Rxivist logo

Rxivist combines preprints from bioRxiv with data from Twitter to help you find the papers being discussed in your field. Currently indexing 62,502 bioRxiv papers from 277,505 authors.

MICOM: metagenome-scale modeling to infer metabolic interactions in the gut microbiota

By Christian Diener, Sean Gibbons, Osbaldo Resendis-Antonio

Posted 04 Jul 2018
bioRxiv DOI: 10.1101/361907

Compositional changes in the gut microbiota have been associated with a variety of medical conditions such as obesity, Crohn’s disease and diabetes. However, connecting microbial community composition to ecosystem function remains a challenge. Here, we introduce MICOM – a customizable metabolic model of the human gut microbiome. By using a heuristic optimization approach based on L2 regularization we were able to obtain a unique set of realistic growth rates that corresponded well with observed replication rates. We integrated adjustable dietary and taxon abundance constraints to generate personalized metabolic models for individual metagenomic samples. We applied MICOM to a balanced cohort of metagenomes from 186 people, including a metabolically healthy population and individuals with type 1 and type 2 diabetes. Model results showed that individual bacterial genera maintained conserved niche structures across humans, while the community-level production of short chain fatty acids (SCFAs) was heterogeneous and highly individual-specific. Model output revealed complex cross-feeding interactions that would be difficult to measure in vivo . Metabolic interaction networks differed somewhat consistently between healthy and diabetic subjects. In particular MICOM predicted reduced butyrate and propionate production in a diabetic cohort, with restoration of SCFA production profiles found in healthy subjects following metformin treatment. Overall, we found that changes in diet or taxon abundances have highly personalized effects. We believe MICOM can serve as a useful tool for generating mechanistic hypotheses for how diet and microbiome composition influence community function. All methods are implemented in the open source Python package, which is available at <https://github.com/micom-dev/micom>.

Download data

  • Downloaded 970 times
  • Download rankings, all-time:
    • Site-wide: 7,390 out of 62,502
    • In systems biology: 209 out of 1,774
  • Year to date:
    • Site-wide: 4,656 out of 62,502
  • Since beginning of last month:
    • Site-wide: 3,122 out of 62,502

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide

Sign up for the Rxivist weekly newsletter! (Click here for more details.)