Vol. 28 no. 3 2012, pages 436–438
SitePainter: a tool for exploring biogeographical patterns
Antonio Gonzalez1,∗, Jesse Stombaugh2, Christian L. Lauber3,4, Noah Fierer3,4
and Rob Knight2,5,∗
1Department of Computer Science,2Department of Chemistry and Biochemistry,3Cooperative Institute for Research
in Environmental Sciences,4Department of Ecology and Evolutionary Biology, University of Colorado at Boulder and
5Howard Hughes Medical Institute, Boulder, CO 80309, USA
Associate Editor: Martin Bishop
Advance Access publication December 9, 2011
As microbial ecologists take advantage of high-throughput analytical
increasing numbers of samples, the need for new analysis
tools that reveal the intrinsic spatial patterns and structures of
these populations is crucial. Here we present SitePainter, an
interactive graphical tool that allows investigators to create or
upload pictures of their study site, load diversity analyses data
and display both diversity and taxonomy results in a spatial
context. Features of SitePainter include: visualizing α-diversity,
using taxonomic summaries; visualizing β-diversity, using results
from multidimensional scaling methods; and animating relationships
among microbial taxa or pathways overtime. SitePainter thus
increases the visual power and ability to explore spatially explicit
Supplementary information: Supplementary data are available at
Contact: email@example.com, Rob.Knight@colorado.edu
Received on September 19, 2011; revised on November 30, 2011;
accepted on December 4, 2011
As sequencing capacity increases, microbial community analysis
is undergoing a revolution in throughput. Where it was once a
monumental task just to analyze microbial communities in a handful
of samples, we can now process thousands of samples in a single
sequencing run (Caporaso et al., 2011). Accordingly, instead of
describing those communities in a few samples collected under a
few specific conditions, we can now perform comprehensive spatial
sampling, exploring interactions among members of the microbial
communities and revealing large-scale spatial dynamics (Brodie
et al., 2007; Caporaso et al., 2011; DeLong 2009; Gilbert et al.,
2010; Gonzalez et al., 2011).
However, the power of these analyses is often limited by our
ability to visualize the data. When analyzing biogeographical data,
researchers would like to know where specific microbes live, how
abundant they are and whether there are any notable patterns
relating microbial taxa to each other or to environmental conditions.
Traditional ordination methods for visualizing large numbers of
microbial communities such as Principal Coordinates Analysis
∗To whom correspondence should be addressed.
(PCoA) (Gower and Legendre, 1986) are extremely useful for
reducing the dimensionality of vast multivariate datasets, but the
patterns are often unclear, especially when the results do not map
easily onto the sampling structure. For example, it may not be clear
surface of a complex shape without closely examining the sample
identifiers (Fig. 1C1).
To address this key barrier to microbial community research,
we developed SitePainter that provides compelling visualizations
that allow the researcher to better explore spatial patterns in
microbial communities. SitePainter allows the user to draw or
import representative images for biogeographical sites [in the
example, we show a human hand (Fig. 1B–C), although any
picture from micro- to landscape-scale is possible] and regions
within that site (e.g. a finger, palm, finger tips, etc.). Once the
biogeographical image is drawn, the user can import tab-delimited
text files, such as those produced by Quantitative Insights Into
Microbial Community (QIIME) (Caporaso et al., 2010) to map
onto their image. For example, it is often useful to plot the α-
diversity (the number of species present in each sample), taxonomy
summaries (the abundance of particular kinds of species in the
sample) and/or β-diversity (the overall similarities and differences
SitePainter is a software package with an interactive user interface,
Vector Graphics (SVG) images and annotates them with per sample
data defined by the user. As an example of usage, we show the
biogeography of the human hand and assign bacterial information
based on QIIME-formatted files.The main work areas of SitePainter
are: (i) file image manipulation that allows the user to load and
save SVG images; (ii) the metadata loading and processing menu,
where the user can select the metadata (tab-delimited format) to be
displayed in combination with the images; (iii) the coloring scheme,
where the user can select how to represent low and high values, reset
the image color or show/hide the display of lines and absent paths,
to show more compelling images; and (iv) the interactive menu that
allows the user to work with previous selections to find patterns that
may not have been apparent when using other visualization tools
and techniques (Fig. 1A).
To illustrate the effectiveness of SitePainter, we show the results
of a study in which different regions from a subject’s hand were
sampled for microbes. The subject’s hand was outlined, and each
© The Author(s) 2011. Published by Oxford University Press.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/
by-nc/3.0), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Fig. 1. SitePainter illustration of microbial community patterns in a human hand. (A) SitePainter user interface: (1) file image manipulation, (2) metadata
loading and processing menu, (3) coloring scheme and (4) interactive menu. (B) The α-diversity analysis showing relative abundance of bacterial taxa on the
hand: low values in blue an high values in red (1) Gammaproteobacteria, (2) Bacilli and (3) Actinomycetales. (C) The β-diversity analysis showing overall
similarities and differences among samples: (1) 3D PCoA axis where each point represents a sample and each sample is colored independently and (2) first
two axes of the PCoA analysis, where similar colors represent samples similar to each other along a given axis in the abstract ordination space, with low
values in blue and high values in red.
region divided into smaller areas that were then sampled using
cotton swabs dipped in saline solution. Using SitePainter, we were
able to rapidly deduce which bacteria were in different regions
of the hand, along with a graphical representation of microbial
abundances by processing the sequence data with QIIME (Caporaso
et al., 2010) and mapping the abundance of each taxon onto the
hand image (Fig. 1B). Using this display, we can immediately
see that Gammaproteobacteria is present in the palm, Bacilli are
more visible toward the fingertips and Actinomycetales is present
in only one area of the hand. This feature thus allows the user to
easily visualize the distribution and abundance of microbes across
Going beyond the abundances of individual microbial taxa,
SitePainter also allows the user to view similarities and differences
at the whole-community level by loading the PCoA axis of the
microbial data to reveal patterns that are not obvious in the PCoA
plot but are immediately obvious when displayed in the context of
the site itself (Fig. 1C: compare left panel to the right two panels.
In this case, we can see a clear gradient from the thumb and index
fingers to the left-bottom corner of the hand; additionally we can
see that the distal phalanges are similar, maybe due to constant
In addition to analyses of microbial taxa, as shown here,
SitePainter can also be used to interpret the abundance of genes
or pathways across sites for metabolic studies. Several further
examples can be found in the Human Microbiome Project web
site: http://hmpdacc.org/sp/. These examples analyze the entire
Human Microbiome Project (Peterson et al., 2009; Turnbaugh
et al., 2007) dataset at several taxonomic levels, and at several
different levels of functional classification for shotgun metagenomic
reads. We expect that this will be a useful community resource
for those trying to interpret the complexities of the human
SitePainter provides a user-friendly tool that allows rapid, clear
visualization of microbial data on arbitrary user-supplied images.
Additionally, it contains several user interface features, such as
rapid switching between taxa/coordinates and animations of e.g.,
of spatial data. We believe that SitePainter will have a large impact
on the field, especially in the case of diseases with complex spatial
structure such as psoriasis and ulcerative colitis, and also in large-
Project (Gilbert et al., 2010).
We thank Dirk Gevers, Curtis Huttenhower, Owen White, Lita
Proctor and Jesse Zaneveld for feedback and assistance with the
HMPimplementation of SitePainter, and Dan Sisco, ClotildeTeiling
and Chinnappa Kodira at 454 Life Sciences and personnel at
the company for carrying out the sequencing associated with this
Funding: National Institutes of Health (Grants 5-R01-HL0090480
& 3-R01-HG004872); the Crohns and Colitis Foundation of
America (Grant WU-09-72/2905035N); The Bill & Melinda Gates
Foundation (Grant WU-09-187/51678); and the Howard Hughes
Conflict of Interest: none declared.
Brodie,E.L. et al. (2007) Urban aerosols harbor diverse and dynamic bacterial
populations. Proc. Natl Acad. Sci. USA, 104, 299–304.
A.Gonzalez et al.
Caporaso,J.G. et al. (2010) QIIME allows analysis of high-throughput community
sequencing data. Nat. Methods, 7, 335–336.
Caporaso,J.G. et al. (2011) Moving pictures of the human microbiome. Genome Biol.,
DeLong,E.F. (2009) The microbial ocean from genomes to biomes. Nature, 459,
Gilbert,J.A. et al. (2010) The taxonomic and functional diversity of microbes at a
temperate coastal site: a ‘multi-omic’study of seasonal and diel temporal variation.
PLoS One, 5, e15545.
Gilbert,J.A. et al. (2010) Meeting report: the terabase metagenomics workshop and the
vision of an Earth microbiome project. Stand. Genomic Sci., 3, 243–248.
Gonzalez,A. et al. (2011) Our microbial selves: what ecology can teach us. EMBO Rep.,
Gower,J.C. and Legendre,P. (1986) Metric and Euclidean properties of dissimilarity
coefficients, J. Classification, 3, 5–48.
Peterson,J. et al. (2009) The NIH Human Microbiome Project. Genome Res., 19,
Turnbaugh,P.J. et al. (2007) The human microbiome project. Nature, 449, 804–810.