1. Genetics and Genomics
  2. Microbiology and Infectious Disease

A phylogenetic transform enhances analysis of compositional microbiota data

  1. Justin D Silverman
  2. Alex D Washburne
  3. Sayan Mukherjee
  4. Lawrence A David  Is a corresponding author
  1. Duke University, United States
https://doi.org/10.7554/eLife.21887
Share this article
Cite this article
  1. Justin D Silverman
  2. Alex D Washburne
  3. Sayan Mukherjee
  4. Lawrence A David
(2017)
A phylogenetic transform enhances analysis of compositional microbiota data
eLife 6:e21887.
https://doi.org/10.7554/eLife.21887
  2. Download BibTeX
  3. Download .RIS

Abstract

Surveys of microbial communities (microbiota), typically measured as relative abundance of species, have illustrated the importance of these communities in human health and disease. Yet, statistical artifacts commonly plague the analysis of relative abundance data. Here, we introduce the PhILR transform, which incorporates microbial evolutionary models with the isometric log-ratio transform to allow off-the-shelf statistical tools to be safely applied to microbiota surveys. We demonstrate that analyses of community-level structure can be applied to PhILR transformed data with performance on benchmarks rivaling or surpassing standard tools. Additionally, By decomposing distance in the PhILR transformed space, we identified neighboring clades that may have adapted to distinct human body sites. Decomposing variance revealed that covariation of bacterial clades within human body sites increases with phylogenetic relatedness. Together, these findings illustrate how the PhILR transform combines statistical and phylogenetic models to overcome compositional data challenges and enable evolutionary insights relevant to microbial communities.

Data availability

The following previously published data sets were used
    1. Human Microbiome Project Consortium
    (2010) Human Microbiome Project
    Publicly available at HMPDACC (v35 download of files 6, 9, and 10).
    http://hmpdacc.org/HMQCP/
    1. Costello EK
    2. Lauber CL
    3. Hamady M
    4. Fierer N
    5. Gordon JI
    6. Knight R
    (2009) Costello Skin Sites
    Publicly available as part of the FEMS Benchmark dataset (2011) provided Dan Knights.
    http://www.knightslab.org/data

Article and author information

Author details

  1. Justin D Silverman

    Program in Computational Biology and Bioinformatics, Duke University, Durham, United States
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3063-2098

  2. Alex D Washburne

    Nicholas School of the Environment, Duke University, Durham, United States
    Competing interests
    The authors declare that no competing interests exist.

  3. Sayan Mukherjee

    Program in Computational Biology and Bioinformatics, Duke University, Durham, United States
    Competing interests
    The authors declare that no competing interests exist.

  4. Lawrence A David

    Program in Computational Biology and Bioinformatics, Duke University, Durham, United States
    For correspondence
    lawrence.david@duke.edu
    Competing interests
    The authors declare that no competing interests exist.
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3570-4767

Funding

Global Probiotics Council (Young Investigator Grant for Probiotics Research)

  • Lawrence A David

Searle Scholars Program (15-SSP-184 Research Agreement)

  • Lawrence A David

Alfred P. Sloan Foundation (BR2014-003)

  • Lawrence A David

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Reviewing Editor

  1. Anthony Fodor, University of North Carolina at Charlotte

Version history

  1. Received: September 27, 2016
  2. Accepted: February 13, 2017
  3. Accepted Manuscript published: February 15, 2017 (version 1)
  4. Version of Record published: February 27, 2017 (version 2)

Copyright

© 2017, Silverman et al.

This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 11,403
    views
  • 1,717
    downloads
  • 236
    citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Justin D Silverman
  2. Alex D Washburne
  3. Sayan Mukherjee
  4. Lawrence A David
(2017)
A phylogenetic transform enhances analysis of compositional microbiota data
eLife 6:e21887.
https://doi.org/10.7554/eLife.21887

Share this article

https://doi.org/10.7554/eLife.21887

Categories and tags

Research organism

Further reading

    1. Genetics and Genomics
    2. Neuroscience

    Metabolic and neurobehavioral disturbances induced by purine recycling deficiency in Drosophila

    Céline Petitgas, Laurent Seugnet ... Serge Birman
    Research Article

    Adenine phosphoribosyltransferase (APRT) and hypoxanthine-guanine phosphoribosyltransferase (HGPRT) are two structurally related enzymes involved in purine recycling in humans. Inherited mutations that suppress HGPRT activity are associated with Lesch–Nyhan disease (LND), a rare X-linked metabolic and neurological disorder in children, characterized by hyperuricemia, dystonia, and compulsive self-injury. To date, no treatment is available for these neurological defects and no animal model recapitulates all symptoms of LND patients. Here, we studied LND-related mechanisms in the fruit fly. By combining enzymatic assays and phylogenetic analysis, we confirm that no HGPRT activity is expressed in Drosophila melanogaster, making the APRT homolog (Aprt) the only purine-recycling enzyme in this organism. Whereas APRT deficiency does not trigger neurological defects in humans, we observed that Drosophila Aprt mutants show both metabolic and neurobehavioral disturbances, including increased uric acid levels, locomotor impairments, sleep alterations, seizure-like behavior, reduced lifespan, and reduction of adenosine signaling and content. Locomotor defects could be rescued by Aprt re-expression in neurons and reproduced by knocking down Aprt selectively in the protocerebral anterior medial (PAM) dopaminergic neurons, the mushroom bodies, or glia subsets. Ingestion of allopurinol rescued uric acid levels in Aprt-deficient mutants but not neurological defects, as is the case in LND patients, while feeding adenosine or N6-methyladenosine (m6A) during development fully rescued the epileptic behavior. Intriguingly, pan-neuronal expression of an LND-associated mutant form of human HGPRT (I42T), but not the wild-type enzyme, resulted in early locomotor defects and seizure in flies, similar to Aprt deficiency. Overall, our results suggest that Drosophila could be used in different ways to better understand LND and seek a cure for this dramatic disease.

    1. Genetics and Genomics

    Mono-methylated histones control PARP-1 in chromatin and transcription

    Gbolahan Bamgbose, Guillaume Bordet ... Alexei Tulin
    Research Article

    PARP-1 is central to transcriptional regulation under both normal and stress conditions, with the governing mechanisms yet to be fully understood. Our biochemical and ChIP-seq-based analyses showed that PARP-1 binds specifically to active histone marks, particularly H4K20me1. We found that H4K20me1 plays a critical role in facilitating PARP-1 binding and the regulation of PARP-1-dependent loci during both development and heat shock stress. Here, we report that the sole H4K20 mono-methylase, pr-set7, and parp-1 Drosophila mutants undergo developmental arrest. RNA-seq analysis showed an absolute correlation between PR-SET7- and PARP-1-dependent loci expression, confirming co-regulation during developmental phases. PARP-1 and PR-SET7 are both essential for activating hsp70 and other heat shock genes during heat stress, with a notable increase of H4K20me1 at their gene body. Mutating pr-set7 disrupts monomethylation of H4K20 along heat shock loci and abolish PARP-1 binding there. These data strongly suggest that H4 monomethylation is a key triggering point in PARP-1 dependent processes in chromatin.

    1. Cancer Biology
    2. Genetics and Genomics

    Germline cis variant determines epigenetic regulation of the anti-cancer drug metabolism gene dihydropyrimidine dehydrogenase (DPYD)

    Ting Zhang, Alisa Ambrodji ... Steven M Offer
    Research Article

    Enhancers are critical for regulating tissue-specific gene expression, and genetic variants within enhancer regions have been suggested to contribute to various cancer-related processes, including therapeutic resistance. However, the precise mechanisms remain elusive. Using a well-defined drug-gene pair, we identified an enhancer region for dihydropyrimidine dehydrogenase (DPD, DPYD gene) expression that is relevant to the metabolism of the anti-cancer drug 5-fluorouracil (5-FU). Using reporter systems, CRISPR genome-edited cell models, and human liver specimens, we demonstrated in vitro and vivo that genotype status for the common germline variant (rs4294451; 27% global minor allele frequency) located within this novel enhancer controls DPYD transcription and alters resistance to 5-FU. The variant genotype increases recruitment of the transcription factor CEBPB to the enhancer and alters the level of direct interactions between the enhancer and DPYD promoter. Our data provide insight into the regulatory mechanisms controlling sensitivity and resistance to 5-FU.