In:
PLOS Computational Biology, Public Library of Science (PLoS), Vol. 18, No. 4 ( 2022-4-21), p. e1010066-
Abstract:
Machine learning-based classification approaches are widely used to predict host phenotypes from microbiome data. Classifiers are typically employed by considering operational taxonomic units or relative abundance profiles as input features. Such types of data are intrinsically sparse, which opens the opportunity to make predictions from the presence/absence rather than the relative abundance of microbial taxa. This also poses the question whether it is the presence rather than the abundance of particular taxa to be relevant for discrimination purposes, an aspect that has been so far overlooked in the literature. In this paper, we aim at filling this gap by performing a meta-analysis on 4,128 publicly available metagenomes associated with multiple case-control studies. At species-level taxonomic resolution, we show that it is the presence rather than the relative abundance of specific microbial taxa to be important when building classification models. Such findings are robust to the choice of the classifier and confirmed by statistical tests applied to identifying differentially abundant/present taxa. Results are further confirmed at coarser taxonomic resolutions and validated on 4,026 additional 16S rRNA samples coming from 30 public case-control studies.
Type of Medium:
Online Resource
ISSN:
1553-7358
DOI:
10.1371/journal.pcbi.1010066
DOI:
10.1371/journal.pcbi.1010066.g001
DOI:
10.1371/journal.pcbi.1010066.g002
DOI:
10.1371/journal.pcbi.1010066.g003
DOI:
10.1371/journal.pcbi.1010066.g004
DOI:
10.1371/journal.pcbi.1010066.g005
DOI:
10.1371/journal.pcbi.1010066.g006
DOI:
10.1371/journal.pcbi.1010066.t001
DOI:
10.1371/journal.pcbi.1010066.s001
DOI:
10.1371/journal.pcbi.1010066.s002
DOI:
10.1371/journal.pcbi.1010066.s003
DOI:
10.1371/journal.pcbi.1010066.s004
DOI:
10.1371/journal.pcbi.1010066.s005
DOI:
10.1371/journal.pcbi.1010066.s006
DOI:
10.1371/journal.pcbi.1010066.s007
DOI:
10.1371/journal.pcbi.1010066.s008
DOI:
10.1371/journal.pcbi.1010066.s009
DOI:
10.1371/journal.pcbi.1010066.s010
DOI:
10.1371/journal.pcbi.1010066.s011
DOI:
10.1371/journal.pcbi.1010066.s012
DOI:
10.1371/journal.pcbi.1010066.s013
DOI:
10.1371/journal.pcbi.1010066.s014
DOI:
10.1371/journal.pcbi.1010066.s015
DOI:
10.1371/journal.pcbi.1010066.s016
DOI:
10.1371/journal.pcbi.1010066.s017
DOI:
10.1371/journal.pcbi.1010066.s018
DOI:
10.1371/journal.pcbi.1010066.s019
DOI:
10.1371/journal.pcbi.1010066.s020
DOI:
10.1371/journal.pcbi.1010066.s021
DOI:
10.1371/journal.pcbi.1010066.s022
DOI:
10.1371/journal.pcbi.1010066.s023
DOI:
10.1371/journal.pcbi.1010066.s024
DOI:
10.1371/journal.pcbi.1010066.s025
DOI:
10.1371/journal.pcbi.1010066.r001
DOI:
10.1371/journal.pcbi.1010066.r002
DOI:
10.1371/journal.pcbi.1010066.r003
DOI:
10.1371/journal.pcbi.1010066.r004
DOI:
10.1371/journal.pcbi.1010066.r005
DOI:
10.1371/journal.pcbi.1010066.r006
Language:
English
Publisher:
Public Library of Science (PLoS)
Publication Date:
2022
detail.hit.zdb_id:
2193340-6
Permalink