes surpasses the pairwise interactions and correlations and there are already dynamic models which are mostly based on GLVM, such as BEEM-Static, there is still a challenge with drawing inferences from the fitted coefficients. Better quality of dynamic models is essential, and this research study continues the BEEM-Static research in the direction of building a stronger Bayesian model of the gut microbiota. A formulation of a fully Bayesian model allows to statistically evaluate the extent to which carry capacities and interaction coefficient differ significantly between two populations. The model was tested on a BeemDemo simulated database and on a Human Microbiome Project database that contains information about a person’s body weight. We can conclude that, while slower, modelling of microbiome interactions is significantly better when done in a Bayesian framework

Chapter2

Chapter3

Insight into Dynamics and Stability of Intestinal Microbiota Linked to Obesity

Chapter4

Data and code for Chapter3

Dynamics-of-urinary-microbiota-associated-with-cystitis

Detailed analysis steps

enmelvan/Dynamics-of-urinary-microbiota-associated-with-cystitis

Detailed analysis steps

enmelvan/Chapter-5

The human microbiome is an essential factor to human health and disruption of the gut microbiota can lead to diseases such as obesity, diabetes, and irritable bowel syndrome. Understanding the interactions of these bacteria can lead to a better understanding of how they might be managed to achieve certain goals. Even though our current knowledge of both commensal and pathogenic microbes surpasses the pairwise interactions and correlations and there are already dynamic models which are mostly based on GLVM, such as BEEM-Static, there is still a challenge with drawing inferences from the fitted coefficients. Better quality of dynamic models is essential, and this research study continues the BEEM-Static research in the direction of building a stronger Bayesian model of the gut microbiota. A formulation of a fully Bayesian model allows to statistically evaluate the extent to which carry capacities and interaction coefficient differ significantly between two populations. The model was tested on a BeemDemo simulated database and on a Human Microbiome Project database that contains information about a person’s body weight. We can conclude that, while slower, modelling of microbiome interactions is significantly better when done in a Bayesian framework

enmelvan/Chapter2

enmelvan/Chapter3

Insight into Dynamics and Stability of Intestinal Microbiota Linked to Obesity

enmelvan/Chapter4

Data and code for Chapter3