Patterns and Processes of Microbial Community Assembly

101128MMBR00051-12

2013 77(3)342 DOIMicrobiol Mol Biol Rev and Scott FerrenbergE Knelman John L Darcy Ryan C Lynch Phillip WickeySean P ONeill Teresa M Bilinski Lee F Stanish Joseph Diana R Nemergut Steven K Schmidt Tadashi Fukami Community AssemblyPatterns and Processes of Microbial

httpmmbrasmorgcontent773342Updated information and services can be found at

These include

REFERENCEShttpmmbrasmorgcontent773342ref-list-1at

This article cites 151 articles 40 of which can be accessed free

CONTENT ALERTS moreraquoarticles cite this article)

Receive RSS Feeds eTOCs free email alerts (when new

httpjournalsasmorgsitemiscreprintsxhtmlInformation about commercial reprint orders httpjournalsasmorgsitesubscriptionsTo subscribe to to another ASM Journal go to

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Patterns and Processes of Microbial Community Assembly

Diana R Nemergutab Steven K Schmidtc Tadashi Fukamid Sean P OrsquoNeillac Teresa M Bilinskiac Lee F Stanishab

Joseph E Knelmanac John L Darcyc Ryan C Lynchc Phillip Wickeyab Scott Ferrenbergc

Institute of Arctic and Alpine Research (INSTAAR)a Environmental Studies Programb and Ecology and Evolutionary Biologyc University of Colorado Boulder ColoradoUSA Department of Biology Stanford University Stanford California USAd

SUMMARY 342INTRODUCTION 342HOW ARE MICROBES UNIQUE 343DEFINITIONS 343

Communities 343Biodiversity 344

BIOGEOGRAPHICAL PATTERNS 345Abundance 345Taxon Turnover 345Phylogenetic Structure 346

VELLENDrsquoS CONCEPTUAL SYNTHESIS OF COMMUNITY ECOLOGY 346SELECTION 347DISPERSAL 347DIVERSIFICATION 348DRIFT 348COMBINING FORCES COMMUNITY ASSEMBLY 348TEMPORAL AND SPATIAL SCALES 349IMPLICATIONS FOR FUNCTION 350IMPLICATIONS FOR BIODIVERSITY 350ACKNOWLEDGMENTS 351REFERENCES 351AUTHOR BIOS 355

SUMMARY

Recent research has expanded our understanding of microbial com-munity assembly However the field of community ecology is inac-cessible to many microbial ecologists because of inconsistent and of-ten confusing terminology as well as unnecessarily polarizing debatesThus we review recent literature on microbial community assemblyusing the framework of Vellend (Q Rev Biol 85183ndash206 2010) inan effort to synthesize and unify these contributions We begin bydiscussing patterns in microbial biogeography and then describe fourbasic processes (diversification dispersal selection and drift) thatcontribute to community assembly We also discuss different combi-nations of these processes and where and when they may be mostimportant for shaping microbial communities The spatial and tem-poral scales of microbial community assembly are also discussed inrelation to assembly processes Throughout this review paper wehighlight differences between microbes and macroorganisms andgenerate hypotheses describing how these differences may be impor-tant for community assembly We end by discussing the implicationsof microbial assembly processes for ecosystem function and biodiver-sity

INTRODUCTION

Molecular phylogenetic approaches continue to revolutionizethe field of microbiology we now possess the tools to un-

derstand high-resolution details about the degree of variation inmicrobial community structure in both space and time (1ndash5)Sequencing costs have plummeted while the amount of publiclyavailable data has increased exponentially in recent years Com-

putational advances (6 7) as well as new standards for contextu-alizing environmental microbial community composition datasets (8) will allow us to make the most of these data facilitatingcross-investigator and cross-system meta-analyses Indeed afteryears of citing the many limitations of studying such complexsystems microbiologists now enjoy many advantages that our col-leagues who study macrobial communities actually lack Admit-tedly we are still a long way from a ldquocompleterdquo understanding ofany but the most simple of microbial communities which willrequire continual improvements in both technology and compu-tation Thus despite these recent advances we are faced withquestions about how to best sample microbial communities tomaximize what we can learn about how they are structured howthey function and how they change through time (9)

A unified conceptual framework of microbial community as-semblymdash one that incorporates our understanding of communityassembly from a macrobial ecology perspective while recognizingthe attributes that make microorganisms uniquemdashis needed tohelp direct the field of microbial ecology through this new eraThis is not an easy task and we argue that it is made more difficultby unnecessarily polarizing debates (eg the false dichotomy ofthe niche-versus-neutral debate [10] as well as the debates over

Address correspondence to Diana R Nemergut nemergutcoloradoedu

doi101128MMBR00051-12

342 mmbrasmorg Microbiology and Molecular Biology Reviews p 342ndash356 September 2013 Volume 77 Number 3

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

null models [11]) as well as the use of inconsistent and sometimesredundant terminology (eg niche based deterministic environ-mental filters and stabilizing mechanisms which all refer to sim-ilar phenomena) However we believe that Vellendrsquos (5) concep-tual synthesis of community ecology which distills the myriad ofprocesses affecting community assembly into four basic categories(diversification dispersal selection and drift) and can be appliedon different temporal and spatial scales is a step in the right di-rection (see also reference 12) The purpose of this review is tointegrate microorganisms into this simple framework with thehope of providing microbiologists with a coherent picture of thepotential mechanisms governing microbial community assembly

Here we begin with a general discussion of the features thatmake microorganisms unique and definitions of the relevantterms used to describe community structure Next we describethe patterns observed in terms of microbial community composi-tion across systems highlighting the many parallels between mi-crobial and macrobial biogeography We then describe Vellendrsquosconceptual synthesis of community ecology which we use toframe our discussion of how diversification dispersal selectionand drift affect microbial community assembly We also describesome examples of how these processes combine to affect commu-nity assembly Next we highlight the importance of spatial andtemporal dynamics in assembly processes We end by discussingthe relationships between community assembly and microbialfunction and biodiversity In this review we deal almost exclu-sively with molecular-based analyses of microbial communitycomposition As noted above these approaches have tremen-dously expanded our appreciation of microbial diversity andcommunity complexity over past culture-based studies but theyare not without their own limitations (13ndash15)

HOW ARE MICROBES UNIQUE

There is copious literature on the mechanisms governing macro-bial community assembly and given the similarities in the ob-served biogeographical patterns (see below) we use this literatureas a starting point to consider the processes guiding microbialcommunity assembly However there are some differences be-tween micro- and macroorganisms that could lead to disparities inthe importance of different processes to the community ecology ofmicrobes It should be emphasized that several of these traits arealso shared to some degree with at least some members of themacrobial world and that not all microbial taxa exhibit these phe-notypes However we highlight these traits here because somemicrobial taxa exhibit all of these traits and the extent to whichthese phenotypes are expressed has the potential to affect assemblydynamics in any community

First the passive dispersal (eg via wind and air) of microbes islikely to be easier than that of macrobes simply because of theirsmall size (16ndash19) For example Finlay (17) used morphologicalcharacterizations of free-living protozoa to hypothesize that mi-crobes with lengths of 2 mm are likely to be globally distributedMore recently Wilkinson and colleagues (19) used computermodels of atmospheric circulation to simulate the airborne dis-persal of different-sized microorganisms They found that parti-cles 20 m in diameter were unlikely to be dispersed betweencontinents within 1 year while particles 50 m in diameter werenever passively transferred between continents While active dis-persal is rare in microbes (20) microorganisms hitch rides on allmacrobes that disperse Thus the dispersal potential of most mi-

croorganisms is likely to be much larger than that of macroorgan-isms

Likewise many microorganisms can experience dormancy inwhich they enter a reversible state of reduced activity in responseto environmental stressors (21 22) Indeed several examples ofcells being revived after decades to millennia exist in the literature(22) Although this is true of some macrobes as well the phenom-enon is more phylogenetically widespread (ie not limited to aspecific clade or clades) for microorganisms (22) It has been es-timated that less than 10 of a typical microbial community maybe active at any one time (23) thus the dormant componentpotentially represents a vast reservoir of genetic diversity

Also some microbes exhibit significant phenotypic plasticitywith regard to the use of electron donors and acceptors (24 25)and the genetic diversity present within a population of a givenldquospeciesrdquo can be large (26ndash28) Consider a typical macrobial com-munity which is composed of organisms that can either photo-synthesize or undergo heterotrophic respiration Microorganismsperform these functions as well but can use a remarkable suite ofadditional electron donors and acceptors including H- Fe- S-and N-based compounds This diversity coupled with short gen-eration times can allow ecological responses to shifting environ-mental gradients on time scales (hours to days) not attainable bymacrobes (29) Finally microbes can undergo rapid evolution(30) and some can exchange genetic material readily even withdistantly related organisms (31) All of these attributes are likely toaffect the ecology and evolution of microbes in ways that maydistinguish them from macrobial communities

In addition to the biological differences highlighted abovethere are some artifacts introduced by the ways in which we studymicrobes that will make an understanding of their communityecology more difficult In most cases we know little about thespatial and temporal structure of ecological communities onscales that are relevant to microbesmdashwhich organisms are inter-acting with others for example Additionally given our generalsampling approaches we are almost always answering questionsabout communities on the macro scale Consider for example theextreme chemical gradients that exist within a single soil particle(32) When we sample a gram of soil we are averaging across all ofthis variation and attempting to uncover patterns between com-munity structure and biogeochemistry at a large scale relative tothe size of an organism Moreover because our approaches typi-cally focus on the sequencing of pools of 16S rRNA genes it ismuch more difficult to link traits and taxonomy than it is formacroorganisms Methodological developments that allow for thecharacterization of microbial community structure at finer spatialscales including catalyzed reporter deposition-fluorescence insitu hybridization (CARD-FISH) (33) and nano-secondary-ionmass spectrometry (nanoSIMS) (34) are steps in the right direc-tion to overcome these challenges

DEFINITIONS

Communities

We define a community as a group of potentially interacting spe-cies that cooccur in space and time (35) (Table 1) Investigatorswho study macrobial communities recognize that communitiesare not discrete and that their boundaries may vary both spatiallyand temporally (36) As highlighted above some of these compli-cations may be heightened for microbes given their potential for

Microbial Community Assembly

September 2013 Volume 77 Number 3 mmbrasmorg 343

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

rapid turnover With regard to geography community definitionsare often arbitrary and reflect our inability to sample at a level thatis both small enough to be relevant for microorganisms and largeenough to be relevant to ecosystem processes (eg 1 g of soil)While communities are not restricted to taxonomic groups (egplants animals or microbes) they are often studied in this man-ner because of logistical constraints (37) For example althoughan understanding of macrobial-microbial interactions is desiredfor a more integrated understanding of community ecology thevast differences in scales of the factors affecting these organismsalso introduce statistical complications into data analysis (38)

Indeed Fauth et al (39) suggested a classification scheme inwhich they define an array of community-related terms based onall possible combinations of shared geography phylogeny andorresources (Table 1) In their scheme ldquocommunitiesrdquo are definedby geography and not by phylogeny or resource use In terms ofphylogeny what we refer to as ldquomicrobial communitiesrdquo are oftenjust bacterial communities as our understanding of membershipis typically restricted to the use of certain molecular probes (ldquouni-versalrdquo bacterial primers for example) Moreover different DNAextraction techniques are more effective for different organisms(40) which can complicate our appreciation of community struc-ture Thus in the suggested terms of Fauth et al most studiesexamine ldquomicrobial (bacterial) assemblagesrdquo or a phylogeneti-cally defined group of organisms that cooccur in space and timeDepending on the tools usedmdash denitrifying gene probes for ex-amplemdashmicrobiologists also occasionally focus on guilds orgroups of organisms that share common resources

For all of these organizationally and operationally defined units(eg community and assemblage) (Table 1) we use the term rich-ness to refer to the number of taxonomic units (eg species) in acommunity standardized for the number of individuals sampled(41) By contrast we use the term structure to describe the com-position of taxonomic units present as well as their relative abun-dances It is worth noting an inherent assumption of the molecu-lar methods typically used to describe microbial communitiesspecifically that one 16S rRNA gene equals one organism a rela-tionship which can vary by an order of magnitude between differ-ent bacterial taxa (42) However these methods also enable phy-logenetic-based examinations of richness and structure

Biodiversity

Quantifying and comparing biodiversity allow us to tease apartthe effects of different ecological processes on community struc-ture There are many different ways to assess biodiversity (35) but

all break down into two general classes either inventory diversityor differentiation diversity Inventory diversity metrics describediversity within an environment (alpha diversity sensu Whittaker[43]) while differentiation diversity describes the turnover in di-versity between environments (beta diversity [see reference 44 fora recent review]) Thus a community displaying high inventorydiversity harbors high biodiversity within a habitat at a definedspatial scale while two distinct communities exhibiting high dif-ferentiation diversity have relatively few species in common

There are a variety of inventory diversity metrics that describebiodiversity using a suite of parameters All consider the numberof different types of taxa present in a given sample and others alsoinclude information on the evenness in terms of relative abun-dance (eg Shannon index and heterogeneity measures) Still oth-ers take into account the amount of phylogenetic diversity (PD)within samples which may be particularly important for diversemicrobial communities (45 46) It is beyond the scope of thispaper to describe these metrics in detail but others have com-mented extensively on the strengths and weaknesses of variousdiversity metrics in general (35) and more specifically as they re-late to microbial communities (47ndash49)

Importantly inventory diversity metrics can be applied to ex-amine biodiversity at any scale Typically alpha diversity some-times called ldquolocal diversityrdquo refers to diversity at the smallestspatial scale of analysis and gamma diversity is a metric for re-gional (landscape) diversity These scales (and their definitions)are subjective and defined by the researcher However because ofthe near-universal relationship between species richness and area(see below) it is essential that samples of the same size be analyzedto facilitate meaningful comparisons between systems Also un-dersampling can lead to issues for diversity comparisons andmathematical methods to overcome these problems have beenproposed (50)

Likewise there are a variety of methods to assess differentiationdiversity The Z value which is the slope of the relationship be-tween the log of area versus the log of species richness is a geo-graphically explicit method with which to examine species turn-over for hierarchically sampled communities It is important tonote that as is the case with inventory diversity differentiationdiversity is also sensitive to sampling intensity (159) Some met-rics feature pairwise comparisons between samples relative to to-tal diversity and the degree of overlap in community structure isrepresented on a scale of 0 to 1 as either a dissimilarity (sometimesdistance) or similarity value Like inventory diversity metrics dif-

TABLE 1 Definitions of community-related terms used in this paper

Term Definition Reference(s)

Community A group of potentially interacting species that cooccur in space and time 35Assemblage A phylogenetically defined group of organisms that cooccur in space and time 39Guild Groups of organisms that cooccur in space and time and that share common resources 39Richness No of taxonomic units (eg species) in a community standardized to the no of individuals sampled 41Structure No of taxonomic units (eg species) in a community as well as their relative abundances This paperInventory diversity Diversity within an environment 43Differentiation diversity Turnover in diversity between environments 43Alpha diversity An inventory metric that expresses the amount of local diversity or diversity at the smallest spatial scale of

analysis gamma diversity is a similar metric used to refer to diversity at the regional scale35 43

Beta diversity A differentiation diversity metric that refers to turnover at the landscape scale ie turnover between localpopulations

Nemergut et al

344 mmbrasmorg Microbiology and Molecular Biology Reviews

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ferentiation diversity metrics can take into account relative abun-dance as well as the degree of phylogenetic overlap between twocommunities (51 52) Again differentiation diversity can be cal-culated at a variety of spatial scales at the discretion of the re-searcher Beta diversity is typically used to refer to either speciesturnover or the difference in species composition between sam-pling sites at the regional scale (ie alpha diversity relative togamma diversity [44 53])

As described above both classes of diversity metrics are inter-related Beta diversity can be affected by changes in both alpha andgamma diversity for example an increase in beta diversity canreflect both a decrease in alpha diversity as well as an increase ingamma diversity An understanding of what is driving changes indiversity can be central to an appreciation of community assemblyprocesses (54ndash56)

BIOGEOGRAPHICAL PATTERNS

After several decades of using molecular phylogenetic tools to ex-amine microbial community composition we now know thatthere are similarities in biogeographical patterns in macrobial andmicrobial communities (20 57) Although there are still manyquestions about how our past and current methodological limita-tions may affect our observations we review some of the commonpatterns that are observed in the microbial world below It is alsoworth noting that there are some patterns that have been shownfor many macrobial systems that have not been shown for mi-crobes including relationships between latitude and diversity aswell as elevation and diversity (58ndash60) It is unknown why thesedifferences exist but for example it could be the case that latitudeserves as a proxy for another driver of macrobial community com-position that does not covary with latitude at the same scale formicrobes However these disparities could also reflect real differ-ences between assembly mechanisms and biogeography in mac-robial and microbial communities

Abundance

Nearly all communities examined to date feature species abun-dance distributions (SADs) in which the majority of taxa tend tobe found in low relative abundances (we avoid the use of the termldquorarerdquo here as it has been used to refer to both organisms with lowabundance and patchy distribution patterns) and only a few aremore abundant (35 61) Much discussion in the literature hasfocused on the statistical shape of this relationship (eg log nor-mal versus geometric) as well as the possible biological mecha-nisms driving these relationships Microbial communities are noexception to this rule although they tend to show a longer ldquotailrdquo oflow-abundance species (62) possibly because of the relative scaleover which we examine microbial versus macrobial communities(Fig 1)

Although there is much debate on the definition of and thepotential methodological and computational artifacts associatedwith our understanding of the presence of these low-abundanceorganisms (eg chimeras generated by PCR and overestimationof diversity with certain OTU [operational taxonomic unit]-clus-tering algorithms) it is clear that microbial communities tend toharbor a great number of low-abundance taxa many of whichmay be inactive For example Hubert and colleagues (63) foundhyperthermophilic microorganisms in cold deep-sea sedimentsamples that became active in laboratory experiments after tem-peratures were raised to 50degC Statistical approaches reveal that

removing data on low-abundance taxa can result in better corre-lations between community composition and environmental pa-rameters (64) Although this may suggest that these taxa are notactively interacting with their environment it could also suggestthat abundant organisms act as ldquoecosystem engineersrdquo directlyaltering the environment while the activity of low-abundancetaxa has much less impact Recent work also suggests that low-abundance organisms may be important for the response to dis-turbances in terrestrial (65) and aquatic (66) environments Arethese low-abundance taxa nothing more than seed banks thatbloom when conditions are right There is some evidence for thishowever other studies suggest that low-abundance organismsmay be disproportionately active in comparison to more abun-dant taxa (21)

Taxon Turnover

The species-area relationship reflects the fact that larger areas tendto harbor greater species richness This pattern has been widelyestablished for macrobes and has also been demonstrated for free-living fungi bacteria and archaea (67ndash70) As described abovethe Z value measures the rate of species addition per unit area TheZ values for microbes range and appear to be affected by spatialscale (71) sampling intensity (50) and species definitions (72)Related to the species-area relationship is the distance-decay rela-tionship (Fig 2) in which turnover in microbial community com-position is observed over space and communities become less andless similar in terms of community composition the further apartthey are geographically (68 73)

Similar relationships in terms of species richness and taxonturnover have also been observed over time Preston (74) origi-nally hypothesized a positive relationship between the duration ofobservation and the number of taxa a pattern referred to as thespecies-time relationship This pattern has received much less at-tention than the species-area relationship in the macrobial litera-ture perhaps because of difficulties inherent in observing slower-changing and larger ecosystems over time However recentstudies have shown that such a relationship exists for many taxa(75) Likewise this relationship has been shown for microbial

FIG 1 Typical rank-abundance plot where each point represents the abun-dance of one organism within the community (data from reference 120) Acommon feature of many biological communities is that few organisms arepresent in high abundances while the majority of taxa are found in lowabundances

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

communities including bacteria on the surfaces of leaves andcommunities present in activated sludge (76ndash78)

Both the species-area and the distance-decay relationships aswell as their temporal analogues are so much a part of our expe-rience as observers of the natural environment that they seem tobe common sense yet the mechanisms underlying these relation-ships for both macrobes and microbes are still poorly understood(see discussion below) Given the current sampling challenges formicrobial communities it may be premature to compare the turn-over rates of macrobial and microbial communities but somework does suggest that Z values may be somewhat lower for mi-croorganisms (60 70) perhaps because of the high degree of dor-mancy in microbial communities (21)

Phylogenetic Structure

Microbial communities tend to be more phylogenetically clus-tered than expected by chance (68 79) harboring groups ofclosely related taxa that exhibit microscale differences in genomicdiversity However a few communities show the opposite pat-terns in which taxa are less clustered and are less related thanexpected by chance (ie overdispersed) (28 80) Both types ofpatterns have also been observed in macrobial communities (81)

VELLENDrsquoS CONCEPTUAL SYNTHESIS OF COMMUNITYECOLOGY

What processes are driving the biogeographical patterns describedabove The field of community ecology seeks to understand themechanisms of assembly and how they produce patterns in bothspace and time Given the overlap in the fields of biogeography

and community ecology it is unfortunate that inconsistent termi-nology is sometimes used to describe the same patterns and pro-cesses across fields As the training of many microbiologists re-flects a reductionist approach emphasizing the genetics andphysiology of individual taxa rather than their ecology these fieldscan be even more difficult to navigate and unify Here we make aneffort to clarify synonyms that have been used in the literature andpresent a consistent framework with which to discuss patterns andprocesses in community ecology A more complete integration ofmicroorganisms into these fields will allow researchers to testbroader theories on organisms some of which can be easier tomanipulate and most of which are faster to respond than mac-robes

We begin with what has been called one of the only ldquolawsrdquo inecology the species-area relationship and use it to illustrate howthe processes involved in community assembly are actually quitesimple (5 82) We note that many of these examples could alsoapply to the species-time relationship We use Vellendrsquos (5) ap-proach (Table 2) to classify the possible drivers of this relation-ship First larger areas are more likely to encompass greater diver-sities of habitat types allowing for a greater diversity of organismsto coexist through selection defined as ldquodeterministic fitness dif-ferences between individualsrdquo (in other work this has been re-ferred to as niche processes environmental filters and determin-istic processes) Larger areas may also provide a larger ldquotargetrdquo forthe dispersal of organisms from outside the ecosystem Thus thespecies-area relationship could reflect greater dispersal or theldquomovement of organisms across spacerdquo Likewise as larger areasallow for larger population sizes the role of extinction throughdrift or the ldquorandom changes in organism abundancesrdquo will beless likely Finally larger areas may provide more chances for di-versification through both larger population sizes and more di-verse niches This represents a slight modification of the Vellendmodel which identified ldquospeciationrdquo as the process of interestHowever evolutionary change can alter community dynamicseven if new species are not created (83) Likewise as the Vellendmodel operates at the individual level it seems more appropriateto consider diversification rather than speciation

All of these processes operate in combination but some pro-cesses may vary in relative importance across different groups oforganisms and for different systems Like Roughgarden (12) Vel-lend (5) acknowledges the two forces that act to bring new organ-isms into communities (speciation and dispersal) and the pro-cesses that affect changes in the presence and absence as well as therelative abundance of organisms over time (drift and selection)This framework is analogous to the theory of population geneticsin which allele frequencies are explained through a combinationof mutation gene flow genetic drift and natural selection

FIG 2 Variogram showing how phylogenetic distance between soil rotifercommunities (community dissimilarity) varies with the log of geographic dis-tance between communities Weighted UniFrac values (a measure of phyloge-netic distance between communities [51 52]) close to 1 indicate very differentcommunities and values close to 0 indicate almost identical communities Thered vertical line is an estimate of the autocorrelation range (60 m) beyondwhich communities show very little autocorrelation Replotted from data re-ported by Robeson et al (143)

TABLE 2 Vellendrsquos four processes for community assembly

Process Description

Diversification Generation of new genetic variationDispersal Movement of organisms across spaceSelection Changes in community structure caused

by deterministic fitness differencesbetween taxa

Drift Stochastic changes in the relativeabundances of different taxa within acommunity through time

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Although this conceptual model was not borne out of a specificconsideration of microbial communities the same basic forcesshould also guide their assembly (84) We find this framework tohave a number of strengths particularly with regard to microor-ganisms Importantly rather than including evolutionary forcesas an afterthought this conceptual model recognizes the centralrole that diversification plays in driving community ecology (83)Vellendrsquos approach also avoids unnecessary and polarizing de-bates while still allowing for the possibility if not the probabilitythat unique combinations of these processes will likely drive as-sembly in different systems For example this model encompassesboth niche-based processes (driven by fitness differences betweenorganisms) and neutral processes (driven by stochastic processes)in shaping communities This is important because these debatescan often be misleading when not presented within a simplifiedframework Consider for example Baas-Beckingrsquos ldquoeverything iseverywhererdquo hypothesis which emphasizes selection as a drivingforce in microbial community assembly (85) While this is tradi-tionally thought of as a niche-based hypothesis neutral processesare also central to this model as dispersal is considered to be aconstant across all organisms in space in time Likewise the neu-tral theory of biodiversity posits that organisms at the sametrophic level are equivalent with respect to fitness within a specificenvironment (86 87) This model explicitly includes three of Vel-lendrsquos processes dispersal drift and evolutionary diversificationIn the sections that follow we highlight how the life history traitsof microbes may affect the relative importance of the four pro-cesses for microbial community assembly

SELECTION

Selection is a large force shaping microbial community assembly(20 59 88) Different habitat types (eg seawater versus soils)harbor different suites of microorganisms (89) and a copiousamount of data supports the role of a variety of environmentalfactors in determining bacterial assemblage structure and diver-sity including pH salinity and the abundance and quality of car-bon (59 88 90 91) Together these relationships support theimportance of selection via abiotic factors in determining micro-bial community structure

However we know much less about how biotic interactions(eg commensalism mutualism and parasitism) shape microbialcommunities While there is a rich history of studying these pro-cesses in macrobial communities such interactions are muchmore difficult to observe and document in microbial communi-ties Many examples of microbial interactions have been describedfrom the organismal perspective (eg H2 syntrophy and endo-symbiosis of plastids) yet we know little about how these scale toshape entire communities Weiher and Keddy (11) have proposeda continuum of how abiotic and biotic factors may drive commu-nity composition and how this may relate to phylogenetic struc-ture They hypothesized that phylogenetically overdispersed com-munities are characterized by strong species-species interactionsand that competition for similar resources or facilitation may re-sult in these patterns (92) excluding more similar taxa that aremore likely to feature niche overlap By contrast it has been hy-pothesized that communities that are phylogenetically clusteredlike many microbial communities (79) are driven by strong selec-tion acting over broad phylogenetic scales Indeed Philippot andcoworkers (93) have shown the deep phylogenetic coherence ofecological traits suggesting that lineages of microbes may display

ecological similarities over large phylogenetic distances poten-tially accounting for the clustering of such communities

By contrast others have hypothesized that such clusters maydevelop over time through evolutionary processes and may actu-ally be a reflection of weak selection (27) Another factor that mayweaken the effects of selection in microbial communities is dor-mancy because dormant cells are essentially invisible to selectionprocesses Consider persister cells which are bacteria that are ge-netically sensitive to antibiotics but because they are in an inactivestate can persist in populations following exposure (94) Al-though there is a physiological cost to dormancy mechanisms(22) given the large selective advantage of being able to persistunder harsh conditions dormancy may be a common phenotypein microorganisms

Although selection should be similar in macrobial and micro-bial systems the specifics of these processes will no doubt be quitedifferent due to the vast metabolic diversity harbored within mi-crobial communities and even within individual organisms Thusthe complexities of the potential environmental and biologicaldrivers of fitness are greatly magnified for these communitiesIndeed the metabolic breadth of microorganisms has been hy-pothesized to be a key factor in the generation and maintenance ofmicrobial diversity Support for this hypothesis has emerged fromthe discovery that sediment environments which feature strongspatial gradients in electron donors and acceptors harbor themost diverse of microbial communities (88)

Finally the prominent role of horizontal gene transfer (HGT)(and recombination in general) in microbial diversification mayaffect microbial community assembly through selection For ex-ample recent work demonstrates that selection acts on traits thatare subject to horizontal gene transfer (27 95) Burke and cowork-ers (95) found that patterns in microbial assembly processes wererelated to functional genes (ie traits) rather than taxonomy Thisfinding highlights the need for trait-based approaches to under-stand community assembly processes as recombination canscramble the relationship between phylogenetics and function(96)

DISPERSAL

Because of the small size high abundance and short generationtime of microorganisms dispersal processes have not been rigor-ously studied much less quantified Thus the distributions ofmicrobes are often used as proxies for dispersal Given that it isdifficult to conclude that an organism is absent from a specificenvironment and that the current distribution of organismscould also reflect selection processes that have excluded less ldquofitrdquoorganisms andor speciation there are severe limitations to ourunderstanding of the role of dispersal processes in communityassembly Thus here we emphasize that ldquodispersalrdquo is differentfrom migration in which a new organism is incorporated into acommunity from outside Migration events are the result of dis-persal as well as selection and possibly drift We also highlight thatit is often the case that dispersal is discussed only in terms of howlimited it is (ldquodispersal limitationrdquo) However dispersal can haveother dimensions with consequences for community assemblyincluding rates and the order in which taxa are added to commu-nities

Microbial dispersal is typically a passive process While somemicrobes can propel themselves to some degree these processesare unlikely to result in long-distance dispersal events (20) Trans-

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

port via wind water and hitchhiking onto mobile macrobes areall common mechanisms for microbial dispersal As noted abovemany have asserted that free-living microorganisms do not expe-rience passive dispersal limitations (17 58) This is supported tosome degree by the distribution of ldquounlikely inhabitantsrdquo includ-ing the presence of spore-forming hyperthermophiles in inhospi-table locations such as arctic fjord sediments (63) or temperatesoils (97) However microbial communities in air and water showdifferences in both space and time (98 99) thereby affecting dis-persal patterns Curtis and coworkers (61) estimate that the low-est-abundance soil organisms may be present at densities ap-proaching 1 cell in every 27 km2 Given the estimates put forth byPapke and Ward (100) it would take 2 to 220 times the age of theearth for all microbes to disperse through the atmosphere

Likewise while passive dispersal is often considered stochasticit is not entirely so taxa vary in dispersal ability making dispersalprobability not entirely random among species For example dor-mancy mechanisms may make organisms more resistant to theenvironmental stressors (eg extremes of temperature wateravailability and UV exposure) encountered during dispersal (2223) Indeed examinations of microbial biogeographical patternsprovide support for the potential role of microbial life historytraits in affecting dispersal Darcy and coworkers (16) found Be-taproteobacteria from the genus Polaromonas to be globally dis-tributed among high-altitude alpine environments and identifieda potential dormancy mechanism that could permit these bacteriato travel through the upper atmosphere Likewise Bissett and col-leagues (71) found that the distribution of spore-forming organ-isms was decoupled from environmental parameters By contrastthe relative abundances of organisms within the RhizobiaceaeBradyrhizobiaceae and Xanthomonadaceae which they classifiedas poor dispersers because of their tendency to form associationswith macroorganisms were correlated with edaphic factorsWhile there are caveats to interpretations of these data includingthe fact that abundant organisms are easier to detect they presentan interesting case for the differences in dispersal potential be-tween organisms

DIVERSIFICATION

The process of evolutionary diversification is fundamental to bio-geography and community ecology but historically evolution hasbeen given only lip service as a potential factor in ecological pro-cesses However scientists are increasingly recognizing that bothprocesses can act on the same spatial and temporal scales (101)Unfortunately as is the case with dispersal we understand littleabout the spatial and temporal dynamics of how microbes evolveas these processes are difficult to study empirically Again we typ-ically make inferences regarding evolution based on the distribu-tion of microbial genetic diversity in space and time While theinfluence of evolutionary history in explaining community struc-ture over long temporal scales is well known recent work demon-strates that active processes of diversification can play a more im-mediate role in microbial community assembly (27)

The process of dormancy is likely to affect microbial evolution-ary processes with potential implications for community assem-bly As mentioned above dormancy can protect cells from thepossibility of death at least temporarily Dormant cells can sitfrozen in time until favorable conditions or stochastic factors(102) lead to their growth This could result in dramatic variationin evolutionary rates over time More importantly it could also

result in raw material for evolutionary processes that are some-what decoupled from the recent history of the environmentwhich may produce more rapid and dramatic differences in thephylogenetic and functional diversity of community members

The process of horizontal gene transfer (HGT) may alsochange community assembly dynamics for microorganisms com-pared to macrobes In addition to changing the rate and tempo ofevolution as HGT can act as both a diversifying and a homoge-nizing force (103) this mechanism may alter the role of historicalprocesses in community assembly For example consider one mi-crobial species that has evolved to cope with an environmentaltoxin through selection for a particular gene sequence providingit with a competitive advantage over other organisms If anotherorganism acquires this resistance determinant through horizontalgene transfer it is no longer subject to the same genomic environ-ment in which the allele evolved potentially allowing this organ-ism to explore new fitness landscapes

Also microorganisms can evolve through mutation rapidlywhich may have implications for community assembly For exam-ple single-species biofilms can quickly generate diversity that pro-motes ecological stability (104 105) Some bacteria can also initi-ate increased rates of mutation and horizontal gene transferparticularly as a strategy in facing inhospitable environmentsagain increasing diversity at variable evolutionary rates over timewith implications for adaptation in such communities (106) Al-though it is important to emphasize that to some degree rapidevolution may be an artifact of nutrient-rich laboratory condi-tions (107) the evolution of microorganisms in response to theintroduction and proliferation of antibiotic use demonstrates thatthese processes can happen rapidly under more ldquoreal-worldrdquo con-ditions

Ecological drift or stochastic changes in the relative abundance oforganisms may play an important role in microbial communityassembly Empirical and theoretical studies of macrobial systemshave demonstrated that drift is most important when selection isweak alpha diversity is low and the total number of communitymembers is small (reviewed in reference 54) These conditions canbe met in certain types of microbial communities including nu-trient-rich systems such as wastewater treatment facilities (108109) as well as host-associated environments (110) The vast ma-jority of taxa in microbial communities are found in low relativeabundances Low-abundance microorganisms are more vulnera-ble to the effects of drift since slight negative changes in theirabundance could result in their extinction on a local scale (111)However low-abundance individuals may exist in dormant statesprotecting them from extinction A better understanding of thedynamics between dormancy and local extinctions is vital to anappreciation of the role of drift in community assembly particu-larly for microbial assemblages

COMBINING FORCES COMMUNITY ASSEMBLY

How do combinations of these processes (Table 2) influence mi-crobial community assembly Specifically which features of acommunity may make certain processes more or less importantWhat aspects of these processes need to be quantified and inte-grated with existing models to better understand microbial com-munity assembly In the following sections we attempt to addresssome of these questions

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

First several factors should affect the relative importance ofselection in microbial community assembly Recent reviews havehighlighted the fact that many microbial communities are sensi-tive to disturbances to abiotic and biotic features of their environ-ment and that community resilience is related to the severity andduration of the disturbance community diversity disturbancehistory and abiotic factors (112ndash114) Drift should be most im-portant in assembly when the degree of ecological equivalencewithin community members is high (115) As many microbialcommunities are phylogenetically underdispersed (68 79 116) itis possible that many ecologically equivalent organisms coexist inmicrobial communities Importantly even if the genetic potentialfor functional differences between these taxa exists if the ecosys-tem does not feature variation to make these differences relevantfor fitness the taxa are effectively ecologically equivalent The po-tential for the community to undergo significant turnoverthrough either drift or selection is related to growth rates and willthus be sensitive to the nutrient status of the environment Bycontrast the effects of evolutionary diversification and dispersalon assembly processes may be most important in new or changingenvironments when newly arriving organisms face less competi-tion from resident organisms (115)

As noted above however all of these processes operate in com-bination and changes in the interactions of dispersal drift selec-tion and diversification will have large impacts of communityassembly While little is known about the dynamics of microbialinvasions some work suggests that communities may experiencemore dramatic shifts in response to the introduction of new indi-viduals (via diversification and dispersal) following a disturbanceevent likely due to changes in selection (114 117) Indeed distur-bance has also been shown to increase the relative importance ofselection in structuring communities (118 119) However otherstudies suggest that neutral processes increase following distur-bance events We recently examined the response of soil microbial

communities to a wildfire and found evidence for an increase inthe effects of neutral processes 1 month following the disturbancebut an increase in the effects of selection after 4 months (120)Such transitions over very short time scales may reconcile the factthat different studies have yielded disparate patterns in commu-nity assembly in response to disturbance

TEMPORAL AND SPATIAL SCALES

Two aspects of assembly that have not been explored extensivelyfor microbial communities are the role of temporal and spatialscales For example temporal scales can be important for under-standing the roles of ldquopriority effectsrdquo (121ndash124) Here selectioncauses changes in community structure when dispersal driftandor diversification introduces variation in the initial relativeabundance (ie frequencies) of species (Fig 3) Gleason (125) wasthe first to imply that the order of species colonization can result indivergence between communities even when environmental con-ditions and regional species pools are identical In these commu-nities early-colonizing organisms can have inhibitory or facilita-tive effects on late-arriving organisms (123) either through directinteractions (eg competition and symbiosis) or through envi-ronmental modification (126)

The importance of assembly history for community ecologyhas been likened to that of evolutionary processes ldquoWhile notdenying the importance of current adaptations we cannot ignorethe long and apparently capricious pathways taken by evolution-ary lineagesrdquo (127) Some work has even demonstrated the exis-tence of ldquoHumpty-Dumptyrdquo communities (128) or communitiesthat cannot be put back together with only the species that theycontain (129) further supporting the role of assembly history incommunity structure The role of priority effects on model micro-bial communities has been studied extensively (129ndash134) Micro-bial systems have played a central role as a model system to gen-erate test and refine general hypotheses on community assembly

FIG 3 Contrasting hypotheses of community assembly Numbers represent hypothetical species arrows represent species immigration letters representdifferent immigration histories and roman numerals represent variations in habitat conditions (Top) Local communities converge in species compositionunder the same environmental conditions regardless of immigration history (Bottom) Local communities diverge in species composition when immigrationhistory is variable even under the same environmental conditions (ie priority effects) (Adapted from references 123 [p 45] and 158 with permission)

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

because the short generation time and other logistical ease associ-ated with microbes make them convenient experimental tools(135ndash137) Some studies demonstrate that assembly order canaffect not only community structure but also ecosystem-level pro-cesses such as decomposition and carbon flux (138) Howevermuch of this research has been conducted in the laboratory andexperiments are needed that test hypotheses under more realisticconditions

Also spatial scales are important in community ecology (123)and metacommunity theory is the study of spatially distinct com-munities that are linked through dispersal (36) Given the highdispersal potential of many microorganisms these linkages arelikely to happen over multiple spatial scales The same basic pro-cesses are influencing community assembly but selection driftand evolution can be decoupled spatially but linked through dis-persal For example extinction events in one community can beldquorescuedrdquo through dispersal from a nearby patch Thus tradeoffscan exist in which less-fit organisms persist within metacommu-nities because of high dispersal rates

These types of spatially explicit differences in patterns and pro-cesses likely exist for microbial communities as well and may ex-plain some of the variation in microbial community compositionRecently Martiny and coworkers (139) suggested that geographicdistance contributed to community differences within saltmarshes (local scale) but not between marshes (regional scale)Various environmental factors however contributed to commu-nity difference at all scales As has been determined for some mac-robial communities (eg see reference 55) a hierarchical under-standing of the links between microbial populations andcommunities should be developed focusing on rates of dispersaland genetic change over space and time Spatial factors such as thesize of local patches (140ndash142) the distance or isolation betweencommunities (143ndash145) and how continuous habitats are acrossthe landscape (146) are all features that will affect the metacom-munity dynamics of microorganisms

IMPLICATIONS FOR FUNCTION

Microbes regulate all major biogeochemical cycles and directlyinfluence plant animal and human welfare However our under-standing of how assembly processes might ultimately influenceecosystem function remains limited Since microbial communitystructure and function are inextricably linked some argue thatthere is intrinsic value in knowing ldquowho does whatrdquo to understandbroader controls over ecosystem processes (147) Indeed the be-lief that community composition determines or at least influ-ences ecosystem function is widely held in community ecologyand is supported by recent work with microbial systems (138148) However many studies have documented correlations be-tween microbial community structure and an array of environ-mental factors (59 88 90) and a large body of literature demon-strates that many of these same parameters are important incontrolling rates of ecosystem processes (149) On the other handBurke and colleagues (95) recently showed that microbial com-munity succession on algae in the ocean showed functional con-vergence but lacked taxonomic coherence This suggests a highdegree of functional redundancy in microbial communitieswhich may decouple structure and function Thus the questionremains does information on microbial community structureprovide added value beyond data on key chemical and physicalfactors in an environment Put another way if the environment

dictates microbial community structure do we need to knowldquowho is thererdquo to predict ldquowhat they will dordquo

Answering this question will require a more complete under-standing of the links between structure and function which maybe affected by the degree of functional redundancy (150) as well ashorizontal gene transfer within communities Moreover we arguethat this will require disentangling the roles of the four processesin community assembly (Table 2) and developing a better under-standing of where and when these matter Some of the argumentsagainst the added value of community composition data are pred-icated on the assumption that microbial community assembly isentirely selection based Communities that are assembled primar-ily through neutral processes should be less affected by differencesin environmental parameters Thus while environmental factorswill still affect processes in these communities to some degree byregulating the physiologies of individual organisms microbialcommunities assembled via stochastic processes may exhibit lessof a direct link between the environment and processes (Fig 4)Describing communities based on functional traits in addition tospecies richness patterns could be a particularly useful addition tocommunity ecology as a whole Microbial species traits may yieldnew insights into various measures of ecosystem function (151ndash153) and remedy the shortcomings of the use of species richnessfor evaluating ecosystem function and microbial ecology in gen-eral (96) However selecting and measuring microbial traits thatare relevant to community assembly and ecosystem function areformidable tasks

IMPLICATIONS FOR BIODIVERSITY

We conclude by returning to the age old question of ldquoWhy arethere so many species of microorganismsrdquo also called the ldquopara-dox of the planktonrdquo (154) Ecological community theory suggeststhat a high level of biodiversity within the same trophic level isunlikely due to competition and drift Coexistence should requirethe partitioning of resources in time and space or tradeoffs be-tween species-species interactions and dispersal (155 156) Sowhy are microbial communities so diverse

As mentioned above the microbial communities with thehighest alpha diversity occur in sediments (88) Our analyses alsosuggest that sediments feature more beta diversity than other mi-crobial communities (T M Legg and D R Nemergut unpub-lished data) Although the source of this beta diversity is un-known one hypothesis is that local- and regional-scaleheterogeneity in terms of redox gradients and nutrient availabilitycould be an important driver of diversity (88) So-called ldquostorageeffectsrdquo through dormancy mechanisms may be strong for micro-organisms given that a changing environment permits coexis-tence as no single species can be competitively superior under allconditions (101 156)

However the spatial scale under which we sample microbialcommunities may confound comparisons of patterns in diversitybetween macrobes and microbes Sampling over an entire gram ofsoil for example will homogenize across many communitiesthereby reducing beta diversity estimates (89) Thus the high al-pha diversity observed for environmental microbial analyses mayrepresent high gamma diversity driven by high beta diversity As-pects of our approach to examining microbial communities mayalso influence observed patterns of diversity For example 16SrRNA gene phylogenetic analyses and our working definitions formicrobial species do not capture ecological differentiation among

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

closely related taxa (79 157) In total an understanding of therelationship between patterns of microbial diversity and commu-nity assembly demands the implementation of a framework thataccounts for the unique biology and scales of microbial ecology toproperly discern meaningful patterns in microbial communitiesas opposed to patterns that are artifacts of sampling methods

While high-resolution data on microbial communities are in-creasingly revealing a multitude of patterns which suggest thatvarious mechanisms in community assembly are at work acrossdifferent environments and scales Vellendrsquos framework providesa foundation to unify wide-ranging observations in the context offundamental testable ecological processes The framework de-scribed here will help microbial ecology to move from a largelyobservational and correlative field to one with more mechanisticinsights If executed with the proper experimental approachesthis effort will not only advance ecological theory but also take amajor step forward in demystifying microbial communities andtherefore the building blocks of ecosystems our environmentsand even ourselves

ACKNOWLEDGMENTS

We acknowledge funding from the National Science Foundation (grantDEB1545913 to DRN and SKS)

We thank Elizabeth Costello for helpful discussions The comments ofthree anonymous reviewers were instrumental in improving the quality ofthe manuscript

REFERENCES1 Brown MV Philip GK Bunge JA Smith MC Bissett A Lauro FM

Fuhrman JA Donachie SP 2009 Microbial community structure in theNorth Pacific Ocean ISME J 31374 ndash1386

2 Gilbert JA Field D Swift P Thomas S Cummings D Temperton BWeynberg K Huse S Hughes M Joint I Somerfield PJ Muhling M2010 The taxonomic and functional diversity of microbes at a temperatecoastal site a lsquomulti-omicrsquo study of seasonal and diel temporal variationPLoS One 5e15545 doi101371journalpone0015545

3 Lee CK Barbier BA Bottos EM McDonald IR Cary SC 2012 The

Inter-Valley Soil Comparative Survey the ecology of Dry Valley edaphicmicrobial communities ISME J 61046 ndash1057

4 Pommier T Neal PR Gasol JM Coll M Acinas SG Pedros-Alio C2010 Spatial patterns of bacterial richness and evenness in the NW Med-iterranean Sea explored by pyrosequencing of the 16S rRNA AquatMicrob Ecol 61212ndash224

5 Vellend M 2010 Conceptual synthesis in community ecology Q RevBiol 85183ndash206

6 Caporaso JG Kuczynski J Stombaugh J Bittinger K Bushman FDCostello EK Fierer N Pena AG Goodrich JK Gordon JI Huttley GAKelley ST Knights D Koenig JE Ley RE Lozupone CA McDonald DMuegge BD Pirrung M Reeder J Sevinsky JR Tumbaugh PJ WaltersWA Widmann J Yatsunenko T Zaneveld J Knight R 2010 QIIMEallows analysis of high-throughput community sequencing data NatMethods 7335ndash336

7 Giongo A Crabb DB Davis-Richardson AG Chauliac D MobberleyJM Gano KA Mukherjee N Casella G Roesch LFW Walts B Riva AKing G Triplett EW 2010 PANGEA pipeline for analysis of nextgeneration amplicons ISME J 4852ndash 861

8 Yilmaz P Kottmann R Field D Knight R Cole JR Amaral-Zettler LGilbert JA Karsch-Mizrachi I Johnston A Cochrane G Vaughan RHunter C Park J Morrison N Rocca-Serra P Sterk P Arumugam MBailey M Baumgartner L Birren BW Blaser MJ Bonazzi V Booth TBork P Bushman FD Buttigieg PL Chain PSG Charlson E CostelloEK Huot-Creasy H Dawyndt P DeSantis T Fierer N Fuhrman JAGallery RE Gevers D Gibbs RA Gil IS Gonzalez A Gordon JIGuralnick R Hankeln W Highlander S Hugenholtz P Jansson J KauAL Kelley ST Kennedy J Knights D Koren O Kuczynski J et al2011 Minimum information about a marker gene sequence (MIMARKS) andminimum information about any (x) sequence (MIxS) specifications Nat Bio-technol 29415ndash420

9 Jansson JK Prosser JI 2013 Microbiology the life beneath our feetNature 49440 ndash 41

10 Wennekes PL Rosindell J Etienne RS 2012 The neutral-niche debatea philosophical perspective Acta Biotheor 60257ndash271

11 Weiher E Keddy PA 1995 Assembly rules null models and trait dis-persionmdashnew questions from old patterns Oikos 74159 ndash164

12 Roughgarden J 2009 Is there a general theory of community ecologyBiol Philos 24521ndash529

13 Hugenholtz P Goebel BM Pace NR 1998 Impact of culture-independent studies on the emerging phylogenetic view of bacterial di-versity J Bacteriol 1804765ndash 4774

FIG 4 Conceptual model outlining the relationship between edaphic factors soil microbial community composition and ecosystem processes for deterministicand stochastic assembly mechanisms

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

14 Kirk JL Beaudette LA Hart M Moutoglis P Khironomos JN Lee HTrevors JT 2004 Methods of studying soil microbial diversity J Micro-biol Methods 58169 ndash188

15 Schloss PD Handelsman J 2004 Status of the microbial census Mi-crobiol Mol Biol Rev 68686 ndash 691

16 Darcy JL Lynch RC King AJ Robeson MS Schmidt SK 2011 Globaldistribution of Polaromonas phylotypesmdash evidence for a highly success-ful dispersal capacity PLoS One 6e23742 doi101371journalpone0023742

17 Finlay BJ 2002 Global dispersal of free-living microbial eukaryote spe-cies Science 2961061ndash1063

18 Finlay BJ Clarke KJ 1999 Ubiquitous dispersal of microbial speciesNature 400828

19 Wilkinson DM Koumoutsaris S Mitchell EAD Bey I 2012 Modellingthe effect of size on the aerial dispersal of microorganisms J Biogeogr3989 ndash97

20 Martiny JBH Bohannan BJM Brown JH Colwell RK Fuhrman JAGreen JL Horner-Devine MC Kane M Krumins JA Kuske CRMorin PJ Naeem S Ovreas L Reysenbach AL Smith VH Staley JT2006 Microbial biogeography putting microorganisms on the map NatRev Microbiol 4102ndash112

21 Jones SE Lennon JT 2010 Dormancy contributes to the maintenanceof microbial diversity Proc Natl Acad Sci U S A 1075881ndash5886

22 Lennon JT Jones SE 2011 Microbial seed banks the ecological andevolutionary implications of dormancy Nat Rev Microbiol 9119 ndash130

23 Locey KJ 2010 Synthesizing traditional biogeography with microbialecology the importance of dormancy J Biogeogr 371835ndash1841

24 Fredrickson JK Romine MF Beliaev AS Auchtung JM Driscoll MEGardner TS Nealson KH Osterman AL Pinchuk G Reed JL Rodi-onov DA Rodrigues JLM Saffarini DA Serres MH Spormann AMZhulin IB Tiedje JM 2008 Towards environmental systems biology ofShewanella Nat Rev Microbiol 6592ndash 603

25 Heidelberg JF Paulsen IT Nelson KE Gaidos EJ Nelson WC ReadTD Eisen JA Seshadri R Ward N Methe B Clayton RA Meyer TTsapin A Scott J Beanan M Brinkac L Daugherty S DeBoy RTDodson RJ Durkin AS Haft DH Kolonay JF Madupu R PetersonJD Umayam LA White O Wolf AM Vamathevan J Weidman JImpraim M Lee K Berry K Lee C Mueller J Khouri H Gill JUtterback TR McDonald LA Feldblyum TV Smith HO Venter JCNealson KH Fraser CM 2002 Genome sequence of the dissimilatorymetal ion-reducing bacterium Shewanella oneidensis Nat Biotechnol201118 ndash1123

26 Polz MF Hunt DE Preheim SP Weinreich DM 2006 Patterns andmechanisms of genetic and phenotypic differentiation in marine mi-crobes Philos Trans R Soc Lond B Biol Sci 3612009 ndash2021

27 Shapiro BJ Friedman J Cordero OX Preheim SP Timberlake SCSzabo G Polz MF Alm EJ 2012 Population genomics of early events inthe ecological differentiation of bacteria Science 33648 ndash51

28 Thompson JR Pacocha S Pharino C Klepac-Ceraj V Hunt DEBenoit J Sarma-Rupavtarm R Distel DL Polz MF 2005 Genotypicdiversity within a natural coastal bacterioplankton population Science3071311ndash1313

29 Schmidt SK Costello EK Nemergut DR Cleveland CC Reed SCWeintraub MN Meyer AF Martin AM 2007 Biogeochemical conse-quences of rapid microbial turnover and seasonal succession in soilEcology 881379 ndash1385

30 Yoshida T Ellner SP Jones LE Bohannan BJM Lenski RE HairstonNG 2007 Cryptic population dynamics rapid evolution masks trophicinteractions PLoS Biol 5e235 doi101371journalpbio0050235

31 Ochman H Lawrence JG Groisman EA 2000 Lateral gene transfer andthe nature of bacterial innovation Nature 405299 ndash304

32 Sexstone AJ Revsbech NP Parkin TB Tiedje JM 1985 Direct mea-surement of oxygen profiles and denitrification rates in soil aggregatesSoil Sci Soc Am J 49645ndash 651

33 Pernthaler A Pernthaler J Amann R 2002 Fluorescence in situ hy-bridization and catalyzed reporter deposition for the identification ofmarine bacteria Appl Environ Microbiol 683094 ndash3101

34 Behrens S Losekann T Pett-Ridge J Weber PK Ng WO StevensonBS Hutcheon ID Relman DA Spormann AM 2008 Linking micro-bial phylogeny to metabolic activity at the single-cell level by using en-hanced element labeling-catalyzed reporter deposition fluorescence insitu hybridization (EL-FISH) and NanoSIMS Appl Environ Microbiol743143ndash3150

35 Magurran AE 2003 Measuring biological diversity Blackwell Publish-ing Victoria Australia

36 Leibold MA Holyoak M Mouquet N Amarasekare P Chase JMHoopes MF Holt RD Shurin JB Law R Tilman D Loreau MGonzalez A 2004 The metacommunity concept a framework for multi-scale community ecology Ecol Lett 7601ndash 613

37 Lawton JH Bignell DE Bolton B Bloemers GF Eggleton P Ham-mond PM Hodda M Holt RD Larsen TB Mawdsley NA Stork NESrivastava DS Watt AD 1998 Biodiversity inventories indicator taxaand effects of habitat modification in tropical forest Nature 39172ndash76

38 Odum EP 1968 Energy flow in ecosystemsmdasha historical review AmZool 811ndash18

39 Fauth JE Bernardo J Camara M Resetarits WJ VanBuskirk J Mc-Collum SA 1996 Simplifying the jargon of community ecology a con-ceptual approach Am Nat 147282ndash286

40 Martin-Laurent F Philippot L Hallet S Chaussod R Germon JCSoulas G Catroux G 2001 DNA extraction from soils old bias for newmicrobial diversity analysis methods Appl Environ Microbiol 672354 ndash2359

41 Gotelli NJ Graves GR 1996 Null models in ecology SmithsonianInstitution Press Washington DC

42 Klappenbach JA Saxman PR Cole JR Schmidt TM 2001 rrndb theRibosomal RNA Operon Copy Number Database Nucleic Acids Res29181ndash184

43 Whittaker RH 1956 Vegetation of the great smoky mountains EcolMonogr 261ndash 69

44 Anderson MJ Crist TO Chase JM Vellend M Inouye BD FreestoneAL Sanders NJ Cornell HV Comita LS Davies KF Harrison SPKraft NJB Stegen JC Swenson NG 2011 Navigating the multiplemeanings of beta diversity a roadmap for the practicing ecologist EcolLett 1419 ndash28

45 Faith DP 1992 Conservation evaluation and phylogenetic diversityBiol Conserv 611ndash10

46 Kembel SW Eisen JA Pollard KS Green JL 2011 The phylogeneticdiversity of metagenomes PLoS One 6e23214 doi101371journalpone0023214

47 Bent SJ Forney LJ 2008 The tragedy of the uncommon understandinglimitations in the analysis of microbial diversity ISME J 2689 ndash 695

48 Lozupone CA Hamady M Kelley ST Knight R 2007 Quantitativeand qualitative beta diversity measures lead to different insights intofactors that structure microbial communities Appl Environ Microbiol731576 ndash1585

49 Lozupone CA Knight R 2008 Species divergence and the measurementof microbial diversity FEMS Microbiol Rev 32557ndash578

50 Sloan WT Lunn M Woodcock S Head IM Nee S Curtis TP 2006Quantifying the roles of immigration and chance in shaping prokaryotecommunity structure Environ Microbiol 8732ndash740

51 Lozupone C Hamady M Knight R 2006 UniFracmdashan online tool forcomparing microbial community diversity in a phylogenetic contextBMC Bioinformatics 7371 doi1011861471-2105-7-371

52 Lozupone C Knight R 2005 UniFrac a new phylogenetic method forcomparing microbial communities Appl Environ Microbiol 718228 ndash8235

53 Tuomisto H 2010 A diversity of beta diversities straightening up aconcept gone awry Part 1 Defining beta diversity as a function of alphaand gamma diversity Ecography 332ndash22

54 Chase JM Myers JA 2011 Disentangling the importance of ecologicalniches from stochastic processes across scales Philos Trans R SocLond B Biol Sci 3662351ndash2363

55 Crist TO Veech JA Gering JC Summerville KS 2003 Partitioningspecies diversity across landscapes and regions a hierarchical analysis ofalpha beta and gamma diversity Am Nat 162734 ndash743

56 Kraft NJB Comita LS Chase JM Sanders NJ Swenson NG Crist TOStegen JC Vellend M Boyle B Anderson MJ Cornell HV Davies KFFreestone AL Inouye BD Harrison SP Myers JA 2011 Disentanglingthe drivers of beta diversity along latitudinal and elevational gradientsScience 3331755ndash1758

57 Astorga A Oksanen J Luoto M Soininen J Virtanen R Muotka T2012 Distance decay of similarity in freshwater communities do macro-and microorganisms follow the same rules Glob Ecol Biogeogr 21365ndash375

58 Fenchel T Finlay BJ 2004 The ubiquity of small species patterns oflocal and global diversity Bioscience 54777ndash784

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

59 Fierer N Jackson RB 2006 The diversity and biogeography of soilbacterial communities Proc Natl Acad Sci U S A 103626 ndash 631

60 Fuhrman JA 2009 Microbial community structure and its functionalimplications Nature 459193ndash199

61 Curtis TP Sloan WT Scannell JW 2002 Estimating prokaryotic di-versity and its limits Proc Natl Acad Sci U S A 9910494 ndash10499

62 McGill BJ Etienne RS Gray JS Alonso D Anderson MJ Benecha HKDornelas M Enquist BJ Green JL He FL Hurlbert AH Magurran AEMarquet PA Maurer BA Ostling A Soykan CU Ugland KI WhiteEP 2007 Species abundance distributions moving beyond single pre-diction theories to integration within an ecological framework EcolLett 10995ndash1015

63 Hubert C Loy A Nickel M Arnosti C Baranyi C Bruchert VFerdelman T Finster K Christensen FM de Rezende JR VandiekenV Jorgensen BB 2009 A constant flux of diverse thermophilic bacteriainto the cold Arctic seabed Science 3251541ndash1544

64 Gobet A Quince C Ramette A 2010 Multivariate cutoff level analysis(MultiCoLA) of large community data sets Nucleic Acids Res 38e155doi101093nargkq545

65 Cleveland CC Nemergut DR Schmidt SK Townsend AR 2007Increases in soil respiration following labile carbon additions linked torapid shifts in soil microbial community composition Biogeochemistry82229 ndash240

66 Szabo KE Itor POB Bertilsson S Tranvik L Eiler A 2007 Importanceof rare and abundant populations for the structure and functional po-tential of freshwater bacterial communities Aquat Microb Ecol 471ndash10

67 Bell T Newman JA Silverman BW Turner SL Lilley AK 2005 Thecontribution of species richness and composition to bacterial servicesNature 4361157ndash1160

68 Bryant JA Lamanna C Morlon H Kerkhoff AJ Enquist BJ Green JL2008 Microbes on mountainsides contrasting elevational patterns ofbacterial and plant diversity Proc Natl Acad Sci U S A 10511505ndash11511

69 Green JL Holmes AJ Westoby M Oliver I Briscoe D Dangerfield MGillings M Beattie AJ 2004 Spatial scaling of microbial eukaryotediversity Nature 432747ndash750

70 Horner-Devine MC Lage M Hughes JB Bohannan BJM 2004 Ataxa-area relationship for bacteria Nature 432750 ndash753

71 Bissett A Richardson AE Baker G Wakelin S Thrall PH 2010 Lifehistory determines biogeographical patterns of soil bacterial communi-ties over multiple spatial scales Mol Ecol 194315ndash 4327

72 Storch D Sizling AL 2008 The concept of taxon invariance in ecologydo diversity patterns vary with changes in taxonomic resolution FoliaGeobot 43329 ndash344

73 King AJ Freeman KR McCormick KF Lynch RC Lozupone CKnight R Schmidt SK 2010 Biogeography and habitat modelling ofhigh-alpine bacteria Nat Commun 153 doi101038ncomms1055

74 Preston FW 1948 The commonness and rarity of species Ecology29254 ndash283

75 Adler PB Lauenroth WK 2003 The power of time spatiotemporalscaling of species diversity Ecol Lett 6749 ndash756

76 Ayarza JM Erijman L 2011 Balance of neutral and deterministic com-ponents in the dynamics of activated sludge floc assembly Microb Ecol61486 ndash 495

77 Redford AJ Fierer N 2009 Bacterial succession on the leaf surface anovel system for studying successional dynamics Microb Ecol 58189 ndash198

78 van der Gast CJ Ager D Lilley AK 2008 Temporal scaling of bacterialtaxa is influenced by both stochastic and deterministic ecological factorsEnviron Microbiol 101411ndash1418

79 Horner-Devine MC Bohannan BJM 2006 Phylogenetic clustering andoverdispersion in bacterial communities Ecology 87S100 ndashS108 doi1018900012-9658(2006)87[100PCAOIB]20CO2

80 Chaffron S Rehrauer H Pernthaler J von Mering C 2010 A globalnetwork of coexisting microbes from environmental and whole-genomesequence data Genome Res 20947ndash959

81 Cardillo M Gittleman JL Purvis A 2008 Global patterns in the phy-logenetic structure of island mammal assemblages Proc Biol Sci 2751549 ndash1556

82 Acinas SG Sarma-Rupavtarm R Klepac-Ceraj V Polz MF 2005PCR-induced sequence artifacts and bias insights from comparison of

two 16S rRNA clone libraries constructed from the same sample ApplEnviron Microbiol 718966 ndash 8969

83 Rainey PB Travisano M 1998 Adaptive radiation in a heterogeneousenvironment Nature 39469 ndash72

84 Hanson CA Fuhrman JA Horner-Devine MC Martiny JBH 2012Beyond biogeographic patterns processes shaping the microbial land-scape Nat Rev Microbiol 10497ndash506

85 Baas-Becking LGM 1934 Geobiologie of inleiding tot de milieukundeWP Van Stockum amp Zoon The Hague The Netherlands

86 Hubbell SP 2001 The unified neutral theory of biodiversity and bioge-ography Princeton University Press Princeton NJ

87 Rosindell J Hubbell SP Etienne RS 2011 The unified neutral theory ofbiodiversity and biogeography at age ten Trends Ecol Evol 26(7)340 ndash348

88 Lozupone CA Knight R 2007 Global patterns in bacterial diversityProc Natl Acad Sci U S A 10411436 ndash11440

89 Nemergut DR Costello EK Hamady M Lozupone C Jiang LSchmidt SK Fierer N Townsend AR Cleveland CC Stanish L KnightR 2011 Global patterns in the biogeography of bacterial taxa EnvironMicrobiol 13135ndash144

90 Logue JB Lindstrom ES 2010 Species sorting affects bacterioplanktoncommunity composition as determined by 16S rDNA and 16S rRNAfingerprints ISME J 4729 ndash738

91 Nemergut DR Cleveland CC Wieder WR Washenberger CLTownsend AR 2010 Plot-scale manipulations of organic matter inputsto soils correlate with shifts in microbial community composition in alowland tropical rain forest Soil Biol Biochem 422153ndash2160

92 Violle C Nemergut DR Pu ZC Jiang L 2011 Phylogenetic limitingsimilarity and competitive exclusion Ecol Lett 14782ndash787

93 Philippot L Andersson SGE Battin TJ Prosser JI Schimel JP Whit-man WB Hallin S 2010 The ecological coherence of high bacterialtaxonomic ranks Nat Rev Microbiol 8523ndash529

94 Lewis K 2010 Persister cells Annu Rev Microbiol 64357ndash37295 Burke C Steinberg P Rusch D Kjelleberg S Thomas T 2011 Bacterial

community assembly based on functional genes rather than speciesProc Natl Acad Sci U S A 10814288 ndash14293

96 Green JL Bohannan BJM Whitaker RJ 2008 Microbial biogeographyfrom taxonomy to traits Science 3201039 ndash1043

97 Marchant R Banat IM Rahman TJ Berzano M 2002 The frequencyand characteristics of highly thermophilic bacteria in cool soil environ-ments Environ Microbiol 4595ndash 602

98 Bowers RM Sullivan AP Costello EK Collett JL Knight R Fierer N2011 Sources of bacteria in outdoor air across cities in the MidwesternUnited States Appl Environ Microbiol 776350 ndash 6356

99 Lindstrom ES Ostman O 2011 The importance of dispersal for bacte-rial community composition and functioning PLoS One 6e25883 doi101371journalpone0025883

100 Papke RT Ward DM 2004 The importance of physical isolation tomicrobial diversification FEMS Microbiol Ecol 48293ndash303

101 Cavender-Bares J Kozak KH Fine PVA Kembel SW 2009 Themerging of community ecology and phylogenetic biology Ecol Lett12693ndash715

102 Buerger S Spoering A Gavrish E Leslin C Ling L Epstein SS 2012Microbial scout hypothesis and microbial discovery Appl Environ Mi-crobiol 783229 ndash3233

103 Papke RT Gogarten JP 2012 How bacterial lineages emerge Science33645ndash 46

104 Boles BR Thoendel M Singh PK 2004 Self-generated diversity pro-duces ldquoinsurance effectsrdquo in biofilm communities Proc Natl Acad SciU S A 10116630 ndash16635

105 Hansen SK Rainey PB Haagensen JAJ Molin S 2007 Evolution ofspecies interactions in a biofilm community Nature 445533ndash536

106 Rensing C Newby DT Pepper IL 2002 The role of selective pressureand selfish DNA in horizontal gene transfer and soil microbial commu-nity adaptation Soil Biol Biochem 34285ndash296

107 Prosser JI Bohannan BJM Curtis TP Ellis RJ Firestone MK Freck-leton RP Green JL Green LE Killham K Lennon JJ Osborn AMSolan M van der Gast CJ Young JPW 2007 The role of ecologicaltheory in microbial ecology Nat Rev Microbiol 5384 ndash392

108 Ofiteru ID Lunn M Curtis TP Wells GF Criddle CS Francis CASloan WT 2010 Combined niche and neutral effects in a microbialwastewater treatment community Proc Natl Acad Sci U S A 10715345ndash15350

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

109 Yang C Zhang W Liu RH Li Q Li BB Wang SF Song CJ Qiao CLMulchandani A 2011 Phylogenetic diversity and metabolic potential ofactivated sludge microbial communities in full-scale wastewater treat-ment plants Environ Sci Technol 457408 ndash7415

110 Lankau EW Hong PY Mackie RI 2012 Ecological drift and localexposures drive enteric bacterial community differences within species ofGalapagos iguanas Mol Ecol 211779 ndash1788

111 Pedros-Alio C 2006 Marine microbial diversity can it be determinedTrends Microbiol 14257ndash263

112 Allison SD Martiny JBH 2008 Resistance resilience and redundancyin microbial communities Proc Natl Acad Sci U S A 10511512ndash11519

113 Griffiths BS Philippot L 2013 Insights into the resistance and resilienceof the soil microbial community FEMS Microbiol Rev 37112ndash129

114 Lozupone CA Stombaugh JI Gordon JI Jansson JK Knight R 2012Diversity stability and resilience of the human gut microbiota Nature489220 ndash230

115 Leibold MA McPeek MA 2006 Coexistence of the niche and neutralperspectives in community ecology Ecology 871399 ndash1410

116 DeAngelis KM Firestone MK 2012 Phylogenetic clustering of soilmicrobial communities by 16S rRNA but not 16S rRNA genes ApplEnviron Microbiol 782459 ndash2461

117 Litchman E 2010 Invisible invaders non-pathogenic invasive microbesin aquatic and terrestrial ecosystems Ecol Lett 131560 ndash1572

118 Chase JM 2007 Drought mediates the importance of stochastic com-munity assembly Proc Natl Acad Sci U S A 10417430 ndash17434

119 Jiang L Brady L Tan JQ 2011 Species diversity invasion and alter-native community states in sequentially assembled communities AmNat 178411ndash 418

120 Ferrenberg S OrsquoNeill SP Knelman JE Todd B Bradley D RobinsonT Schmidt SK Townsend AR Williams MW Cleveland CC Mel-bourne BA Jiang L Nemergut D 2013 Changes in assembly processesin soil bacterial communities following a wildfire disturbance ISME J71102ndash1111

121 Chase JM 2003 Community assembly when should history matterOecologia 136489 ndash 498

122 Drake JA 1990 Communities as assembled structures do rules governpattern Trends Ecol Evol 5(5)159 ndash164

123 Fukami T 2010 Community assembly dynamics in space p 45ndash53 InVerhoef HA Morin PJ (ed) Community ecology processes models andapplications Oxford University Press Oxford United Kingdom

124 Fukami T Nakajima M 2011 Community assembly alternative stablestates or alternative transient states Ecol Lett 14973ndash984

125 Gleason HA 1927 Further views on the succession concept Ecology8299 ndash326

126 Peterson CH 1984 Does a rigorous criterion for environmental identitypreclude the existence of multiple stable points Am Nat 124127ndash133

127 Luh HK Pimm SL 1993 The assembly of ecological communities aminimalist approach J Anim Ecol 62749 ndash765

128 Pimm SL 1991 The balance of nature Ecological issues in the conser-vation of species and communities The University of Chicago PressChicago IL

129 Warren PH Law R Weatherby AJ 2003 Mapping the assembly ofprotist communities in microcosms Ecology 841001ndash1011

130 Drake JA 1991 Community assembly mechanics and the structure of anexperimental species ensemble Am Nat 1371ndash26

131 Jiang L Patel SN 2008 Community assembly in the presence of distur-bance a microcosm experiment Ecology 891931ndash1940

132 Peay KG Belisle M Fukami T 2012 Phylogenetic relatedness predictspriority effects in nectar yeast communities Proc Biol Sci 279749 ndash758

133 Robinson JV Dickerson JE 1987 Does invasion sequence affect com-munity structure Ecology 68587ndash595

134 Tan JQ Pu ZC Ryberg WA Jiang L 2012 Species phylogeneticrelatedness priority effects and ecosystem functioning Ecology 931164 ndash1172

135 Cadotte MW Drake JA Fukami T 2005 Constructing nature labora-

tory models as necessary tools for investigating complex ecological com-munities Adv Ecol Res 37333ndash353

136 Drake JA Huxel GR Hewitt CL 1996 Microcosms as models forgenerating and testing community theory Ecology 77670 ndash 677

137 Jessup CM Kassen R Forde SE Kerr B Buckling A Rainey PBBohannan BJM 2004 Big questions small worlds microbial modelsystems in ecology Trends Ecol Evol 19(4)189 ndash197

138 Fukami T Dickie IA Wilkie JP Paulus BC Park D Roberts ABuchanan PK Allen RB 2010 Assembly history dictates ecosystemfunctioning evidence from wood decomposer communities Ecol Lett13675ndash 684

139 Martiny JBH Eisen JA Penn K Allison SD Horner-Devine MC 2011Drivers of bacterial beta-diversity depend on spatial scale Proc NatlAcad Sci U S A 1087850 ndash7854

140 Fukami T 2004 Assembly history interacts with ecosystem size to influ-ence species diversity Ecology 853234 ndash3242

141 Lomolino MV 1990 The target area hypothesis the influence of islandarea on immigration rates of non-volant mammals Oikos 57297ndash300

142 Petraitis PS Latham RE 1999 The importance of scale in testing theorigins of alternative community states Ecology 80429 ndash 442

143 Robeson MS King AJ Freeman KR Birky CW Martin AP SchmidtSK 2011 Soil rotifer communities are extremely diverse globally butspatially autocorrelated locally Proc Natl Acad Sci U S A 1084406 ndash4410

144 Robinson JV Edgemon MA 1988 An experimental evaluation of theeffect of invasion history on community structure Ecology 691410 ndash1417

145 Whitaker RJ Grogan DW Taylor JW 2003 Geographic barriers isolateendemic populations of hyperthermophilic archaea Science 301976 ndash978

146 Cottenie K 2005 Integrating environmental and spatial processes inecological community dynamics Ecol Lett 81175ndash1182

147 Meyer O 1994 Functional groups of microorganisms In Schulze EDMooney HA (ed) Biodiversity and ecosystem function Springer-VerlagBerlin Germany

148 Jones CM Hallin S 2010 Ecological and evolutionary factors underly-ing global and local assembly of denitrifier communities ISME J 4633ndash641

149 Paul EA Clark FE 1996 Soil microbiology and biochemistry 2nd edAcademic Press London United Kingdom

150 Langenheder S Lindstrom ES Tranvik LJ 2005 Weak coupling be-tween community composition and functioning of aquatic bacteriaLimnol Oceanogr 50957ndash967

151 Allison SD 2012 A trait-based approach for modelling microbial litterdecomposition Ecol Lett 151058 ndash1070

152 Follows MJ Dutkiewicz S Grant S Chisholm SW 2007 Emergentbiogeography of microbial communities in a model ocean Science 3151843ndash1846

153 Raes J Letunic I Yamada T Jensen LJ Bork P 2011 Toward molec-ular trait-based ecology through integration of biogeochemical geo-graphical and metagenomic data Mol Syst Biol 7473 doi101038msb20116

154 Hutchinson G 1961 Paradox of plankton Am Nat 95137ndash145155 Chase JM Leibold MA 2002 Spatial scale dictates the productivity-

biodiversity relationship Nature 416427ndash 430156 Chesson P 2000 Mechanisms of maintenance of species diversity

Annu Rev Ecol Syst 31343ndash366157 Fraser C Alm EJ Polz MF Spratt BG Hanage WP 2009 The bacterial

species challenge making sense of genetic and ecological diversity Sci-ence 323741ndash746

158 Fukami T 2008 Stochasticity in community assembly and spatial scalep 51ndash71 In Ohgushi T Kondoh M Noda T (ed) Community ecologyvol 5 Kyoto University Press Kyoto Japan (In Japanese)

159 Woodcock S Curtis TP Head IM Lunn M Sloan WT 2006 Taxa-area relationships for microbes the unsampled and the unseen EcolLett 9805ndash 812

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Diana R Nemergut is an Associate Professor atthe University of Colorado in the Environmen-tal Studies program and the Institute of Arcticand Alpine Research She is currently serving asa Program Director in Antarctic Organisms andEcosystems at the National Science FoundationHer research focuses on the evolution and ecol-ogy of environmental microbial communitiesShe received her PhD from the University ofColorado and was a postdoctoral fellow at Rut-gers University She was inspired to go into mi-crobiology after working as a technician in Dr Joan Bennettrsquos laboratory atTulane University in 1997

Steven K Schmidt is Professor and Chair of theDepartment of Ecology and Evolutionary Biol-ogy University of Colorado Boulder His re-search interests include biogeochemistry of ex-treme ecosystems fungal ecology soilmicrobiology and microbe-plant interactionsDr Schmidt has published extensively on thebiogeochemistry and microbial diversity ofhigh-mountain ecosystems around the worldincluding the High Himalayas the Andes andmany other ranges His latest research projectsfocus on the function and diversity of microbes found on the Earthrsquos highestvolcanoes (Llullaillaco and Ojos del Salado) and the microbes that are en-demic to high-elevation snow packs

Tadashi Fukami is an Assistant Professor in theDepartment of Biology at Stanford UniversityHe studies how species assemble into ecologicalcommunities focusing on historical contin-gency or when and why the structure and func-tion of communities are contingent on the his-tory of species immigration He earned abachelorrsquos degree from Waseda University in1996 a masterrsquos degree from the University ofTokyo in 1998 and a PhD from the Universityof Tennessee Knoxville in 2003 After postdoc-toral work at Landcare Research New Zealand he was an Assistant Professorat the University of Hawaii at Manoa before moving to Stanford

Sean P OrsquoNeill is an MA student in the De-partment of Ecology and Evolutionary Biologyat the University of Colorado and he is affili-ated with the Institute of Arctic and Alpine Re-search both in Boulder CO His current re-search is focused on the role of neutral andniche processes in microbial community as-sembly as well as the role of microbial commu-nities in landscape ecology and ecosystem man-agement practices He earned his bachelorrsquosdegree at the University of Colorado in 2008and has been engaged with research in microbial ecology since 2006

Teresa M Bilinski is a PhD student in the In-stitute for Arctic and Alpine Research and theDepartment of Ecology and Evolutionary Biol-ogy at the University of Colorado in BoulderHer dissertation research focuses on the role ofbacterial community ecology in groundwaterarsenic cycling She earned a bachelorrsquos degreeat the University of Oregon and a masterrsquos de-gree from the University of Colorado She willbegin a position as an Assistant Professor at StEdwardrsquos University in the Fall of 2013

Lee F Stanish is a postdoctoral researcher in theChemistry and Geochemistry Department atthe Colorado School of Mines Her research fo-cuses on differentiating the relative importanceof physical biogeographic and biological pro-cesses in structuring microbial communitieswith a focus on finding common patterns in theecological responses of phylogenetically diversemicroorganisms She earned a PhD from theUniversity of Colorado at Boulder where sheexplored the role of hydrology and species in-teractions in structuring microbial mat communities in Antarctic streamsHer current research applies ecological principles to engineered systems forenhancing the productivity of algal biofuels

Joseph E Knelman is a PhD student at theUniversity of Colorado Boulder at the Insti-tute of Arctic and Alpine Research and in theDepartment of Ecology and Evolutionary Biol-ogy His research interests center around theecology of plant-soil microbe interactions fromnatural to agricultural systems His past workhas examined the impact of plant colonizationon microbial communities and nitrogen fixa-tion in early successional soils He received abachelorrsquos degree from Northwestern Univer-sity and a masterrsquos degree from the University of Colorado

John L Darcy is a PhD student in the Depart-ment of Ecology and Evolutionary Biology atthe University of Colorado Boulder He is in-terested in many aspects of microbiology andecology especially microbial phylogenetics andphylogeography His work seeks to understandwhich factors aid microorganisms in attainingtheir spatial distributions and to what extenttaxa differ in this regard He earned his bache-lorrsquos degree at the University of Colorado

Ryan C Lynch is a PhD candidate at the Uni-versity of Colorado in the Department of Ecol-ogy and Evolutionary Biology He is a memberof the Schmidt Laboratory where he studiesdesert soil Actinobacteria He uses experimentaland comparative population genomic ap-proaches to seek to uncover both novel adaptivebacterial traits and underlying evolutionaryprocesses He earned his bachelorrsquos degree atthe University of Colorado

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Philipp Wickey is a bachelorrsquos student with adual major in Physical Geography and Environ-mental Studies at the University of Coloradostudying the interaction between humans andthe environment Specifically he is interested indisturbance ecology and restoration His expe-rience also includes research in forest fire ecol-ogy in the American West megafauna extinc-tion in Australia and arthropod recovery inCosta Rica

Scott Ferrenberg is currently a doctoral candi-date in the Department of Ecology and Evolu-tionary Biology at the University of ColoradoHis research focuses on understanding how dis-turbance events influence speciesrsquo interactionsand how phenotypic variation within popula-tions affects processes within the larger biolog-ical community He earned his MS in 2002from the University of Maryland Department ofEntomology Among various positions hespent 2 years as a fire ecologist with the USGeological Survey and 3 years as an associate research scientist with theUniversity of Californiarsquos Sierra Nevada Research Institute studying the po-tential for speciesrsquo range shifts under climate-warming scenarios

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Diana R Nemergutab Steven K Schmidtc Tadashi Fukamid Sean P OrsquoNeillac Teresa M Bilinskiac Lee F Stanishab

Joseph E Knelmanac John L Darcyc Ryan C Lynchc Phillip Wickeyab Scott Ferrenbergc

Institute of Arctic and Alpine Research (INSTAAR)a Environmental Studies Programb and Ecology and Evolutionary Biologyc University of Colorado Boulder ColoradoUSA Department of Biology Stanford University Stanford California USAd

SUMMARY 342INTRODUCTION 342HOW ARE MICROBES UNIQUE 343DEFINITIONS 343

Communities 343Biodiversity 344

BIOGEOGRAPHICAL PATTERNS 345Abundance 345Taxon Turnover 345Phylogenetic Structure 346

VELLENDrsquoS CONCEPTUAL SYNTHESIS OF COMMUNITY ECOLOGY 346SELECTION 347DISPERSAL 347DIVERSIFICATION 348DRIFT 348COMBINING FORCES COMMUNITY ASSEMBLY 348TEMPORAL AND SPATIAL SCALES 349IMPLICATIONS FOR FUNCTION 350IMPLICATIONS FOR BIODIVERSITY 350ACKNOWLEDGMENTS 351REFERENCES 351AUTHOR BIOS 355

SUMMARY

Recent research has expanded our understanding of microbial com-munity assembly However the field of community ecology is inac-cessible to many microbial ecologists because of inconsistent and of-ten confusing terminology as well as unnecessarily polarizing debatesThus we review recent literature on microbial community assemblyusing the framework of Vellend (Q Rev Biol 85183ndash206 2010) inan effort to synthesize and unify these contributions We begin bydiscussing patterns in microbial biogeography and then describe fourbasic processes (diversification dispersal selection and drift) thatcontribute to community assembly We also discuss different combi-nations of these processes and where and when they may be mostimportant for shaping microbial communities The spatial and tem-poral scales of microbial community assembly are also discussed inrelation to assembly processes Throughout this review paper wehighlight differences between microbes and macroorganisms andgenerate hypotheses describing how these differences may be impor-tant for community assembly We end by discussing the implicationsof microbial assembly processes for ecosystem function and biodiver-sity

INTRODUCTION

Molecular phylogenetic approaches continue to revolutionizethe field of microbiology we now possess the tools to un-

derstand high-resolution details about the degree of variation inmicrobial community structure in both space and time (1ndash5)Sequencing costs have plummeted while the amount of publiclyavailable data has increased exponentially in recent years Com-

putational advances (6 7) as well as new standards for contextu-alizing environmental microbial community composition datasets (8) will allow us to make the most of these data facilitatingcross-investigator and cross-system meta-analyses Indeed afteryears of citing the many limitations of studying such complexsystems microbiologists now enjoy many advantages that our col-leagues who study macrobial communities actually lack Admit-tedly we are still a long way from a ldquocompleterdquo understanding ofany but the most simple of microbial communities which willrequire continual improvements in both technology and compu-tation Thus despite these recent advances we are faced withquestions about how to best sample microbial communities tomaximize what we can learn about how they are structured howthey function and how they change through time (9)

A unified conceptual framework of microbial community as-semblymdash one that incorporates our understanding of communityassembly from a macrobial ecology perspective while recognizingthe attributes that make microorganisms uniquemdashis needed tohelp direct the field of microbial ecology through this new eraThis is not an easy task and we argue that it is made more difficultby unnecessarily polarizing debates (eg the false dichotomy ofthe niche-versus-neutral debate [10] as well as the debates over

Address correspondence to Diana R Nemergut nemergutcoloradoedu

doi101128MMBR00051-12

342 mmbrasmorg Microbiology and Molecular Biology Reviews p 342ndash356 September 2013 Volume 77 Number 3

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DEFINITIONS

Communities

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Biodiversity

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Abundance

Taxon Turnover

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Process Description

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

SELECTION

DISPERSAL

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DEFINITIONS

Communities

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Biodiversity

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Abundance

Taxon Turnover

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Process Description

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

SELECTION

DISPERSAL

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Biodiversity

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Abundance

Taxon Turnover

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Process Description

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

SELECTION

DISPERSAL

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Abundance

Taxon Turnover

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Process Description

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

SELECTION

DISPERSAL

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Process Description

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

SELECTION

DISPERSAL

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

SELECTION

DISPERSAL

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

DIVERSIFICATION

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

ACKNOWLEDGMENTS

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Continued next page

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Nemergut et al

on Septem

ber 4 2013 by SE

OL Lane M

edical Libraryhttpm

mbrasm

ownloaded from

Documents

Patterns and Processes of Microbial Community Assembly