Scaling DEPP phylogenetic placement to ultra-large reference trees: a tree-aware ensemble approach
Bioinformatics · 2024
DOI: 10.1093/bioinformatics/btae361 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2024. Verfasst von Yueyu Jiang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Electrical and Computer Engineering Department, University of California San Diego, 9500 Gilman Dr, La Jolla, CA, 92093, United States. Hauptthemen laut MEDLINE: Phylogeny. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NIGMS NIH HHS, National Institute of Health.
Abstract (Originalsprache)
Motivation Phylogenetic placement of a query sequence on a backbone tree is increasingly used across biomedical sciences to identify the content of a sample from its DNA content. The accuracy of such analyses depends on the density of the backbone tree, making it crucial that placement methods scale to very large trees. Moreover, a new paradigm has been recently proposed to place sequences on the species tree using single-gene data. The goal is to better characterize the samples and to enable combined analyses of marker-gene (e.g., 16S rRNA gene amplicon) and genome-wide data. The recent method DEPP enables performing such analyses using metric learning. However, metric learning is hampered by a need to compute and save a quadratically growing matrix of pairwise distances during training. Thus, the training phase of DEPP does not scale to more than roughly 10 000 backbone species, a problem that we faced when trying to use our recently released Greengenes2 (GG2) reference tree containing 331 270 species. Results This paper explores divide-and-conquer for training ensembles of DEPP models, culminating in a method called C-DEPP. While divide-and-conquer has been extensively used in phylogenetics, applying divide-and-conquer to data-hungry machine-learning methods needs nuance. C-DEPP uses carefully crafted techniques to enable quasi-linear scaling while maintaining accuracy. C-DEPP enables placing 20 million 16S fragments on the GG2 reference tree in 41 h of computation. Availability and implementation The dataset and C-DEPP software are freely available at https://github.com/yueyujiang/dataset_cdepp/.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Electrical and Computer Engineering Department, University of California San Diego, 9500 Gilman Dr, La Jolla, CA, 92093, United States.
- Pediatrics Department, University of California San Diego, 9500 Gilman Dr, La Jolla, CA, 92093, United States.
- Center for Microbiome Innovation, Jacobs School of Engineering, University of California San Diego, 9500 Gilman Dr, La Jolla, CA, 92093, United States.
ORCID-Profile: Yueyu Jiang
Förderung
- NIGMS NIH HHS
- National Institute of Health
Zitieren
APA
Jiang, Y., McDonald, D., Perry, D., Knight, R., & Mirarab, S. (2024). Scaling DEPP phylogenetic placement to ultra-large reference trees: a tree-aware ensemble approach. Bioinformatics. https://doi.org/10.1093/bioinformatics/btae361
Vancouver
Jiang Y, McDonald D, Perry D, Knight R, Mirarab S. Scaling DEPP phylogenetic placement to ultra-large reference trees: a tree-aware ensemble approach. Bioinformatics. 2024. doi:10.1093/bioinformatics/btae361
BibTeX
@article{Jiang2024,
author = {Jiang, Yueyu and McDonald, Daniel and Perry, Daniela and Knight, Rob and Mirarab, Siavash},
title = {Scaling DEPP phylogenetic placement to ultra-large reference trees: a tree-aware ensemble approach},
journal = {Bioinformatics},
year = {2024},
doi = {10.1093/bioinformatics/btae361}
}
EUVIMED