PhyloMix: Enhancing Microbiome-trait Association Prediction Through Phylogeny-mixing Augmentation

Overview

Journal Bioinformatics

Publisher Oxford University Press

Date 2025 Jan 12

PMID 39799515

Authors

Yifan Jiang

Disen Liao

Qiyun Zhu

Yang Young Lu

Affiliations

Soon will be listed here.

Abstract

Motivation: Understanding the associations between traits and microbial composition is a fundamental objective in microbiome research. Recently, researchers have turned to machine learning (ML) models to achieve this goal with promising results. However, the effectiveness of advanced ML models is often limited by the unique characteristics of microbiome data, which are typically high-dimensional, compositional, and imbalanced. These characteristics can hinder the models' ability to fully explore the relationships among taxa in predictive analyses. To address this challenge, data augmentation has become crucial. It involves generating synthetic samples with artificial labels based on existing data and incorporating these samples into the training set to improve ML model performance.

Results: Here, we propose PhyloMix, a novel data augmentation method specifically designed for microbiome data to enhance predictive analyses. PhyloMix leverages the phylogenetic relationships among microbiome taxa as an informative prior to guide the generation of synthetic microbial samples. Leveraging phylogeny, PhyloMix creates new samples by removing a subtree from one sample and combining it with the corresponding subtree from another sample. Notably, PhyloMix is designed to address the compositional nature of microbiome data, effectively handling both raw counts and relative abundances. This approach introduces sufficient diversity into the augmented samples, leading to improved predictive performance. We empirically evaluated PhyloMix on six real microbiome datasets across five commonly used ML models. PhyloMix significantly outperforms distinct baseline methods including sample-mixing-based data augmentation techniques like vanilla mixup and compositional cutmix, as well as the phylogeny-based method TADA. We also demonstrated the wide applicability of PhyloMix in both supervised learning and contrastive representation learning.

Availability And Implementation: The Apache-licensed source code is available at (https://github.com/batmen-lab/phylomix).

References

Rong R, Jiang S, Xu L, Xiao G, Xie Y, Liu D . MB-GAN: Microbiome Simulation via Generative Adversarial Network. Gigascience. 2021; 10(2). PMC: 7931821. DOI: 10.1093/gigascience/giab005. View

Boktor J, Sharon G, Verhagen Metman L, Hall D, Engen P, Zreloff Z . Integrated Multi-Cohort Analysis of the Parkinson's Disease Gut Metagenome. Mov Disord. 2023; 38(3):399-409. DOI: 10.1002/mds.29300. View

Turnbaugh P, Ley R, Mahowald M, Magrini V, Mardis E, Gordon J . An obesity-associated gut microbiome with increased capacity for energy harvest. Nature. 2006; 444(7122):1027-31. DOI: 10.1038/nature05414. View

Hernandez Medina R, Kutuzova S, Nielsen K, Johansen J, Hansen L, Nielsen M . Machine learning and deep learning applications in microbiome research. ISME Commun. 2023; 2(1):98. PMC: 9723725. DOI: 10.1038/s43705-022-00182-9. View

Dutta S, Verma S, Jain V, Surapaneni B, Vinayek R, Phillips L . Parkinson's Disease: The Emerging Role of Gut Dysbiosis, Antibiotics, Probiotics, and Fecal Microbiota Transplantation. J Neurogastroenterol Motil. 2019; 25(3):363-376. PMC: 6657920. DOI: 10.5056/jnm19044. View

Qin J, Li R, Raes J, Arumugam M, Burgdorf K, Manichanh C . A human gut microbial gene catalogue established by metagenomic sequencing. Nature. 2010; 464(7285):59-65. PMC: 3779803. DOI: 10.1038/nature08821. View

Albanese D, De Filippo C, Cavalieri D, Donati C . Explaining diversity in metagenomic datasets by phylogenetic-based feature weighting. PLoS Comput Biol. 2015; 11(3):e1004186. PMC: 4376673. DOI: 10.1371/journal.pcbi.1004186. View

Zhu Q, Hou Q, Huang S, Ou Q, Huo D, Vazquez-Baeza Y . Compositional and genetic alterations in Graves' disease gut microbiome reveal specific diagnostic biomarkers. ISME J. 2021; 15(11):3399-3411. PMC: 8528855. DOI: 10.1038/s41396-021-01016-7. View

Gonzalez A, Navas-Molina J, Kosciolek T, McDonald D, Vazquez-Baeza Y, Ackermann G . Qiita: rapid, web-enabled microbiome meta-analysis. Nat Methods. 2018; 15(10):796-798. PMC: 6235622. DOI: 10.1038/s41592-018-0141-9. View

10.

Reiman D, Metwally A, Sun J, Dai Y . PopPhy-CNN: A Phylogenetic Tree Embedded Architecture for Convolutional Neural Networks to Predict Host Phenotype From Metagenomic Data. IEEE J Biomed Health Inform. 2020; 24(10):2993-3001. DOI: 10.1109/JBHI.2020.2993761. View

11.

Lozupone C, Knight R . UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol. 2005; 71(12):8228-35. PMC: 1317376. DOI: 10.1128/AEM.71.12.8228-8235.2005. View

12.

Knights D, Parfrey L, Zaneveld J, Lozupone C, Knight R . Human-associated microbial signatures: examining their predictive value. Cell Host Microbe. 2011; 10(4):292-6. PMC: 3879110. DOI: 10.1016/j.chom.2011.09.003. View

13.

Zhu Q, Mai U, Pfeiffer W, Janssen S, Asnicar F, Sanders J . Phylogenomics of 10,575 genomes reveals evolutionary proximity between domains Bacteria and Archaea. Nat Commun. 2019; 10(1):5477. PMC: 6889312. DOI: 10.1038/s41467-019-13443-4. View

14.

. The Integrative Human Microbiome Project: dynamic analysis of microbiome-host omics profiles during periods of human health and disease. Cell Host Microbe. 2014; 16(3):276-89. PMC: 5109542. DOI: 10.1016/j.chom.2014.08.014. View

15.

Vogt N, Kerby R, Dill-McFarland K, Harding S, Merluzzi A, Johnson S . Gut microbiome alterations in Alzheimer's disease. Sci Rep. 2017; 7(1):13537. PMC: 5648830. DOI: 10.1038/s41598-017-13601-y. View

16.

Washburne A, Morton J, Sanders J, McDonald D, Zhu Q, Oliverio A . Methods for phylogenetic analysis of microbiome data. Nat Microbiol. 2018; 3(6):652-661. DOI: 10.1038/s41564-018-0156-0. View

17.

Gonzalez C, Mills R, Zhu Q, Sauceda C, Knight R, Dulai P . Location-specific signatures of Crohn's disease at a multi-omics scale. Microbiome. 2022; 10(1):133. PMC: 9400277. DOI: 10.1186/s40168-022-01331-x. View

18.

Sharma D, Paterson A, Xu W . TaxoNN: ensemble of neural networks on stratified microbiome data for disease prediction. Bioinformatics. 2020; 36(17):4544-4550. PMC: 7750934. DOI: 10.1093/bioinformatics/btaa542. View

19.

McCoubrey L, Elbadawi M, Orlu M, Gaisford S, Basit A . Harnessing machine learning for development of microbiome therapeutics. Gut Microbes. 2021; 13(1):1-20. PMC: 7872042. DOI: 10.1080/19490976.2021.1872323. View

20.

Laske C, Muller S, Preische O, Ruschil V, Munk M, Honold I . Signature of Alzheimer's Disease in Intestinal Microbiome: Results From the AlzBiom Study. Front Neurosci. 2022; 16:792996. PMC: 9063165. DOI: 10.3389/fnins.2022.792996. View