Towards virtual physiological humans: integrating metabolism, physiology, and gut-microbiome

Science communication on a recently published paper


Precision Medicine

A milestone in precision medicine would be the ability to explore generative models that can capture the real-world complexity of human physiology. If computational models can successfully integrate physiology, metabolism, and host-microbiome data – thereby achieving personalization of predictive models of virtual physiological humans – this would be a key stepping-stone in yielding promising therapeutic targets that are both individual-specific and case-specific. Thus, the science (and art) of generating and exploring virtual physiological humans could be a boon to systems physiology and precision medicine.

Whole-body metabolic reconstructions

Towards this end, Thiele et al. recently published an intriguing paper. Thiele et al. demonstrate building and validation of sex-specific curated Whole-Body Metabolic (WBM) reconstructions. Computational models derived from such WBM reconstructions offer a novel molecular-level, anatomically and physiologically consistent, sex-specific genome-scale reconstructions of human physiology and metabolism. Thiele et al. further demonstrated that the WBM reconstructions can be ‘personalized’ via integration of quantitative physiological data with multi-omics such as metabolomics, and gut-microbiome. Thus, Thiele et al. introduce Harvey and Harvetta, the male and female WBM reconstructions, respectively. These reconstructions and the derived models and analysis enable novel assessment of, e.g., host-microbiome co-metabolism that is individual-specific and is resolved at an organ-level.

Complexity

The extreme complexity of real-world metabolic modelling of host-microbiome in humans can hardly be overstated. Thiele et al. address this critical challenge via constraint-based reconstruction – their models account for the enormous diversity of microbiome-associated microbial genes that are meaningful both physiologically and is individual-specific such as in host-microbiome co-metabolism. Thus this approach allows case-specific investigation of human metabolism, such as for studying inherited metabolic diseases.

Approach

Previously the Thiele Lab , extensive metabolic modelling of metabolic and physiologically relevant pathways (curated algorithmically and manually) led to a comprehensive model that allows investigation via constraint-based reconstruction of over 80,000 pathways. Thiele et al. approached the problem into a smaller sub-problems that they have previously addressed. First – by compiling organ-specific information from previous literature and including experimental-omics data, Thiele Lab generated WBM reconstructions. Second – Thiele Lab imposed constraints the WBM reconstructions via (large-scale) physiology, dietary constraints and quantitative metabolomic data.

Highlights

A key challenge that Thiele et al. addressed in this current 2020 paper is the integration of more than 80,000 biochemical reactions, that were both anatomically and physiologically consistent. These WBM reconstructions describe 26 organs and 6 blood cell types and capture whole-body organ-resolved metabolism. Importantly, these recapitulate previously known inter-organ metabolic cycles and energy use. Interestingly, WBM models can also predict biomarkers. This might, e.g., elucidate new pathways or targets that are implicated in heritable metabolic diseases.

Take-home-message and future ahead

Exploring such quantitative metabolic models could reveal new molecular insights pertaining to (co-dependence of host-microbiome) metabolism and overall human physiology and homeostasis – both in health and disease. The power of computational models is further exemplified if these allow addressing questions via novel hypothesis-testing that go beyond the scope of wet-lab experiments.


written by: Div Prasad

I’m interested in quantitative biology (at the intersection) of human genetics, machine learning, and host-microbiome interactions.

Twitter: @divyaePrasad
github: divprasad

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s