@article{fdi:010091096,
  title = {{D}eep learning methods in metagenomics : a review},
  author = {{R}oy, {G}aspar and {P}rifti, {E}di and {B}elda, {E}. and {Z}ucker, {J}ean-{D}aniel},
  editor = {},
  language = {{ENG}},
  abstract = {{T}he ever- decreasing cost of sequencing and the growing potential applications of metagenomics have led to an unprecedented surge in data generation. {O}ne of the most prevalent applications of metagenomics is the study of microbial environments, such as the human gut. {T}he gut microbiome plays a crucial role in human health, providing vital information for patient diagnosis and prognosis. {H}owever, analysing metagenomic data remains challenging due to several factors, including reference catalogues, sparsity and compositionality. {D}eep learning ({DL}) enables novel and promising approaches that complement state- of- the- art microbiome pipelines. {DL}- based methods can address almost all aspects of microbiome analysis, including novel pathogen detection, sequence classification, patient stratification and disease prediction. {B}eyond generating predictive models, a key aspect of these methods is also their interpretability. {T}his article reviews {DL} approaches in metagenomics, including convolutional networks, autoencoders and attention- based models. {T}hese methods aggregate contextualized data and pave the way for improved patient care and a better understanding of the microbiome's key role in our health.},
  keywords = {microbiome ; metagenomics ; deep learning ; neural network ; embedding ; binning ; disease prediction.},
  booktitle = {},
  journal = {{M}icrobial {G}enomics},
  volume = {10},
  numero = {4},
  pages = {001231 [28 p.]},
  ISSN = {2057-5858},
  year = {2024},
  DOI = {10.1099/mgen.0.001231},
  URL = {https://www.documentation.ird.fr/hor/fdi:010091096},
}