Mots-Clés
Metagenomics
Dimensionality reduction
Probabilistic models
soil microbiome
Description
Titre du stage : Understanding the effect of agricultural practices on soil prokaryotic biodiversity and its functions through probabilistic modeling of metagenomic data
Début du stage souhaité : janvier - mars 2027 (flexible)
Résumé du projet de stage :
Soil microbial communities are responsible for the cycling of nutrients in soil through organic matter degradation, and are therefore essential for the maintenance of productive land ecosystems. This includes agricultural land, where the role of soil microbiome has long been overlooked (Bertola et al., 2021). To understand the effect of agricultural practices on soil prokaryotic diversity—which encompasses both bacteria and archaea—and the functions it performs, we collected soil samples from both agricultural and uncultivated land across France and Europe and thoroughly characterised their prokaryotic microbiome and local pedological and environmental conditions. Specifically, we performed deep shotgun metagenomic sequencing so as to both identify the prokaryotic taxa found in the samples and reconstruct their genomes, which inform on the molecular functions they perform (Anthony et al., 2024).
The general objective of this internship is to develop a probabilistic model describing variation in the taxonomic composition of soil communities through a reduced number of interpretable latent components. Namely, rather than considering thousands of taxa to describe these microbiomes, we aim to represent them using a much smaller number of taxa groups. Each group can be interpreted as a recurring subcommunity of taxa found together in different samples. The inference of subcommunities exploits the co-occurrence and covariance patterns of different taxa or genes across samples (Sommeria-Klein et al., 2020). By following a Bayesian approach, the subcommunities can be further constrained according to the biological question being addressed, for instance, distinguishing native soils from cultivated soils, or positioning microbiome profiles in the broader context of soil microbiome diversity as represented in existing databases.
A first aim of this internship project is to assess the feasibility of applying the above-mentioned probabilistic modeling approach to a limited number of samples, such as our dataset of 200 soil samples. To facilitate inference and improve the ecological interpretability of the subcommunities, the intern will constrain the distribution of subcommunities across samples with the samples’ pedological and environmental characteristics, so that each subcommunity can be interpreted as a set of taxa that tend to be found together in a certain soil type. If necessary, the intern will also supplement the dataset with samples from public databases to inform the model. The second aim of the project is to use the developed model to understand the effect of agricultural practices on soil prokaryotic biodiversity and its functions.
Compétences souhaitées :
- Master en modélisation pour l’écologie ou en biologie computationelle
- Bases solides en statistiques et probabilités
- Programmation Python
- Bon niveau en anglais
Références
Anthony, W.E., Allison, S.D., Broderick, C.M., Chavez Rodriguez, L., Clum, A., Cross, H., Eloe-Fadrosh, E., Evans, S., Fairbanks, D., Gallery, R., Gontijo, J.B., Jones, J., McDermott, J., Pett-Ridge, J., Record, S., Rodrigues, J.L.M., Rodriguez-Reillo, W., Shek, K.L., Takacs-Vesbach, T. & Blanchard, J.L. (2024) From soil to sequence: filling the critical gap in genome-resolved metagenomics is essential to the future of soil microbial ecology. Environmental Microbiome, 19, 56.
Bertola, M., Ferrarini, A. & Visioli, G. (2021) Improvement of Soil Microbial Diversity through Sustainable Agricultural Practices and Its Evaluation by -Omics Approaches: A Perspective for the Environment, Food Quality and Human Safety. Microorganisms, 9, 1400.
Sommeria-Klein, G., Zinger, L., Coissac, E., Iribar, A., Schimann, H., Taberlet, P. & Chave, J. (2020) Latent Dirichlet Allocation reveals spatial and taxonomic structure in a DNA‐based census of soil biodiversity from a tropical forest. Molecular Ecology Resources, 20, 371–386.