DeepMPTB: a vaginal microbiome-based deep neural network as artificial intelligence strategy for efficient preterm birth prediction.

In recent decades, preterm birth (PTB) has become a significant research focus in the healthcare field, as it is a leading cause of neonatal mortality worldwide. Using five independent study cohorts including 1290 vaginal samples from 561 pregnant women who delivered at term (n = 1029) or prematurely (n = 261), we analysed vaginal metagenomics data for precise microbiome structure characterization. Then, a deep neural network (DNN) was trained to predict term birth (TB) and PTB with an accuracy of 84.10% and an area under the receiver operating characteristic curve (AUROC) of 0.875 ± 0.11. During a benchmarking process, we demonstrated that our DL model outperformed seven currently used machine learning algorithms. Finally, our results indicate that overall diversity of the vaginal microbiota should be taken in account to predict PTB and not specific species. This artificial-intelligence based strategy should be highly helpful for clinicians in predicting preterm birth risk, allowing personalized assistance to address various health issues. DeepMPTB is open source and free for academic use. It is licensed under a GNU Affero General Public License 3.0 and is available at https://deepmptb.streamlit.app/ . Source code is available at https://github.com/oschakoory/DeepMPTB and can be easily installed using Docker ( https://www.docker.com/ ).

Targeted 16S rRNA Gene Capture by Hybridization and Bioinformatic Analysis.

Next-generation sequencing technologies have impressively unlocked capacities to depict the complexity of microbial communities. Microbial community structure is for now routinely monitored by sequencing of 16S rRNA gene, a phylogenetic marker almost conserved among bacteria and archaea. Nevertheless, amplicon sequencing, the most popular used approach, suffers from several biases impacting the picture of microbial communities. Here, we describe an innovative method based on gene capture by hybridization for the targeted enrichment of 16S rDNA biomarker from metagenomic samples. Coupled to near full-length 16S rDNA reconstruction, this approach enables an exhaustive and accurate description of microbial communities by enhancing taxonomic and phylogenetic resolutions. Furthermore, access of captured 16S flanking regions opens link between structure and function in microbial communities.

RiboTaxa: combined approaches for rRNA genes taxonomic resolution down to the species level from metagenomics data revealing novelties.

Metagenomic classifiers are widely used for the taxonomic profiling of metagenomics data and estimation of taxa relative abundance. Small subunit rRNA genes are a gold standard for phylogenetic resolution of microbiota, although the power of this marker comes down to its use as full-length. We aimed at identifying the tools that can efficiently lead to taxonomic resolution down to the species level. To reach this goal, we benchmarked the performance and accuracy of rRNA-specialized versus general-purpose read mappers, reference-targeted assemblers and taxonomic classifiers. We then compiled the best tools (BBTools, FastQC, SortMeRNA, MetaRib, EMIRGE, VSEARCH, BBMap and QIIME 2's Sklearn classifier) to build a pipeline called RiboTaxa. Using metagenomics datasets, RiboTaxa gave the best results compared to other tools (i.e. Kraken2, Centrifuge, METAXA2, phyloFlash, SPINGO, BLCA, MEGAN) with precise taxonomic identification and relative abundance description without false positive detection (F-measure of 100% and 83.7% at genus level and species level, respectively). Using real datasets from various environments (i.e. ocean, soil, human gut) and from different approaches (e.g. metagenomics and gene capture by hybridization), RiboTaxa revealed microbial novelties not discerned by current bioinformatics analysis opening new biological perspectives in human and environmental health.