Skip to main content Skip to navigation

Longitudinal Analysis of UK Anaerobic Digestion Microbiomes

Longitudinal Analysis of UK Anaerobic Digestion Microbiomes

Anaerobic Digestion (AD) is an important renewable energy technology involving the breakdown of waste products by microbes to produce biogas consisting mostly of methane. AD reactors contain whole communities of microbes, which rely on each other to survive. In particular, methane production is achieved by so-called methanogens, but these organisms cannot directly access the complex organic waste input to the reactor. Thus, the AD microbial communities can be seen as complex ecosystems implementing a food chain from complex organics all the way down to methane and other output products.

The health of the microbial communities residing in AD reactors is crucial to the final output of AD reactors. Changes in conditions can cause the microbial community to change and reduce the output and stability of the reactor. We do not currently have detailed understanding of which microbes constitute the communities in the AD reactors, how these microbes interact with each other, and how the interactions within the community and stability of the community change over time.

In this BBSRC funded project, led by Professors Orkun SoyerLink opens in a new window and Chris QuinceLink opens in a new window, we used state-of-the-art genetic sequencing techniques and extensive longitudinal sampling of AD reactors to help answers these questions. We took weekly samples from participating AD reactors over 12 months and also recorded meta-data on operational conditions, such as amount of organics fed into reactors and methane production. The longitudinal and large amount of high- quality data has allowed us to characterise which species of microbes inhabit the AD reactors, what their functions are and, importantly, how they interact with each other.

This longitudinal study will provide highly useful insights not only for academics but also for the UK AD industry. It provides an unbiased and transparent source of information on the performance of large numbers of industrial AD reactors, which will help to enhance and expand this technology in the UK. From a scientific perspective, the longitudinal data collected and analysed in this project constitutes the most detailed dataset of this type on any microbial community (as of writing, in 2021).

The results so far from this analysis are summarised in the following publications. These results not only provide biological insights on species composition, species-species interactions, and dynamics of AD microbial communities, but also include new methods that we have (and are) developing to analyse this type of meta genomic data. Some of the highlights of our findings so far include:

  • Identification of over 2000 new species of microbes
  • Identification of first representative genomes consituting novel microbial classes and orders
  • First time prediction of microbial interactions from longitudinal meta-genomic data
  • Prediction of over 500 novel microbial interactions or associations
  • Novel and better methods for meta genome assembly at strain level
  • Novel methods for assignment of species functions from their genomes


Novel microbial syntrophies identified by longitudinal metagenomics
Sebastien Raguideau, Anna Trego, Fred Farrell, Gavin Collins, Christopher Quince, Orkun S Soyer. Posted July 05, 2021 | bioRxiv linkLink opens in a new window
STRONG: metagenomics strain resolution on assembly graphs

Quince, C., Nurk, S., Raguideau, S., James R., Soyer OS., Summers JK., Limasset A., Eren AM., Chikhi R & Darling A. Genome Biol 22, 214 (2021)

Machine learning based prediction of functional capabilities in metagenomically assembled microbial genomes.

Fred Farrell, Orkun S Soyer, Christopher Quince. Posted April 25, 2018 | bioRxiv linkLink opens in a new window

Datasets & Tools

The metagenomic and 16S datasets described in Raguideau et al. (2021) are deposited in the European Nucleotide Archive (ENA) under project PRJEB39861.

The Metadata described in Raguideau et al (2021) is available as a supplementary material (link).

Our pipeline used for the meta genome analyses is freely available from

We developed a machine learning based tool to infer species functional capabilities from their genomes, available at