Figures
In Figs 4 and 5, there are infographics included to clarify the workflow, however, the labels on these graphics are incorrect. The authors have provided a corrected version of Figs 4 and 5 here.
Embeddings were trained on American Gut data, and the predictive models were trained and tested on Halfvarson dataset (A). Transforming microbiome data into GloVe embedding space (100 features) prior to model training produces more accurate models than using ASVs (26,251 features) (B).
Two models, one embedding-based and one ASV-based, were trained on American Gut data and tested on two independent query datasets (A). Embedding-based models outperform ASV-based models significantly when testing on Halfvarson dataset (B) and Schirmer dataset (C).
Reference
- 1. Tataru CA, David MM (2020) Decoding the language of microbiomes using word-embedding techniques, and applications in inflammatory bowel disease. PLoS Comput Biol 16(5): e1007859. https://doi.org/10.1371/journal.pcbi.1007859 pmid:32365061
Citation: Tataru CA, David MM (2020) Correction: Decoding the language of microbiomes using word-embedding techniques, and applications in inflammatory bowel disease. PLoS Comput Biol 16(11): e1008423. https://doi.org/10.1371/journal.pcbi.1008423
Published: November 2, 2020
Copyright: © 2020 Tataru, David. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.