@article {Michel-Mata2021.06.17.448886, author = {Sebastian Michel-Mata and Xu-Wen Wang and Yang-Yu Liu and Marco Tulio Angulo}, title = {Predicting microbiome compositions through deep learning}, elocation-id = {2021.06.17.448886}, year = {2021}, doi = {10.1101/2021.06.17.448886}, publisher = {Cold Spring Harbor Laboratory}, abstract = {Microbes can form complex communities that perform critical functions in maintaining the integrity of their environment1,2 or the well-being of their hosts3{\textendash}6. Successfully managing these microbial communities requires the ability to predict the community composition based on the species assemblage7. However, making such a prediction remains challenging because of our limited knowledge of the diverse physical8, biochemical9, and ecological10,11 processes governing the microbial dynamics. To overcome this challenge, here we present a deep learning framework that automatically learns the map between species assemblages and community compositions from training data. First, we systematically validate our framework using synthetic data generated by classical population dynamics models. Then, we apply it to experimental data of both in vitro and in vivo communities, including ocean and soil microbial communities12,13, Drosophila melanogaster gut microbiota14, and human gut and oral microbiota15. Our results demonstrate how deep learning can enable us to understand better and potentially manage complex microbial communities.Competing Interest StatementThe authors have declared no competing interest.}, URL = {https://www.biorxiv.org/content/early/2021/06/17/2021.06.17.448886}, eprint = {https://www.biorxiv.org/content/early/2021/06/17/2021.06.17.448886.full.pdf}, journal = {bioRxiv} }