Metagenomics enables the study of gene abundances in complex mixtures of microorganisms and has become a standard methodology for the analysis of the human microbiome. However, gene abundance data is inherently noisy and contains high levels of biological and technical variability as well as an excess of zeros due to non-detected genes. This makes the statistical analysis challenging. In this study, we present a new hierarchical Bayesian model for inference of metagenomic gene abundance data. The model uses a zero-inflated overdispersed Poisson distribution which is able to simultaneously capture the high gene-specific variability as well as zero observations in the data. By analysis of three comprehensive datasets, we show that zero-inflation is common in metagenomic data from the human gut and, if not correctly modelled, it can lead to substantial reductions in statistical power. We also show, by using resampled metagenomic data, that our model has, compared to other methods, a higher and more stable performance for detecting differentially abundant genes. We conclude that proper modelling of the gene-specific variability, including the excess of zeros, is necessary to accurately describe gene abundances in metagenomic data. The proposed model will thus pave the way for new biological insights into the structure of microbial communities.
HandelsmanJRondonMBradyS, et al.Molecular biological access to the chemistry of unknown soil microbes: a new frontier for natural products. Chem Biol1998; 5: R245–R249. .
2.
SchlossPHandelsmanJ. Metagenomics for studying unculturable microorganisms: cutting the Gordian knot. Genome Biol2005; 6: 229.
3.
AlbertsenMHugenholtzPSkarshewskiA, et al.Genome sequences of rare, uncultured bacteria obtained by differential coverage binning of multiple metagenomes. Nature Biotechnol2013; 31: 533–538.
4.
Human Microbiome Project Consortium. Structure, function and diversity of the healthy human microbiome. Nature2012; 486: 207–214.
5.
KarlssonFTremaroliVNookaewI, et al.Gut metagenome in European women with normal, impaired and diabetic glucose control. Nature2013; 498: 99–103.
6.
PalCBengtsson-PalmeJKristiansonE, et al.The structure and diversity of human, animal and environmental resistomes. Microbiome2016; 4(54): 1–15. .
7.
BhattAPRedinboMRBultmanSJ. The role of the microbiome in cancer development and therapy. CA: Cancer J Clin2017; 67: 326–344. .
8.
FiererNLadauJClementeJC, et al.Reconstructing the microbial diversity and function of pre-agricultural tallgrass prairie soils in the United States. Science2013; 342: 621–624.
9.
SunagawaSCoelhoLPChaffronS, et al.Structure and function of the global ocean microbiome. Science2015; 348: 1261359.
10.
GibbonsSMJonesEBearquiverA, et al.Human and environmental impacts on river sediment microbial communities. PLoS One2014; 9: e97435.
11.
KuczynskiJLauberCLWaltersWA, et al.Experimental and analytical tools for studying the human microbiome. Nat Rev Genet2012; 13: 47–58.
12.
BoulundFSjögrenAKristianssonE. Tentacle: distributed quantification of genes in metagenomes. GigaScience2015; 4: 1.
13.
HatemABozdağDTolandAE, et al.Benchmarking short sequence mapping tools. BMC Bioinform2013; 14: 1.
14.
DinsdaleEAEdwardsRAHallD, et al.Functional metagenomic profiling of nine biomes. Nature2008; 452: 629–632.
15.
YatsunenkoTReyFManaryM, et al.Human gut microbiome viewed across age and geography. Nature2012; 486: 222–227.
16.
HaftDHSelengutJDRichterRA, et al.Tigrfams and genome properties in 2013. Nucleic Acid Res2013; 41: D387–D395.
17.
Huerta-CepasJSzklarczykDForslundK, et al.eggnog 4.5: a hierarchical orthology framework with improved functional annotations for eukaryotic, prokaryotic and viral sequences. Nucl Acid Res2015; 44: D289–D293. .
18.
FinnRDCoggillPEberhardtRY, et al.The Pfam protein families database: towards a more sustainable future. Nucleic Acid Res2016; 44: D279–D285.
19.
Österlund T, Jonsson V and Kristiansson E. Hirbin: High-resolution identification of differentially abundant functions in metagenomes. BMC Genomics 2016; 18(316): 1–11.
20.
BoulundFPereiraMBJonssonV, et al.Computational and statistical considerations in the analysis of metagenomic data. In: Nagarajan M (ed) Metagenomics, London, UK: Elsevier, 2018, pp. 81–102. .
21.
KnightRJanssonJFieldD, et al.Unlocking the potential of metagenomics through replicated experimental design. Nat Biotechnol2012; 30: 513–520.
22.
NayfachSPollardKS. Toward accurate and quantitative comparative metagenomics. Cell2016; 166: 1103–1116.
23.
Jonsson V, Österlund T, Nerman O, et al. Variability in metagenomic count data and its influence on the identification of differentially abundant genes. J Computat Biol 2016 (in preparation).
24.
WhiteJNagarajanNPopM. Statistical methods for detecting differentially abundant features in clinical metagenomic samples. PLoS Computat Biol2009; 5: e1000352.
25.
PaulsonJNStineOCBravoHC, et al.Differential abundance analysis for microbial marker-gene surveys. Nat Meth2013; 10: 1200–1202.
26.
ChenELiH. A two-part mixed-effects model for analyzing longitudinal microbiome compositional data. Bioinformatics2016; 32: 2611–2617. .
27.
JonssonVÖsterlundTNermanO, et al.Statistical evaluation of methods for identification of differentially abundant genes in comparative metagenomics. BMC Genomics2016; 17: 1.
28.
SohnMBDuRAnL. A robust approach for identifying differentially abundant features in metagenomic samples. Bioinformatics2015; 31: 2269–2275. .
29.
SmythG. Linear models and empirical Bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol2004; 3: 1–25. .
30.
FangRWagnerBHarrisJ, et al.Zero-inflated negative binomial mixed model: an application to two microbial organisms important in oesophagitis. Epidemiol Infect2016; 144: 1–9. .
31.
PengXLiGLiuZ. Zero-inflated beta regression for differential abundance analysis with metagenomics data. J Computat Biol2016; 23: 102–110.
32.
OhJByrdALDemingC, et al.Biogeography and individuality shape function in the human skin metagenome. Nature2014; 514: 59–64.
33.
PereiraMWallrothMJonssonV, et al.Comparison of normalization methods for the analysis of metagenomic gene abundance data. BMC Genomics2018; 19(274): 1–17. .
34.
AitchisonJHoC. The multivariate Poisson-log normal distribution. Biometrika1989; 76: 643–653.
35.
McMurdiePJHolmesS. Waste not, want not: why rarefying microbiome data is inadmissible. PLoS Comput Biol2014; 10: e1003531.
36.
SonesonCDelorenziM. A comparison of methods for differential expression analysis of rna-seq data. BMC Bioinform2013; 14: 1.
37.
RobinsonMMcCarthyDSmythG. edger: A bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics2010; 26: 139–140.
38.
LoveMIHuberWAndersS. Moderated estimation of fold change and dispersion for rna-seq data with deseq2. Genome Biol2014; 15: 1–21.
39.
GelmanACarlinJSternH, et al.Bayesian data analysis, 2nd ed. New York: Chapman & Hall, 2003. .
40.
QinJLiRRaesJ, et al.A human gut microbial gene catalogue established by metagenomic sequencing. Nature2010; 464: 59–65.
41.
QinJLiYCaiZ, et al.A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature2012; 490: 55–60.
MeyerFPaarmannDD'SouzaM, et al.The metagenomics rast server – a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinform2008; 9(386): 1–8. .
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
