... Received: 16 May 2023 | Accepted: 1 December 2023 DOI: 10.1111/2041-210X.14279 RESEARCH ARTICLE A conceptual framework for host-associated microbiomes of hybrid organisms Benjamin T. Camper1 | Zachary Laughlin1 Robert Denton2 | Sharon Bewick1 1 Department of Biological Sciences, Clemson University, Clemson, South Carolina, USA 2 Biology Department, Marian University, Indianapolis, Indiana, USA Correspondence Benjamin T. Camper Email: btcampers@gmail.com Funding information NSF: Division of Integrative Organismal Systems, Grant/Award Number: 2105604; Clemson University Support for Early Exploration and Development (CUSEED) Grant; Clemson University Creative Inquiry (CI) Program Handling Editor: Antonino Malacrin | Daniel Malagon1 | Abstract 1. Hybridization between organisms from evolutionarily distinct lineages can have profound consequences on organismal ecology, with cascading effects on fitness and evolution. Most studies of hybrid organisms have focused on organismal traits, for example, various aspects of morphology and physiology. However, with the recent emergence of holobiont theory, there has been growing interest in understanding how hybridization impacts and is impacted by host-associated microbiomes. Better understanding of the interplay between host hybridization and host-associated microbiomes has the potential to provide insight into both the roles of host-associated microbiomes as dictators of host performance as well as the fundamental rules governing host-associated microbiome assembly. Unfortunately, there is a current lack of frameworks for understanding the structure of host-associated microbiomes of hybrid organisms. 2. In this paper, we develop four conceptual models describing possible relationships between the host-associated microbiomes of hybrids and their progenitor or parent taxa. We then integrate these models into a quantitative 4H index and present a new R package for calculation, visualization and analysis of this index. 3. We demonstrate how the 4H index can be used to compare hybrid microbiomes across disparate plant and animal systems. Our analyses of these data sets show variation in the 4H index across systems based on host taxonomy, host site and microbial taxonomic group. 4. Our four conceptual models, paired with our 4H index and associated visualization tools, facilitate comparison across hybrid systems. This, in turn, allows for systematic exploration of how different aspects of host hybridization impact the host-associated microbiomes of hybrid organisms. KEYWORDS Aitchison simplex, holobiont, host-associated microbiome, hybrid, R package This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. 2024 The Authors. Methods in Ecology and Evolution published by John Wiley & Sons Ltd on behalf of British Ecological Society. Methods Ecol Evol. 2024;15:511529. wileyonlinelibrary.com/journal/mee3 | 511 | 1 | CAMPER et al. I NTRO D U C TI O N be important determinants of dietary niche (Blyton et al., 2019; Greene et al., 2020; Heys et al., 2021; Kohl et al., 2014; Moeller & Hybridization is increasingly recognized as an important compo- Sanders, 2020), either by provisioning hosts with key nutrients (Hu nent of ecological and evolutionary processes. Consequences of et al., 2018; Jing et al., 2020; Ju et al., 2020) or by detoxifying de- hybridization span the fitness spectrum ranging from infertility and fensive compounds found in host food sources (Zheng et al., 2016). death (Brucker & Bordenstein, 2013; Zhang et al., 2014) to innova- Beyond diet and metabolism, HA microbiomes influence a range of tion and adaptation (Abbott et al., 2013; Dowling & Secor, 1997; other host traits as well (Archie & Theis, 2011; Bravo et al., 2011; Patton et al., 2020; Seehausen, 2004). Ultimately, these fitness Davidson et al., 2018; Ezenwa et al., 2012; Gaona et al., 2016; consequences dictate the role that hybridization plays in the suc- Grinberg et al., 2022; Jia et al., 2021; Kirchoff et al., 2019; Neufeld cess or failure of different genetic lineages (Seehausen, 2004; Pala et al., 2011; Sampson & Mazmanian, 2015; Sharon et al., 2010). & Coelho, 2005; Larouche et al., 2020; Todesco et al., 2016). If, for Healthy gut (Chen et al., 2018; Kamada et al., 2013), skin (Chen example, hybridization produces sterile offspring, then it can drive et al., 2018; Harris et al., 2006; Kueneman et al., 2014) and vagi- the emergence of genetic sinks and evolutionary dead ends (Tripp & nal microbiomes (Brotman et al., 2010), for example, provide patho- Manos, 2008) and thus serve as a brake for evolution. Alternatively, gen resistance across a broad spectrum of animal species (Buffie & if hybridization facilitates ecological release and/or sexual isola- Pamer, 2013; Ubeda et al., 2017; Woodhams et al., 2016). Indeed, tion (either directly through mating barriers or indirectly through amphibian skin microbiomes have been extensively studied as altered temporal or spatial proximity), then it can promote lineage a means of defending hosts from devastating fungal pathogen diversification and thus serve as a motor for evolution (Heard & (Batrachochytrium dendrobatidis and B. salamandrivorans) epidemics Hauser, 1995). (Bates et al., 2018, 2022; Rebollar et al., 2016, 2020). In humans, Most early research on hybrid organisms focused on under- disruptions to healthy HA microbiomes also underly a range of non- standing how hybridization impacts host fitness through effects infectious diseases (Ahn et al., 2013; Zackular et al., 2013) such as on host traits, for example, fecundity (Campbell et al., 2006; rheumatoid arthritis (Bergot et al., 2019; Scher & Abramson, 2011) Dobzhansky, 1934; Forejt, 1996; Hovick & Whitney, 2014; Reed and irritable bowel syndrome (Chong et al., 2019; Pimentel & & Sites Jr, 1995), physiology (Brown & Bouton, 1993; Cooper & Lembo, 2020). Ultimately, the cascading effects of HA microbiomes Shaffer, 2021; Lafarga-De la Cruz et al., 2013; Martins et al., 2019; on host traits and processesranging from host energy balance and Pereira et al., 2014), morphology (Capblancq et al., 2020; Carreira dietary niche through disease risk and immune dysfunctionhave et al., 2008; Jackson, 1973; Mrot et al., 2020) and behaviour (Robbins strong consequences on host ecological success (Abbott et al., 2021) et al., 2010, 2014). Recently, however, there has been growing recog- and, by extension, host evolution (Kolodny et al., 2020; Opstal, & nition that macroorganisms are not autonomous units. Rather, they Bordenstein, 2015; Zilber-Rosenberg & Rosenberg, 2008). are collectives or holobionts compromised of both a host and all of Although there has been substantial literature document- its host-associated (HA) microbes (Baedke et al., 2020; Bordenstein ing both coevolutionary (Ehrlich & Raven, 1964; Janz, 2011; & Theis, 2015; Bosch & Miller, 2016; Margulis & Fester, 1991). Thus, Janzen, 1980; Thompson, 1994, 2005) processes and codiversi- just as it is important to understand how hybridization impacts fication patterns (Janz, 2011; Nishida & Ochman, 2021; Suzuki the traits of the host, it is equally important to understand how et al., 2022; Thompson, 1989) between hosts and their HA micro- hybridization impacts the traits of the holobiont, including charac- biomes (Apprill et al., 2020; Chiarello et al., 2018; Ley et al., 2008; teristics of the HA microbiome (Miller et al., 2021). Indeed, the eco- Meadows, 2022; Moran & Sloan, 2015; Ochman et al., 2010; Phillips evolutionary basis for holobionts has led to entirely new branches of et al., 2012; Sanders et al., 2014; Scheelings et al., 2020; Walker research in areas as diverse as human health (Postler & Ghosh, 2017; et al., 2019), the study of how HA microbiomes respond when di- Walter et al., 2013), conservation (Bahrndorff et al., 2016; Banerjee vergent host lineages reunite, or admix, through hybridization is et al., 2020; Carthey et al., 2020; Jimnez & Sommer, 2017; Jin Song relatively new (Malukiewicz et al., 2019). One of the earliest inves- et al., 2019; Maebe et al., 2021; Redford et al., 2012; Trevelline tigations into hybrid microbiomes was in Nasonia wasps (Brucker & et al., 2019; West et al., 2019; Zhu et al., 2021) and biotechnology Bordenstein, 2013). In this system, up to 90% lethality is observed in (Bredon et al., 2020; Ren et al., 2022), and it is currently poised to do F2 males of N. vitripennis/N. giraulti crosses. However, rearing wasps so within the field of hybridization research as well. under germ-free conditions results in near complete rescue of the The importance of the holobiont concept stems from the many same F2 males. This suggests a microbial basis to hybrid lethality. host traits and processes that are either partially or fully depen- Interestingly, the 10% of hybrid N. vitripennis/N. giraulti males that dent on host-associated microbes (Fontaine & Kohl, 2020; Friesen survive under natural conditions exhibit highly transgressive micro- et al., 2011; Nobs et al., 2019; Walters et al., 2020). As an example, bial phenotypes. This includes both the appearance of novel micro- gut microbiomes are strong regulators of host metabolic phenotype bial taxa in hybrid microbiomes as well as shifts in the abundances of (Claus et al., 2008; Li et al., 2008; Mayneris-Perxachs et al., 2016). microbial taxa that are shared among parents and hybrids. This, in turn, impacts host energy balance (Corbin et al., 2020; More recent studies on hybrid vertebrates paint a similar picture. Nieuwdorp et al., 2014; Xifra et al., 2019), including both energy For example, hybrid house mice (Mus musculus musculus and Mus m. intake as well as use and expenditure. Gut microbiomes can also domesticus) in central Europe (Wang et al., 2015) exhibit widespread 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 512 transgressive microbiomes. Furthermore, like the Nasonia wasp 513 host genetics and the environment impact HA microbiomes across system, there is evidence that the altered microbial pheno- a Neotoma woodrat hybrid zone, Nielsen et al. (2023) demonstrated types of hybrid individuals at least partially explain their poor fit- that HA microbial composition was predominately driven by host ness outcomes (Baird et al., 2012; Britton-Davidian et al., 2005; genetics (genotypic classes), while HA microbial richness was pre- Forejt & Ivnyi, 1974; Good et al., 2008; Sage et al., 1986; Turner dominately driven by the environment (core diet + vegetation com- et al., 2012). In particular, there is an interaction between inflamma- munities). Applying similar approaches to other hybrid systems may tion, immune gene expression and the gut microbiome that appears be a fruitful avenue for disentangling the long-standing nature ver- to cause hybrid mice to exhibit defects in immunoregulation. This sus nurture paradigm as it applies to HA microbiomes and HA micro- may be one reason why hybrid individuals are restricted to a nar- biome assembly. row tension zone where the two parent subspecies co-occur (Balard Despite the increasing recognition that HA microbiomes are & Heitlinger, 2022; Barton & Hewitt, 1985). A range of additional an important facet of hybridization and that hybrid organisms are studies, including hybridization of sika deer (Cervus nippon) and elk valuable systems for understanding HA microbiome structure and (Cervus elaphus) (Li et al., 2016), lake whitefish lineages (Coregonus function, there is a lack of frameworks for describing and compar- clupeaformis) (Sevellec et al., 2019), blunt snout bream (Megalobrama ing hybrid HA microbiomes across the tree of life. In this paper, we amblycephala) and topmouth culter (Culter alburnus) (Li et al., 2018) develop four conceptual models delineating potential relationships and desert (Neotoma lepida) and Bryant's (Neotoma bryanti) woodrats between hybrid microbiomes and the microbiomes of their progen- (Nielsen et al., 2023) have reiterated the finding that hybrid animals itors. We discuss the underlying implications of each model, how often exhibit altered microbiomes relative to their progenitors (i.e. each model might arise based on fundamental host mechanisms and parent lineages or parent taxa). Indeed, even beyond the animal how each model could impact host fitness. We then integrate these kingdom, hybrid macroorganisms are commonly associated with four models into a quantitative 4H index that can be used to assess perturbations to the HA microbiome (Cregger et al., 2018; O'Brien the relative importance of each model across widely disparate hybrid et al., 2019; Wagner et al., 2020). systems. Finally, we introduce an R package, HybridMicrobiomes As suggested above, the study of hybridization and its impact (https://c ran.r-p rojec t.o rg/w eb/p ackag es/H ybrid Micro biome s/ on HA microbiomes is important for understanding host fitness and index.html), containing a series of functions that allow researchers evolution (Baeckens, 2019; Muoz & Bodensteiner, 2019). However, to apply the 4H index to their own hybrid microbiome data sets. even beyond host success, hybrid systems are of interest because they facilitate an understanding of genotypephenotype interactions (Kearney, 2005; Kratochwil & Meyer, 2015). Many hybrid zones (Cooper & Shaffer, 2021; Robbins et al., 2010; Walls, 2009), especially systems where F2 individuals readily admix with their 2 | M ATE R I A L S A N D M E TH O DS 2.1 | Conceptual models progenitors, provide variable genetic combinations (Lee et al., 2017; Pfennig, 2021) and degrees of heterozygosity across hybrid individ- We propose four conceptual modelsthe Union Model, the uals. Consequently, these systems serve as natural laboratories for Intersection Model, the Gain Model and the Loss Modelto describe understanding how host genetics and environmental characteris- the potential relationships between the HA microbiomes of hybrid tics influence host traits. For example, in a study investigating how individuals and those of their progenitors (see Figure 1). These four F I G U R E 1 The limiting scenarios for each of our four conceptual models describing the host-associated microbiomes of hybrid organisms. Hybrid organisms can (a) host all of the microbial taxa found on either progenitor (Union), (b) host only those microbial taxa found on both progenitors (Intersection), (c) host only novel microbial taxa found on neither progenitor (Gain) or (d) be missing all microbial taxa found on one or both progenitors (Loss). Note that, in the final scenario, assuming that there is no gain of microbial taxa, the hybrid has no microbiome at all. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | CAMPER et al. models represent extreme or limiting scenarios, each portraying an invasion (McCann, 2000). If this is true for HA microbial communi- idealized relationship between hybrid and progenitor HA microbi- ties, then hybrids following the Union Model may gain the advantage omes. Realistic hybrid microbiomes can then be described as differ- of having a more resilient microbiome that is more resistant to colo- ing combinations of these four idealized models. In what follows, we nization by pathogens (Harrison et al., 2019). delineate the four models, discuss possible underlying mechanisms However, there are likely costs to the Union Model as well. and outline their potential for impacting hybridization outcomes. For Most obvious are the challenges of bringing together large num- the sake of simplicity, we introduce each of the models within the bers of distinct microbial taxa from different progenitors. Consider framework of microbiome membership (taxon incidence) rather than BatesonDobzhanskyMuller (BDM) incompatibilities (Muller, 1942; composition (taxon abundance). However, comparable arguments Orr, 1995; Orr & Turelli, 2001), which emerge in hybrid organisms can be made for abundance relationships between progenitor and due to mismatches between the genes from their two progenitors. hybrid microbiomes as well. If BDM incompatibilities are a common outcome of combining different progenitor genomes, then analogous mismatches that result from combining different microbial metagenomes should also 2.1.1 | The Union Model be possible. Furthermore, there could be mismatches between the microbial metagenome from one progenitor and the host genome In its most extreme form, this model implies that hybrid microbiomes from the other, as outlined in the microbial-assisted BDM model are comprised of all microbial taxa present on at least one progenitor by Brucker and Bordenstein (2012). Indeed, if BDM incompatibilities and nothing else (see Figure 1a). This could occur if carrying a particu- scale with genome size (a tenuous assumption), hostmicrobe and lar host genome fosters colonization by associated microbial taxa. microbemicrobe incompatibilities may be more likely than tradi- Notably, such fostering could emerge either directly through host tional BDM incompatibilities simply because the microbial metag- interactions with the microbe (e.g. if specific hybrid and/or progeni- enome is typically much larger than the genome of the host itself. tor morphologies provide housing for symbiotic microbes (Belcaid Whether or not this is the case, the Mus musculus and Nasonia sys- et al., 2019; Delaux & Schornack, 2021; Fronk & Sachs, 2022)) or tems suggest that microbial-assisted BDMs are certainly a possibility indirectly through effects on host behaviour or ecology (e.g. if hy- among hybrid HA microbiome systems. brids colonize a progenitor's environment and subsequently acquire environmental microbes). To the extent that hybrid individuals share genetic material from both progenitors (note that this may vary de- 2.1.2 | The Intersection Model pending on the extent of back-crossing), hybrids should support all microbes present on either progenitor. Said differently, in the Union In its most extreme form, this model suggests that hybrid microbi- Model, the host genome acts as a ticket for acquiring a particu- omes are comprised of all microbial taxa simultaneously present on lar microbiome. Having two tickets (i.e. each representing a unique both progenitors and nothing else (see Figure 1b). Note that, within genomic component) results in the acquisition of two microbiomes, the framework of our conceptual models, the Intersection Model is one from each progenitor. a subset of the Union Model (i.e. the Union Model comprises all mi- Hybrids characterized exclusively by the Union Model (see crobial taxa found on one or both progenitors, while the Intersection Figure 1a) should have more taxonomically diverse microbiomes than Model comprises only those microbial taxa found on both progeni- either progenitor. Importantly, greater taxonomic diversity could re- tors). We describe our conceptual models in this way because it best sult in greater functional diversity as well (Petchey & Gaston, 2002), reflects potential mechanisms by which Union and Intersection of with important consequences for host health and ecological perfor- HA microbiomes may emerge. However, a slightly different defini- mance. Consider a thought experiment wherein two different insect tion of the Union model (not including microbial taxa found on both species are each limited to a distinct set of host plants based on the progenitors) is used for the 4H index. This is done to avoid dou- need for gut microbial detoxification of plant defensive compounds. ble counting components of microbial diversity (see Section 2.2). If the hybrid offspring of these two insect species harbour the gut Further note that, unlike the Union Model, a hybrid cannot be exclu- microbiomes of both progenitors, then hybrid microbiomes should sively characterized by the Intersection Model unless there are no be able to detoxify both sets of host plants, allowing hybrids to uti- microbial taxa unique to one of the two progenitors. This is because lize all resources open to either progenitor. More broadly, greater a hybrid that only harbours microbial taxa found on both progenitors functional capacity of hybrid microbiomes could enable hybrids must, in addition, have lost all microbial taxa found on only one of to persist in habitats that are intermediate to their progenitors or the two progenitors. The inter-relatedness of the Intersection and across all habitats colonized by either progenitor. Beyond expanded Loss models is discussed more below. function, a more diverse hybrid microbiome may have other bene- The Intersection Model could occur if a particular host genome fits as well. Although contentious, both in microbiome (Deng, 2012; hinders or prevents colonization by unassociated microbial taxa. He et al., 2013; Wagg et al., 2018) and general ecology literature, Again, the underlying mechanism could be direct (e.g. changes in the diversity (Ives & Carpenter, 2007; McCann, 2000) has long been as- host immune system) or indirect (e.g. changes in host behaviour that sociated with lower temporal variability and increased resistance to alter exposure to environmental microbes). In either case, hybrids 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 514 515 that carry genetic material from both progenitors will be more re- potentially different detoxification properties as compared to their fractory to, or isolated from, a wider range of microbial taxa. Said progenitors. Thus, rather than being able to use both of their pro- differently, in the Intersection Model, each host genome acts as a genitor's host plants (Union Model) or neither of their progenitor's gate. Having two gates blocks a wider range of microbes, leaving host plants (Intersection Model), these saltational hybrids could only those taxa that are permitted access by both progenitors. Thus, potentially colonize an entirely novel set of host plants not used by hybrids characterized by the most extreme form of the Intersection either progenitor. Model (see Figure 1b) should have less taxonomically diverse micro- More so than the Union, Intersection or Loss models, the Gain biomes which could have consequences for functional diversity as Model is responsible for phenotypic novelty and thus, provides well. For instance, in our previous insect example (see The Union the building blocks for evolutionary innovation. This could result in Model), the Intersection Model could leave hybrids without the rapid adaptation, escape from competition with their progenitors or ability to detoxify either set of progenitor host plants, placing a sub- even reproductive isolation. Indeed, in some cases, the Gain Model stantial limitation on feeding opportunities. This, in turn, could have may actually accelerate speciation (Mallet, 2007). However, the impacts on fitness, leading to higher rates of starvation, underper- Gain Model may have non- or maladaptive consequences as well. formance due to toxin build-up or even poisoning directly. Similar Notably, there is no a priori reason to believe that the acquisition negative effects on survival could be possible due to more general of large numbers of novel microbial taxa will be generally beneficial mechanisms associated with microbial diversity as well, for example, to a host. In fact, there are many reasons to believe the opposite. the loss of microbiome stability and pathogen resistance. The bene- In particular, the Gain Model describes a scenario of rapid evolu- fit of the Intersection Model, of course, is that it virtually eliminates tionary change (i.e. the introduction of novel microbial metagenomic opportunities for microbehost or microbemicrobe incompatibili- content to the hybrid) that occurs far outside the confines of more ties. This is because, in the Intersection Model, all microbehost and typical hostmicrobe coevolutionary relationships forged over gen- microbemicrobe interactions that occur on the hybrid are already erations of symbiosis. As a result, the Gain Model exemplifies a high present on both progenitors. risk, high reward scenario, and novel microbes acquired by the hybrid could just as easily enhance or reduce host fitness. Thus, like Goldschmidt's hopeful monsters, the Gain Model relies on happy 2.1.3 | The Gain Model accidents (Ross, 1983-1994) meaning that many hybrid individuals are likely to fail for each ecological success. In its most extreme form, this model suggests that hybrid microbiomes only include microbial taxa not present on either progenitor (see Figure 1c). Like the Intersection Model, a hybrid cannot be 2.1.4 | The Loss Model exclusively characterized by the Gain Model unless the progenitor microbiomes are fully devoid of microbial taxa. Again, this is because In its most extreme form, this model suggests that hybrid microbi- a hybrid that only harbours novel microbial taxa must, in addition, omes are missing all microbial taxa that are present on one or both have lost all microbial taxa found on the two progenitors. The Gain progenitors and have gained no new microbial taxa (see Figure 1d). Model is possible if HA microbiomes are idiosyncratically sensitive In other words, the most extreme form of the Loss Model implies to specific gene combinations that arise from merging progenitor that hybrids have no microbiome at all. Again, this is unrealistic. genomes. Broadly speaking, the Gain Model is the microbial equiva- Thus, just as the Intersection and the Gain models cannot occur re- lent of Bateson's saltational evolution (Bateson, 1984, 2002) or alistically independent of the Loss Model, nor can the Loss Model Goldschmidt's hopeful monsters (Goldschmidt, 1933, 1940). Like occur realistically independent of at least one or more of the other Bateson's and Goldshmidt's models, the Gain Model posits that models. The non-independence of the various models reflects the hybridization can yield profound (saltational) changes in pheno- fact that, except in very special and typically non-realistic scenarios type (Theien, 2006, 2009), and that these phenotypic changes (e.g. when the progenitors or hybrids have no HA microbes), real- may enable hybrids to establish an entirely novel ecological niche ized systems will always be combinations of the idealized models. relative to their progenitors (Dittrich-Reed & Fitzpatrick, 2013; The idealized models, however, serve as limits that emerge out of Goldschmidt, 1933; Mallet, 2007). However, unlike Bateson and various scenarios by which host genomes, and in particular hybrid Goldschmidt, who focused on host genes, the Gain Model assumes host genomes, could feasibly impact HA microbiome assembly. As that there are underlying microbial dimensions to the saltational in the Gain Model, the loss of microbes present on both progenitors change. Arguably, adding microbial dimensions provides even more describes a saltational scenario that is possible if HA microbiomes opportunity for saltational change, again because of the vast size are idiosyncratically sensitive to gene combinations of the progeni- and diversity of functions encompassed by the microbial metage- tors. In contrast to the Gain Model, however, the saltational change nome relative to the host genome itself. Once more, consider our invoked by the Loss Model is the deletion, rather than the addition, hypothetical insect example (see The Union Model). Hybrid insects of microbial taxa. characterized by the most extreme form of the Gain Model should The Loss Model gives rise to hybrid microbiomes with lower harbour an entirely new set of gut bacteria with novel taxa and overall diversity and potentially lower functional capacity as well. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | CAMPER et al. Indeed, as suggested above, in the limit of a hybrid organism ex- like to consider abundance thresholds have the option to do so. clusively characterized by the Loss Model, there would be no host- Thus, both the 4H index and the HybridMicrobiomes R package pro- associated microbiome at all. Returning, for the last time, to our vide flexibility that can be decided within the context of a particular hypothetical insect complex (see The Union Model), the Loss Model hybrid system (though a common set of parameters should be used predicts that hybrid insects should lack microbes present on one or for any comparison between hybrid systems). both progenitors. In the case of the latter, hybrids would lose the For the 4H index, , and are the same across all host classes. ability to detoxify host plants that are usable by both of their progen- However, each host class (i.e. each progenitor and the hybrid) is sep- itors. Like the Intersection Model, this could limit opportunities for arately assigned its own core microbiome. Thus, a microbial taxon feeding, cause toxin build-up or result in poisoning of hybrid insects. is part of a host's core microbiome provided it is found on at least Lower microbiome diversity could also lead to a suite of additional N hosts of that host class at a minimum average abundance of challenges like greater microbiome instability and lower pathogen and a minimum abundance on at least one host of , where N is the resistance. Again, however, the costs of low diversity microbiomes number of hosts of each class and should be the same across all host may be balanced out by the benefits of reduced opportunities for classes (i.e. a balanced design with equal numbers of each progenitor hostmicrobe or microbemicrobe incompatibilities. and the hybrid; note that the HybridMicrobiomes package includes bootstrapping steps that will downsample data sets such that a bal- 2.2 | The 4H index and quaternary plots anced design is achieved). In what follows, we describe four versions of the 4H index, two based on incidence of microbial taxa and two based on abundance of microbial taxa. While the four conceptual models in Section 2.1 present limiting, extreme or idealized scenarios, any realistic hybrid system will almost certainly exhibit mixed support across two or more conceptual 2.2.1 | Incidence-based analyses models. To examine the importance of each of the four conceptual models to any given hybrid system, we introduce the 4H index, along Our two incidence-based methods are inspired by the Jaccard index with R package HybridMicrobiomes (https://cran.r-projec t.org/ (Jaccard, 1908) and the Sorensen index, respectively (Dice, 1945; web/packages/HybridMicrobiomes/index.html), which can be used Sorensen, 1948). For any given , and , we define P1, P2 and H as to calculate and graph the 4H index for any hybrid system. The 4H the set of core microbial taxa present on the first progenitor, the index uses the core microbiomes of each host class (where we use second progenitor and hybrids. We then determine the number of host class to refer to any one of the three types of hoststhe first microbial taxa shared by different combinations of hybrid and pro- progenitor, the second progenitor or the hybridin a hybrid com- genitor classes. Specifically, we define: plex) to determine which microbial taxa are lost and gained on hybrid organisms relative to their progenitors. To define the core microbiome, we use a tunable parameter, , which can range from = 1 (i.e. microbial taxa are only considered if they are present on every host of a particular host class) to = 0 (i.e. the full microbiome; all microbial taxa are considered regardless of the number of hosts they are found on). Consistent with the common definition of a core microbiome, we typically select higher values of . This is based on the assumption that microbial taxa with strong consequences for host ecology and/or evolution should be detectable on the majority of hosts within a population. However, researchers who have reason to suspect otherwise can use a lower value of or can compare the 4H index across a range of values (see Figures S2.1S2.4; Tables S2.1 and S2.2). While core microbial taxa are usually defined as those present on a threshold number of hosts, alternate definitions exist that incorporate microbial abundances as well (Shade & Stopnisek, 2019). To ) | |( a = | P1 P2 H|, | | ) | |( b = | P1 P2 H| a, | | (1a) (1b) b1 = ||P1 H|| a, b2 = ||P2 H|| a, c = |H| a b, (1c) d = ||P1 P2 H|| a b c, (1d) d1 = ||P1 || ||P1 P2 || ||P1 H|| + a, allow for this, we include a second threshold, , based on the average d2 = ||P2 || ||P1 P2 || ||P2 H|| + a, relative abundance (across all hosts within a class) that a microbial d12 = d d1 d2 . taxon must reach to be considered part of the core. In addition, we include a third threshold, , based on the minimum relative abundance that a microbial taxon must reach on at least one host to be In Equation (1), |S| denotes the cardinality of set S, where S is any set. Accordingly, a is the number of microbial taxa shared by both progen- considered part of the core. By default, the HybridMicrobiomes R itors and the hybrid, b is the number of microbial taxa shared by one package sets both = 0 and = 0. However, researchers who would progenitor (but not both) and the hybrid, b1 is the number of microbial 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 516 taxa shared by the first progenitor and the hybrid, b2 is the number of c , 3a + 2b + c + d1 + d2 + 2d12 (3c) d1 + d2 + 2d12 = 1 . 3a + 2b + c + d1 + d2 + 2d12 (3d) = microbial taxa shared by the second progenitor and the hybrid, c is the number of microbial taxa found only on the hybrid, d is the number of microbial taxa found only on one or both progenitors, d1 is the number of microbial taxa found only on the first progenitor, d2 is the number of = 517 microbial taxa found only on the second progenitor and d12 is the num- Notice that the difference between the Jaccard-and Sorensen-inspired ber of microbial taxa found only on both progenitors. For the Jaccard- methods is whether taxa found on multiple host classes are weighted inspired method, we define the four dimensions of the 4H index (three according to the number of host classes that they occur on. Similar independent dimensions) as: to the Jaccard index for beta diversity, the Jaccard-inspired 4H index only counts the unique microbial taxa shared by each combination of = b , a+b+c+d (2a) host classes. By contrast, like the Sorensen index for beta diversity, the Sorensen-inspired 4H index triple weights microbial taxa shared by all three host classes and double weights microbial taxa shared by two of b1 1 = , a+b+c+d the three host classes. Function FourHbootstrap in HybridMicrobiomes takes a phyloseq object (McMurdie & Holmes, 2013) and a vector specifying 2 = b2 , a+b+c+d = a , a+b+c+d (2b) c , a+b+c+d (2c) d = 1 , a+b+c+d (2d) progenitor and hybrid classifications. It then calculates the 4H index over bootstrapped samples of hybrid organisms and their progenitors. FourHbootstrap outputs a data frame with the percentage of microbial taxa that fall into each of the four models (see Equation 1). The data frame also includes the fraction of progenitor microbial taxa that are found on both progenitors. Finally, the data frame breaks = = the Union Model into separate components attributable to the first progenitor and the second progenitor, respectively ( 1 + 2 = ). 2.2.2 | Abundance-based analyses where , , and reflect the extent of Union, Intersection, Gain and Loss models, respectively. Briefly, is the fraction of microbial Like incidence, we include two different abundance-based methods taxa found on hybrids and on one (but not both) progenitor (note that for calculating the 4H index, with the first inspired by the Ruzicka this is a slight deviation from the conceptual Union Model, which does index (Legendre, 2014) and the second inspired by the BrayCurtis not distinguish between taxa found on one or both progenitors. This index (Bray & Curtis, 1957). Similar to incidence-based methods, the deviation is necessary to avoid double counting microbial taxa in the two abundance-based methods also focus on core microbial taxa as 4H index. Also note that can be divided into a component that the defined by , and . However, abundance-based methods require hybrid shares with the first progenitor, 1, and a component that the an additional pre-step to find representative microbial abundances hybrid shares with the second progenitor, 2). is the fraction of mi- for each host class. This step is performed on the full microbiome crobial taxa found on hybrids and on both progenitors, is the fraction (i.e. core and non-core microbial taxa) based on either the mean or of microbial taxa only found on hybrids, and is the fraction of micro- median relative abundance of each microbial taxon on each host bial taxa only found on progenitors. class. These representative abundances can then be used raw or can Similarly, for the Sorensen-inspired method, we define the four dimensions of the 4H index (three independent dimensions) as: be renormalized based only on microbial taxa that comprise the core of each host class. Renormalization results in a metric that is density invariant (i.e. does not vary with the number of reads attributed to = 2b , 3a + 2b + c + d1 + d2 + 2d12 (3a) the core microbiome of each host class) (Jost et al., 2011). However, a downside of renormalization is that it constrains the 4H index to a two-dimensional plane (the same is true when using = 0 since the 2b1 1 = , 3a + 2b + c + d1 + d2 + 2d12 FourHbootstrapA function rarefies microbiome data sets, thereby forcing all hybrid classes to have equivalent numbers of reads). By contrast, using raw reads allows the total number of reads to differ 2b2 2 = , 3a + 2b + c + d1 + d2 + 2d12 between host classes. While this results in a metric that is not den- 3a = , 3a + 2b + c + d1 + d2 + 2d12 we view raw reads as the preferred option. This is because full micro- sity invariant (i.e. it changes with the number of reads attributed to the core of a particular host class, see Supplemental Information S1), (3b) biomes are rarefied to the same number of reads prior to selection 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | CAMPER et al. of the core. Thus, differences in read numbers attributed to the core C , A+B+C+D (5c) D = 1 . A+B+C+D (5d) = reflect biologically meaningful differences in composition. To generalize our incidence-based 4H index to account for abundance, we follow the method in Tams et al. (2001) and define: A= B= S { i=1 S i=1 = ( ( ) ) S min min xiP1 , xiP2 , xiH = i . i=1 (4a) Similarly, for the BrayCurtis-inspired method, we define the four dimensions of the 4H index (three independent dimensions) as: ( ) ( )} S { } min xiP1 i , xiH i + min xiP2 i , xiH i = 1i + 2i , i=1 (4b) B1 = S { ( )} S min xiP1 i , xiH i = 1i , B2 = S { ( )} S min xiP2 i , xiH i = 2i , i=1 i=1 C= D= i=1 = 2B , 3A + 2B + C + D1 + D2 + 2D12 1 = 2B1 , 3A + 2B + C + D1 + D2 + 2D12 2 = 2B2 , 3A + 2B + C + D1 + D2 + 2D12 = 3A , 3A + 2B + C + D1 + D2 + 2D12 (6b) = C , 3A + 2B + C + D1 + D2 + 2D12 (6c) D1 + D2 + 2D12 = 1 . 3A + 2B + C + D1 + D2 + 2D12 (6d) i=1 S { } xiH i 1i 2i , (4c) i=1 S { ( )} xiP1 + xiP2 i 1i 2i min xiP1 , xiP2 , i=1 D1 = S { D2 = S { i=1 ( )} ( )} xiP1 1i min xiP1 , xiP2 (4d) , = i=1 D12 = xiP2 2i min xiP1 , xiP2 (6a) , Again, the difference between the Ruzicka-and BrayCurtis-inspired methods is whether reads/fractions of reads found on multiple host S { ( ) } min xiP1 , xiP2 i , classes are weighted according to the number of host classes that i=1 they occur on. Similar to the Ruzicka index for beta diversity, the where xiP1 , xiP2 and xiH are the number/fraction of reads of microbial Ruzicka-inspired 4H index only counts shared microbial reads once taxon i on the first progenitor, the second progenitor and the hybrid re- regardless of the number of host classes that they occur on. By con- spectively, and S is the total number of microbial taxa in the system. A is trast, like the BrayCurtis index for beta diversity, the BrayCurtis- then the number/fraction of reads shared by both progenitors and the hy- inspired 4H index triple weights reads/fractions of reads shared by brid, B is the number/fraction of reads shared by one progenitor (but not all three host classes and double weights reads/fractions of reads both) and the hybrid, B1 is the number/fraction of reads shared by the first shared by two of the three host classes. Function FourHbootstrapA progenitor and the hybrid, B2 is the number/fraction of reads shared by in HybridMicrobiomes performs abundance-based bootstraps of the second progenitor and the hybrid, C is the number/fraction of reads the 4H index with input and output as described for the function found only on the hybrid, D is the number/fraction of reads found only on FourHbootstrap (see above). one or both progenitors, D1 is the number/fraction of reads found only on the first progenitor, D2 is the number/fraction of reads found only on the second progenitor and D12 is the number/fraction of reads found only on 2.2.3 | Bootstrap analysis both progenitors. For the Ruzicka-inspired method, we define the four dimensions of the 4H index (three independent dimensions) as: Function FourHcentroid takes the output from FourHbootstrap or FourHbootstrapA and calculates the centroid of the boot- = B , A+B+C+D (5a) FourHbootstrap or FourHbootstrapA on multiple hybrid systems and uses a PERMANOVA test (Anderson, 2014) on the isometric log- B1 1 = , A+B+C+D ratio transformed (Egozcue et al., 2003) 4H indices (with the option to use a centred log-ratio transformation, an additive log-ratio trans- B2 2 = , A+B+C+D = A , A+B+C+D strapped samples. Function FourHcompare takes the outputs from formation or untransformed data instead (Filzmoser et al., 2010; Quinn et al., 2019)) to determine whether different hybrid systems vary with respect to the importance of the Union, Intersection, Gain (5b) and Loss models, respectively. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 518 2.2.4 | Quaternary plots 2.2.5 | Null planes 519 To visualize the 4H index (see Figures 2 and 3), which can be par- As suggested above (see The Loss Model), both our conceptual ticularly helpful for comparison between systems, we introduce models and the four dimensions of the 4H index conflate microbial a quaternary plotting technique (i.e. a four-d imensional barycen- taxon loss due to the intersection of progenitor microbiomes (i.e. tric plot or an Aitchison Simplex (Aitchison, 1982)). This positions loss of microbial taxa only present on one progenitor) with broader each of our four index dimensions (, , and ) at a vertex microbial taxon loss (i.e. including loss of microbial taxa present on of a triangular prism, with one edge of the prism connecting the both progenitors). Thus, the 4H index does not indicate whether the Gain and Loss models (henceforth termed the transgressive axis) microbial taxa that are lost versus retained by hybrid organisms rep- and the opposite edge connecting the Union and Intersection resent microbial taxa that are shared by both progenitors or taxa that models (henceforth termed the parental axis). Function are only found on one progenitor. Unfortunately, conflation of these FourHquaternary takes the output from FourHbootstrap or different types of loss is necessary to double-counting microbial FourHbootstrapA and generates an interactive and rotatable taxa ( + + + = 1) while still using a maximum of four (ben- quaternary plot of the bootstrapped samples with the option to eficial for visualization) dimensions. To offset this constraint, and include the centroid. Function FourHquaternarycentroid takes better identify the particular microbial taxa that are lost by hybrid the output from FourHbootstrap or FourHbootstrapA and gener- organisms, we develop null planes. Specifically, we assume a null ates an interactive quaternary plot of only the centroids over the model wherein all microbial taxa (or reads in the case of abundance- bootstrapped samples. based methods) present on progenitors are equally likely to be lost F I G U R E 2 Quaternary plots showing 500 bootstrapped genus-level microbial samples (small circles) and the bootstrap centroid (large circles) of the Jaccard-inspired 4H index for (a) gut microbiomes from hybrid Kikihia cicadas (black), Neotoma woodrats (brown) and Aspidoscelis neomexicanus whiptail lizards (green); (b, c) woodrat and lizard systems individually along with the system null planes; (d) leaf (green) and rhizosphere (brown) bacterial/archaeal (16S rRNA, light) and fungal (ITS, dark) microbiomes from B73 line Mo17 line maize hybrids; (e) B73 line Mo17 line maize hybrid leaf and rhizosphere bacterial/archaeal systems along with system null planes; (f) leaf bacterial/archaeal microbiomes from B73 line Mo17 line (red), B73 line CML103 line (green) and B73 line Mo18W line (blue) maize hybrids. For systems in (ac), bootstraps consisted of seven hybrid individuals and seven of each progenitor. For systems in (df), bootstraps consisted of 10 hybrid individuals and 10 of each progenitor. A microbial genus was defined as being part of the core microbiome if at least 50% of hosts from a particular class carried that microbial genus. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | CAMPER et al. F I G U R E 3 Quaternary plots showing 500 bootstrapped genus-level microbial samples (small circles) and the bootstrap centroid (large circles) of the BrayCurtis-inspired 4H index for (a) gut microbiomes from hybrid Kikihia cicadas (black), Neotoma woodrats (brown) and Aspidoscelis neomexicanus whiptail lizards (green); (b, c) woodrat and lizard systems individually along with the system null planes; (d) leaf (green) and rhizosphere (brown) bacterial/archaeal (16S rRNA, light) and fungal (ITS, dark) microbiomes from B73 line Mo17 line maize hybrids; (e) B73 line Mo17 line maize hybrid leaf and rhizosphere bacterial/archaeal systems along with system null planes; (f) leaf bacterial/archaeal microbiomes from B73 line Mo17 line (red), B73 line CML103 line (green) and B73 line Mo18W line (blue) maize hybrids. For systems in (ac), bootstraps consisted of seven hybrid individuals and seven of each progenitor. For systems in (df), bootstraps consisted of 10 hybrid individuals and 10 of each progenitor. A microbial genus was defined as being part of the core microbiome if at least 50% of hosts from a particular class carried that microbial genus. by hybrids. This allows us to define a plane bisecting the quaternary concentrated among microbial taxa only found on one of the two plot at the expected fraction of hybrid microbial taxa that should be progenitors as in the Intersection Model). 4H indices that lie more shared with one versus both progenitors, assuming that there is no towards the vertex relative to the null plane suggest that hybrids preferential loss of one over the other. For any value of = + are disproportionately likely to retain microbes only found on one (i.e. the summed fractions of microbial taxa following the Gain and of the two progenitors (i.e. loss is concentrated among microbial Loss models), the null plane is given by: taxa found on both progenitors and is saltational). By using the null where = = a + d12 a + d12 + d1 + d2 2(a + d12 ) 2(a + d12 ) + d1 + d2 null = (1 )(1 ), (7a) null = (1 ), (7b) for the Jaccard-inspired for the Sorensen-inspired 4H index, = for the Ruzicka-inspired 4H index and = 4H plane as a reference, it is possible to assess the degree to which the loss occurs due to the intersection of progenitor microbiomes versus the broader saltational loss of microbes present on both progenitors. Function FourHnullplane takes the output from index, FourHbootstrap or FourHbootstrapA and graphs the (average) A + D12 A + D12 + D1 + D2 null plane for a particular hybrid system onto a quaternary plot. for the Function FourHplaneD takes the output from FourHbootstrap or BrayCurtis-inspired 4H index. In general, is the fraction of pa- FourHbootstrapA and reports both the average distance between 2(A + D12 ) 2(A + D12 ) + D1 + D2 rental microbial taxa that are found on both progenitors. null and the expected, null, and observed, , value of the intersection di- null are thus the expected fractions of microbial taxa that should mension, as well as the fraction, p, of bootstrap samples that lie be found on only one parent versus both parents under null model further from the vertex than expected (this is useful for testing assumptions. 4H indices that lie more towards the vertex rela- the hypothesis that microbes shared by both progenitors are more tive to the null plane indicate that hybrids are disproportionately likely to be retained by the hybrid than microbes only found on one likely to retain microbes shared by both progenitors (i.e. loss is progenitor). 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 520 3 | R E S U LT S 521 across all systems. Hybrid woodrats, for example, are dominated by the Intersection Model ( = 0.4927), while this is less important for To illustrate the usefulness of the 4H index, we apply functions hybrid cicadas ( = 0.1337) and hybrid lizards ( = 0.1160). Instead, from our HybridMicrobiomes package to a range of plant and animal lizards and cicadas feature a mix of the Gain and Loss models, with hybrid microbiota data sets including results from both 16S rRNA Gain being more important for lizards ( = 0.3438) and Loss being gene and ITS sequencing (see Supplemental Information S2 for ad- more important for cicadas ( = 0.3758). Figure 2b,c illustrates null ditional system pre-analysis steps and Supplemental Information S3 planes for the woodrat and lizard systems. From the null planes and for more information on the organismal systems). These analyses Table 2, we see that the two progenitor woodrat species share a demonstrate how any given hybrid system shows mixed support larger fraction of their core microbes as compared to the two pro- for each of our conceptual models and how the degree of support genitor lizard species. Furthermore, we see that hybrid woodrats are for any particular conceptual model varies from one system to an- biased towards the Intersection Model as compared to the null plane other. First, we consider applying the Jaccard-inspired 4H index ( null = 0.1358, p = 0.002). This means that hybrid woodrats are (incidence-based) to our data sets (see Figure 2; Tables 1 and 2; more likely to retain microbial taxa that are shared by both progeni- Supplemental Video files Figures 2a.mp42f.mp4). Figure 2a shows tors than they are to retain microbial taxa found on only one of the quaternary plots (R version 4.2.1, phyloseq 1.41.1) comparing gut two progenitors. By contrast, hybrid lizards do not show any obvious microbiota from F1 crosses of Neotoma woodrats (brown) (Nielsen bias, and thus, are equally likely to retain (or lose) microbial taxa that et al., 2023), Kikihia cicadas with evidence of mitochondrial intro- are shared by both progenitors or only present on one of the two gression (black) (Haji et al., 2022) and a parthenogenetic Aspidoscelis progenitors. lizard of hybrid origin (green, our own data). Table 1 shows the av- Figure 2d shows quaternary plots of both the phyllosphere erage values of the 4H indices for each system. Table 2 shows the (green) and rhizosphere (brown) of hybrid maize (B73 line Mo17 average values of (the fraction of the overall parental microbiome line) for both bacterial/archaeal (16S rRNA gene, light shade) and found on both progenitors), the mean distance between the pre- fungal (ITS1 gene, dark shade) microbiotas (Wagner et al., 2020). dicted and observed value of and the proportion of bootstrap Figure 2e shows the same bacterial/archaeal microbiotas but in- samples that lie further from the vertex than expected based on cludes their respective null planes, and Figure 2f compares the the null model. Despite the variation in life history (vertebrate vs. in- bacterial/archaeal phyllosphere microbiotas across three different vertebrate, ectotherm vs. endotherm, herbivore vs. insectivore) and maize hybrids: B73 line Mo17 line (red; stiff stalk crossed with mode of hybridization (F1 crosses, mitochondrial introgression, hy- non-stiff stalk varieties (Wagner et al., 2020)), B73 line CML103 brid speciation/parthenogenesis), the 4H index enables comparison line (yellow; temperate crossed with tropical varieties (Woodhouse TA B L E 1 Centroid values of the Jaccard-inspired 4H index as calculated by the FourHcentroid function. Parental axis Neotoma woodrat Transgressive axis + + 0.07296 0.4927 0.5657 0.0811 0.3532 0.4343 Aspidoscelis lizard 0.2941 0.1160 0.4101 0.3438 0.2461 0.5899 Kikihia cicada 0.1999 0.1337 0.3336 0.2906 0.3758 0.6664 Maize B73 Mo17 (leaf, bacteria) 0.1272 0.4949 0.6221 0.0575 0.3205 0.3780 Maize B73 Mo17 (rhizosphere, bacteria) 0.1892 0.5000 0.6892 0.1624 0.1484 0.3108 Maize B73 Mo17 (leaf, fungi) 0.1029 0.5238 0.6267 0.0902 0.2832 0.3734 Maize B73 Mo17 (rhizosphere, fungi) 0.1936 0.4226 0.6162 0.1524 0.2314 0.3838 Maize B73 CML103 (leaf, bacteria) 0.1070 0.5044 0.6114 0.0798 0.3087 0.3885 Maize B73 Mo18W (leaf, bacteria) 0.1601 0.4784 0.6385 0.1632 0.1983 0.3615 Maize B73 CML103 (rhizosphere, fungi) 0.2152 0.4468 0.6620 0.1123 0.2256 0.3379 Maize B73 Mo18W (rhizosphere, fungi) 0.1433 0.4635 0.6068 0.1146 0.2785 0.3931 Note: Summing the values + and + gives totals along the parental axis and the transgressive axis, respectively, and can be used as a broader scale comparison between systems. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | CAMPER et al. TA B L E 2 Fraction of shared microbial taxa among progenitors, , as calculated by the FourHcentroid function for the Jaccard- inspired 4H index. Displacement from the null plane ( null)a and the proportion of bootstrap samples (p) falling further from the vertex than the null plane are both calculated by the FourHnullplaneD function. Next, we consider applying the BrayCurtis-inspired 4H index (abundance-based) to our data sets (see Figure 3; Tables 3 and 4; Supplemental Video files Figures 3a.mp43f.mp4). Similar to the Jaccard-inspired 4H index (see Figure 2; Table 1), hybrid woodrats are dominated by the Intersection Model ( = 0.7745), while this is less important for hybrid cicadas ( = 0.1645) and hybrid lizards null p ( = 0.07921). Instead, lizards and cicadas feature a mix of the Gain Neotoma woodrat 0.6273 0.1358 0.002 and Loss models. For the BrayCurtis-inspired index, however, the Aspidoscelis lizard 0.2490 0.0110 0.364 Gain Model is almost equivalent for lizards ( = 0.1411) and cicadas Kikihia cicada 0.2827 0.0373 0.178 ( = 0.1420). Meanwhile, the Loss Model is slightly more import- Maize B73 Mo17 (leaf, bacteria) 0.5984 0.1208 0 ant for lizards ( = 0.5705) as compared to cicadas ( = 0.4993), Maize B73 Mo17 (rhizosphere, bacteria) 0.6291 0.0663 0 Maize B73 Mo17 (leaf, fungi) 0.6483 0.1167 0 Maize B73 Mo17 (rhizosphere, fungi) 0.5664 0.0742 0 Maize B73 CML103 (leaf, bacteria) 0.6475 0.1077 0 Maize B73 Mo18W (leaf, bacteria) 0.6121 0.0872 0 Maize B73 CML103 (rhizosphere, fungi) 0.5565 0.0781 0 Maize B73 Mo18W (rhizosphere, fungi) 0.6145 0.0910 0 which is the reverse of findings for the Jaccard-inspired 4H index. Subtle differences in results based on the chosen metric are consistent with the different interpretations of the metrics. In this case, a Positive values of null indicate that points lie closer to the intersection vertex than expected by chance, suggesting that hybrids are more likely to retain taxa shared by both progenitors than they are to retain taxa shared by only one of the two progenitors. Negative values indicate the opposite, namely that hybrids are more likely to retain taxa only found on one of the two progenitors than they are to retain taxa found on both progenitors. for instance, community membership differences suggest that lizard hybrids feature more Gain and less Loss, but that these differences are insignificant or reversed when considering abundance changes. Such discrepancies are expected when membership changes occur primarily in rare taxa and thus contribute little to abundance change, which may instead be dominated by shifts in abundance of microbes shared by hybrids and progenitors. From the null planes and Table 4, we see that, consistent with the Jaccard-inspired 4H index, the Bray Curtis-inspired 4H index suggests that the two progenitor woodrat species share a larger fraction of their core microbes as compared to the two progenitor lizard species. Furthermore, hybrid woodrats are biased towards retaining microbes shared by both progenitors ( null = 0.0476518, p = 0.006), whereas hybrid lizards do not preferentially retain microbes based on whether they are shared by one or both progenitors ( null = 0.030863, p = 0.764). Like our animal examples, abundance- and incidence-based 4H indices for maize hybrids exhibit a similar pattern. In particular, with our abundance-based analysis, we again find that the entire maize system is dominated by the Intersection Model. Indeed, like et al., 2021)) and B73 line Mo18W line (blue; flooding sensitive woodrats, the dominance of the Intersection Model is even more crossed with flooding insensitive varieties (Campbell et al., 2015)). apparent for the BrayCurtis-inspired 4H index than it is for the Tables 1 and 2 show the corresponding values of the 4H indices, as Jaccard-inspired 4H index with >60% of model support across all well as relationships of the hybrid microbiotas to their respective comparisons (see Table 3). However, differences between the hybrid null planes. As with our animal examples, our analysis of maize hy- rhizosphere (brown) and hybrid phyllosphere (green) are not as obvi- brids demonstrates the versatility of the 4H index and how the 4H ous and/or are reversed when changes in abundance are accounted index can be used to compare not only between microbiotas from for. Again, this suggests that microbiota membership changes on different host species but also between microbiotas from different the hybrid are sometimes but not always consistent with abundance parts of a single organism (roots vs. leaf) or different microbial tax- changes. onomic groups (bacteria vs. fungi). Notably, the entire maize sys- One of the benefits of the 4H index is the fact that it can be ap- tem is dominated by the Intersection Model as seen by the nearly plied to any hybrid system, regardless of the type of host, the type of 50% or more of model support across all comparisons (see Table 1). microbiome, microbiome composition or even microbiome diversity. However, the hybrid rhizosphere (brown) is more prone to Union and This flexibility follows from the fact that the 4H index is monotonic Gain. This is apparent from its relatively greater clustering nearest to with respect to each vertex/dimension (, , and ), option- the Union and Gain vertices, as well as relatively greater support for ally density invariant and replication invariant (see Supplemental these two models (see Figure 2d; Table 1; Supplemental Video files Information S1) (Magurran & McGill, 2010). Despite this, some Figure 2d.mp4). By contrast, the hybrid phyllosphere is more prone standardization of data sets from different systems is necessary to Loss. This is apparent from its relatively greater clustering near for fair comparison. For example, the 4H index can be applied to the Loss vertex and relatively greater support for the Loss Model microbiomes at any taxonomic scale. As expected, however, higher (see Figure 2d; Table 1; Supplemental Video files Figure 2d.mp4). taxonomic scales predict a greater importance of the Intersection 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 522 TA B L E 3 Centroid values of the Bray Curtis-inspired 4H index as calculated by the FourHcentroid function. Parental axis 523 Transgressive axis + + Neotoma woodrat 0.1011 0.7745 0.8756 0.02852 0.09587 0.12439 Aspidoscelis lizard 0.2092 0.07921 0.28841 0.1411 0.5705 0.7116 Kikihia cicada 0.1942 0.1645 0.3587 0.1420 0.4993 0.6413 Maize B73 Mo17 (leaf, bacteria) 0.2326 0.5790 0.8116 0.02496 0.1635 0.1884 Maize B73 Mo17 (rhizosphere, bacteria) 0.1199 0.7311 0.8510 0.02944 0.1196 0.1490 Maize B73 Mo17 (leaf, fungi) 0.08865 0.7709 0.8596 0.03248 0.1079 0.1404 Maize B73 Mo17 (rhizosphere, fungi) 0.1157 0.6092 0.7248 0.06660 0.2086 0.2752 Maize B73 CML103 (leaf, bacteria) 0.1094 0.7372 0.8467 0.03673 0.1166 0.1533 Maize B73 Mo18W (leaf, bacteria) 0.1611 0.6719 0.8330 0.03272 0.1343 0.1670 Maize B73 CML103 (rhizosphere, fungi) 0.1542 0.6257 0.7800 0.05275 0.1673 0.2200 Maize B73 Mo18W (rhizosphere, fungi) 0.1357 0.6090 0.7447 0.06074 0.1945 0.2553 Note: Summing the values + and + gives totals along the parental axis and the transgressive axis, respectively, and can be used as a broader scale comparison between systems. Model because hybrids are more likely to share distantly related FourHbootstrap and FourHbootstrapA do have the option to rarefy microbial taxa with progenitors than they are to share identical or microbiome samples to a lower read depth than the minimal number near-identical microbial taxa (see Figures S1.1S1.4; Tables S1.1 and of reads of the lowest sample. This allows for standardization of read S1.2). Importantly, because taxonomic scale can have considerable depth across systems. effects, systems should always be compared using the same taxonomic scale, and interpretation of the index should always be within the context of the taxonomic scale chosen. Likewise, defining the 4 | DISCUSSION core microbiome based on a lower fraction of hosts also favours Intersection (at least some hybrids and some of each parental spe- The advent of low-cost sequencing has greatly contributed to our cies are likely to have a particular microbial taxon, even if it is just a understanding of the importance of both hybridization and HA mi- transient acquisition from the environment; see Figures S2.1S2.4; crobiomes on host ecological traits and evolutionary consequences. Tables S2.1 and S2.2). Again, then, it is important to use the same These two fields come together in the study of HA microbiomes of value of for all systems that are being compared. Host sample size hybrid organismsa newly emerging area of research across disci- has a smaller, but still detectable, effect resulting in somewhat dif- plines ranging from agricultural science to ecology and conserva- ferent trends across systems but generally shifting the 4H index to- tion. In this paper, we integrate four conceptual models to develop wards the parental axis and away from the transgressive axis (see a framework for understanding the relationship between hybrid Figures S3.1S3.4; Tables S3.1 and S3.2). Although the effect of microbiomes and the microbiomes of their progenitors. We then host sample size is relatively small, particularly for larger host sam- use these models to develop a four-dimensional (three independ- ple sizes, it is still best to compare systems by subsampling to the ent dimensions) metricthe 4H indexto describe where a par- smallest number of hosts available for any host class across all sys- ticular hybrid complex falls among our four models. Our 4H index tems (e.g. see Figure 2 where we were limited to 7 individuals based borrows inspiration from beta diversity metrics, and thus takes on on the number of available cicada microbiotas). Finally, sequencing four different forms; two are incidence-based (Jaccard-inspired and depth has almost no impact on predictions, at least for >1000 reads Sorensen-inspired), and two are abundance-based (BrayCurtis- or more. This last feature of the 4H index is a benefit of focusing inspired and Ruzicka-inspired). Importantly, the 4H index facilitates on core microbiomes since low abundance microbial taxa that are comparisons across widely disparate systems, ultimately making it likely to be missed at low read depths are unlikely to be part of the possible to identify patterns that emerge across hybrid microbiomes core of any given species. For this reason, it is largely unnecessary to from different organisms. For example, the 4H index could be used standardize for read depth across systems. Nevertheless, functions to determine whether there are systematic differences between 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | CAMPER et al. TA B L E 4 Fraction of shared microbial taxa among progenitors, , as calculated by the FourHcentroid function for the BrayCurtis-inspired 4H index. Displacement from the null plane ( null)a and the proportion of bootstrap samples (p) falling further from the vertex than the null plane are both calculated by the FourHnullplaneD function. null 0.0476518 For this reason, we envision the 4H index as a tool that can be used for exploring and comparing patterns and formulating hypotheses about underlying eco-evolutionary processes of microbiome restructuring after host hybridization. While not explicitly explored in this manuscript, the 4H index could easily be extended to examine HA microbiome functional p change between hybrids and their progenitors. Indeed, when applied 0.006 to function, the 4H index could be useful for forming hypotheses re- 0.764 lated to hybrid ecology and/or ecological success. However, even in Neotoma woodrat 0.830025 Aspidoscelis lizard 0.392932 Kikihia cicada 0.3025042 0.05380129 0.19 this context the 4H index should be interpreted as a tool for charac- Maize B73 Mo17 (leaf, bacteria) 0.6684608 0.03651511 0.132 terizing patterns of change, rather than mechanisms. This is because Maize B73 Mo17 (rhizosphere, bacteria) 0.7801631 0.06678432 0 Maize B73 Mo17 (leaf, fungi) 0.8294745 0.05757388 0 Maize B73 Mo17 (rhizosphere, fungi) 0.7474605 0.06723603 0 Maize B73 CML103 (leaf, bacteria) 0.8038933 0.05631366 0.006 Maize B73 Mo18W (leaf, bacteria) 0.7243933 0.06803136 0.016 Maize B73 CML103 (rhizosphere, fungi) 0.7003835 0.07931675 0 Maize B73 Mo18W (rhizosphere, fungi) 0.7082141 0.0815329 0 0.030863 microbiome function and host ecology can have bidirectional impacts, and thus, it can be challenging to delineate cause and effect. As a result, though function may provide better insight into potential pattern a Positive values of null indicate that points lie closer to the intersection vertex than expected by chance, suggesting that hybrids are more likely to retain taxa shared by both progenitors than they are to retain taxa shared by only one of the two progenitors. Negative values indicate the opposite, namely that hybrids are more likely to retain taxa only found on one of the two progenitors than they are to retain taxa found on both progenitors. generating mechanisms, the 4H index is not a test for causality, but rather an exploratory tool for hypothesis generation. Outside the context of hybridization, it is worth noting that this same framework can be applied to any triplet of host species, where one of the three host species is in some way intermediate to the other two. Thus, for example, a 4H index could be calculated for the microbiomes of organisms from an ecotonal habitat, and then compared to the microbiomes of organisms from the two pure habitat types on either end of the ecotone (O'Brien et al., 2022), even if it is the same host taxon across the entire zone. Likewise, a 4H index could be calculated for species (e.g. swordtail males, Xiphophorus nigrensis) that exhibit three discrete size classes, with one size class being intermediate to the other two (Morris et al., 1992). Similarly, a 4H index could be calculated for captive animals fed two different pure diets as compared to captive animals fed a mixed diet. In these scenarios, the interpretation of our four conceptual models would change. However, because the 4H metric is defined solely based on distributions of microbial presence/absence or abundance across non-overlapping sets hybrid plant versus hybrid animal microbiomes, or between hybrid of host classes, it is valid for any analysis where there is ecological, vertebrate versus hybrid invertebrate microbiomes. Likewise, the evolutionary, morphological or physiological reason to believe that 4H index could be used to determine how phylogenetic and/or phe- one host class falls between the other two host classes. notypic distances between progenitors or ploidy level impact the hybrid microbiome. AU T H O R C O N T R I B U T I O N S Importantly, the intent of each of the four conceptual models Benjamin T. Camper and Sharon Bewick wrote the first draft of the and, indeed, the 4H index in general is to highlight hybridprogeni- manuscript. Sharon Bewick developed the code for the R package. All tor microbiome relationships. Thus, like beta diversity, the 4H index authors contributed substantially to the content of the manuscript. should be taken as a measure of pattern, not process. Just as beta diversity cannot be used to explain why turnover differs among com- AC K N OW L E D G E M E N T S munities, the 4H index should not be used to discriminate among We thank Andrew Kanes, Thomas Dempster, August Spencer microbiome reassembly mechanisms responsible for microbiome and Eva Purcell for their field assistance in New Mexico as well restructuring after host hybridization. In the woodrat system, for as Eva Purcell, Lily Margeson, Simon Dunn, Georgianna Bellinger, example, the 4H index cannot be used to explain why Intersection Henry Egloff, Kaila Hodges, Camryn Lachica and Savannah Utz dominates. It may be that the hybrid woodrat immune system is re- for their assistance assembling drift fence trapping arrays. We fractory to all microbes not found on both progenitors. Alternatively, thank Daniel Nielsen for generously providing additional woodrat it could be that hybrid woodrats are restricted to habitats where data and insight into the woodrat system. All research was ap- both progenitors overlap and the hybrid microbiome reflects micro- proved by Clemson University under IACUC protocol numbers bial exposure patterns of hybrid animals. Regardless, experimental #2020-015 and #2021-0 47. We completed this work under the work will always be needed to understand what drives hybridpro- Sevilleta National Wildlife Refuge Special Use Permit #SEV_ genitor HA microbiome relationships observed using the 4H index. Bewick_Camper_2022_59, the USDA-A RS Jornada study permit 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 524 #592 and the New Mexico Department of Game and Fish permit authorization #3772. This study was funded by the NSF award #2105604, a Clemson University Support for Early Exploration and Development (CUSEED) Grant and the Clemson University Creative Inquiry (CI) Program. C O N FL I C T O F I N T E R E S T S TAT E M E N T No conflicts of interest have been declared. PEER REVIEW The peer review history for this article is available at https://w ww. webofs cience.com/a pi/g ateway/wos/p eer-review/10.1111/2041- 210X.14279. DATA AVA I L A B I L I T Y S TAT E M E N T Cleaned data for cicadas, woodrats, lizards and maize and all scripts to generate the figures and tables for this manuscript are available on GitHub (https://github.com/bewicklab/HybridMicrobiom eFramework) and Zenodo (https://zenodo.org/records/10358091) (Camper et al., 2023). The HybridMicrobiomes R package is available from CRAN (https://cran.r-projec t.org/web/packages/HybridMicr obiomes/index.html). ORCID Benjamin T. Camper Zachary Laughlin Daniel Malagon https://orcid.org/0000-0002-7861-485X https://orcid.org/0009-0003-7931-7130 https://orcid.org/0000-0003-2831-4370 Robert Denton https://orcid.org/0000-0002-8629-1376 Sharon Bewick https://orcid.org/0000-0002-2563-5761 REFERENCES Abbott, K. C., Eppinga, M. B., Umbanhowar, J., Baudena, M., & Bever, J. D. (2021). Microbiome influence on host community dynamics: Conceptual integration of microbiome feedback with classical hostmicrobe theory. Ecology Letters, 24, 27962811. Abbott, R., Albach, D., Ansell, S., Arntzen, J. W., Baird, S. J., Bierne, N., Boughman, J., Brelsford, A., Buerkle, C. A., & Buggs, R. (2013). Hybridization and speciation. Journal of Evolutionary Biology, 26, 229246. Ahn, J., Sinha, R., Pei, Z., Dominianni, C., Wu, J., Shi, J., Goedert, J. J., Hayes, R. B., & Yang, L. (2013). Human gut microbiome and risk for colorectal cancer. Journal of the National Cancer Institute, 105, 19071911. Aitchison, J. (1982). The statistical analysis of compositional data. Journal of the Royal Statistical Society: Series B (Methodological), 44, 139160. Anderson, M. J. (2014). Permutational multivariate analysis of variance (PERMANOVA). Wiley Statsref: Statistics Reference Online, 115. https://doi.org/10.1002/9781118445112.stat07841 Apprill, A., Miller, C. A., Van Cise, A. M., U'Ren, J. M., Leslie, M. S., Weber, L., Baird, R. W., Robbins, J., Landry, S., & Bogomolni, A. (2020). Marine mammal skin microbiotas are influenced by host phylogeny. Royal Society Open Science, 7, 192046. Archie, E. A., & Theis, K. R. (2011). Animal behaviour meets microbial ecology. Animal Behaviour, 82, 425436. Baeckens, S. (2019). Evolution of animal chemical communication: Insights from non-model species and phylogenetic comparative methods. Belgian Journal of Zoology, 149, 6393. 525 Baedke, J., Fbregas-Tejeda, A., & Nieves Delgado, A. (2020). The holobiont concept before Margulis. Journal of Experimental Zoology Part B: Molecular and Developmental Evolution, 334, 149155. Bahrndorff, S., Alemu, T., Alemneh, T., & Lund Nielsen, J. (2016). The microbiome of animals: Implications for conservation biology. International Journal of Genomics, 2016, 17. Baird, S. J., Ribas, A., Macholn, M., Albrecht, T., Pilek, J., & Goy de Bellocq, J. (2012). Where are the wormy mice? A reexamination of hybrid parasitism in the European house mouse hybrid zone. Evolution: International Journal of Organic Evolution, 66, 27572772. Balard, A., & Heitlinger, E. (2022). Shifting focus from resistance to disease tolerance: A review on hybrid house mice. Ecology and Evolution, 12, e8889. Banerjee, A., Cornejo, J., & Bandopadhyay, R. (2020). Emergent climate change impact throughout the world: Call for microbiome conservation before it's too late. Biodiversity and Conservation, 29, 345348. Barton, N. H., & Hewitt, G. M. (1985). Analysis of hybrid zones. Annual Review of Ecology and Systematics, 16, 113148. Bates, K. A., Clare, F. C., O'Hanlon, S., Bosch, J., Brookes, L., Hopkins, K., McLaughlin, E. J., Daniel, O., Garner, T. W., & Fisher, M. C. (2018). Amphibian chytridiomycosis outbreak dynamics are linked with host skin bacterial community structure. Nature Communications, 9, 693. Bates, K. A., Sommer, U., Hopkins, K. P., Shelton, J. M., Wierzbicki, C., Sergeant, C., Tapley, B., Michaels, C. J., Schmeller, D. S., & Loyau, A. (2022). Microbiome function predicts amphibian chytridiomycosis disease dynamics. Microbiome, 10, 116. Bateson, P. (2002). William Bateson: A biologist ahead of his time. Journal of Genetics, 81, 4958. Bateson, P. (1984). Behavioral evolution and integrative levels: The Tc Schneirla conferences series, Volume 1. Psychology Press. Belcaid, M., Casaburi, G., McAnulty, S. J., Schmidbaur, H., Suria, A. M., Moriano-Gutierrez, S., Pankey, M. S., Oakley, T. H., Kremer, N., & Koch, E. J. (2019). Symbiotic organs shaped by distinct modes of genome evolution in cephalopods. Proceedings of the National Academy of Sciences of the United States of America, 116, 30303035. Bergot, A.-S., Giri, R., & Thomas, R. (2019). The microbiome and rheumatoid arthritis. Best Practice & Research. Clinical Rheumatology, 33, 101497. Blyton, M. D., Soo, R. M., Whisson, D., Marsh, K. J., Pascoe, J., Le Pla, M., Foley, W., Hugenholtz, P., & Moore, B. D. (2019). Faecal inoculations alter the gastrointestinal microbiome and allow dietary expansion in a wild specialist herbivore, the koala. Animal Microbiome, 1, 118. Bordenstein, S. R., & Theis, K. R. (2015). Host biology in light of the microbiome: Ten principles of holobionts and hologenomes. PLoS Biology, 13, e1002226. Bosch, T. C., & Miller, D. J. (2016). The holobiont imperative. Springer. Bravo, J. A., Forsythe, P., Chew, M. V., Escaravage, E., Savignac, H. M., Dinan, T. G., Bienenstock, J., & Cryan, J. F. (2011). Ingestion of lactobacillus strain regulates emotional behavior and central GABA receptor expression in a mouse via the vagus nerve. Proceedings of the National Academy of Sciences of the United States of America, 108, 1605016055. Bray, J. R., & Curtis, J. T. (1957). An ordination of the upland forest communities of southern Wisconsin. Ecological Monographs, 27, 326349. Bredon, M., Herran, B., Bertaux, J., Grve, P., Moumen, B., & Bouchon, D. (2020). Isopod holobionts as promising models for lignocellulose degradation. Biotechnology for Biofuels, 13, 114. Britton-Davidian, J., Fel-Clair, F., Lopez, J., Alibert, P., & Boursot, P. (2005). Postzygotic isolation between the two European subspecies of the house mouse: Estimates from fertility patterns in wild and laboratory-bred hybrids. Biological Journal of the Linnean Society, 84, 379393. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. | Brotman, R. M., Klebanoff, M. A., Nansel, T. R., Yu, K. F., Andrews, W. W., Zhang, J., & Schwebke, J. R. (2010). Bacterial vaginosis assessed by gram stain and diminished colonization resistance to incident gonococcal, chlamydial, and trichomonal genital infection. Journal of Infectious Diseases, 202, 19071915. Brown, H., & Bouton, J. H. (1993). Physiology and genetics of interspecific hybrids between photosynthetic types. Annual Review of Plant Biology, 44, 435456. Brucker, R. M., & Bordenstein, S. R. (2012). Speciation by symbiosis. Trends in Ecology & Evolution, 27, 443451. Brucker, R. M., & Bordenstein, S. R. (2013). The hologenomic basis of speciation: Gut bacteria cause hybrid lethality in the genus Nasonia. Science, 341, 667669. Buffie, C. G., & Pamer, E. G. (2013). Microbiota-mediated colonization resistance against intestinal pathogens. Nature Reviews Immunology, 13, 790801. Campbell, L. G., Snow, A. A., & Ridley, C. E. (2006). Weed evolution after crop gene introgression: Greater survival and fecundity of hybrids in a new environment. Ecology Letters, 9, 11981209. Campbell, M. T., Proctor, C. A., Dou, Y., Schmitz, A. J., Phansak, P., Kruger, G. R., Zhang, C., & Walia, H. (2015). Genetic and molecular characterization of submergence response identifies Subtol6 as a major submergence tolerance locus in maize. PLoS ONE, 10, e0120385. Camper, B., Laughlin, Z., Malagon, D., Denton, R., & Bewick, S. (2023). Zenodo HybridMicrobiomesFrameworkv1.1. https://zenodo.org/ records/10358091 Capblancq, T., Desprs, L., & Mavrez, J. (2020). Genetic, morphological and ecological variation across a sharp hybrid zone between two alpine butterfly species. Evolutionary Applications, 13, 14351450. Carreira, V. P., Soto, I. M., Fanara, J. J., & Hasson, E. (2008). A study of wing morphology and fluctuating asymmetry in interspecific hybrids between Drosophila buzzatii and D. koepferae. Genetica, 133, 111. Carthey, A. J., Blumstein, D. T., Gallagher, R. V., Tetu, S. G., & Gillings, M. R. (2020). Conserving the holobiont. Functional Ecology, 34, 764776. Chen, Y. E., Fischbach, M. A., & Belkaid, Y. (2018). Skin microbiotahost interactions. Nature, 553, 427436. Chiarello, M., Auguet, J.-C ., Bettarel, Y., Bouvier, C., Claverie, T., Graham, N. A., Rieuvilleneuve, F., Sucr, E., Bouvier, T., & Villger, S. (2018). Skin microbiome of coral reef fish is highly variable and driven by host phylogeny and diet. Microbiome, 6, 114. Chong, P. P., Chin, V. K., Looi, C. Y., Wong, W. F., Madhavan, P., & Yong, V. C. (2019). The microbiome and irritable bowel syndromeA review on the pathophysiology, current research and future therapy. Frontiers in Microbiology, 10, 1136. Claus, S. P., Tsang, T. M., Wang, Y., Cloarec, O., Skordi, E., Martin, F. P., Rezzi, S., Ross, A., Kochhar, S., & Holmes, E. (2008). Systemic multicompartmental effects of the gut microbiome on mouse metabolic phenotypes. Molecular Systems Biology, 4, 219. Cooper, R. D., & Shaffer, H. B. (2021). Allele-specific expression and gene regulation help explain transgressive thermal tolerance in non-native hybrids of the endangered California tiger salamander (Ambystoma californiense). Molecular Ecology, 30, 9871004. Corbin, K. D., Krajmalnik-Brown, R., Carnero, E. A., Bock, C., Emerson, R., Rittmann, B. E., Marcus, A. K., Davis, T., Dirks, B., & Ilhan, Z. E. (2020). Integrative and quantitative bioenergetics: Design of a study to assess the impact of the gut microbiome on host energy balance. Contemporary Clinical Trials Communications, 19, 100646. Cregger, M., Veach, A., Yang, Z., Crouch, M., Vilgalys, R., Tuskan, G., & Schadt, C. (2018). The Populus holobiont: Dissecting the effects of plant niches and genotype on the microbiome. Microbiome, 6, 114. Davidson, G. L., Cooke, A. C., Johnson, C. N., & Quinn, J. L. (2018). The gut microbiome as a driver of individual variation in cognition and functional behaviour. Philosophical Transactions of the Royal Society B: Biological Sciences, 373, 20170286. CAMPER et al. Delaux, P.-M., & Schornack, S. (2021). Plant evolution driven by interactions with symbiotic and pathogenic microbes. Science, 371, eaba6605. Deng, H. (2012). A review of diversity-stability relationship of soil microbial community: What do we not know? Journal of Environmental Sciences, 24, 10271035. Dice, L. R. (1945). Measures of the amount of ecologic association between species. Ecology, 26, 297302. Dittrich-Reed, D. R., & Fitzpatrick, B. M. (2013). Transgressive hybrids as hopeful monsters. Evolutionary Biology, 40, 310315. Dobzhansky, T. (1934). Studies on hybrid sterility. Zeitschrift fr Zellforschung und Mikroskopische Anatomie, 21, 169223. Dowling, T. E., & Secor, C. L. (1997). The role of hybridization and introgression in the diversification of animals. Annual Review of Ecology and Systematics, 28, 593619. Egozcue, J. J., Pawlowsky-Glahn, V., Mateu-Figueras, G., & Barcelo-V idal, C. (2003). Isometric logratio transformations for compositional data analysis. Mathematical Geology, 35, 279300. Ehrlich, P. R., & Raven, P. H. (1964). Butterflies and plants: A study in coevolution. Evolution, 18, 586608. Ezenwa, V. O., Gerardo, N. M., Inouye, D. W., Medina, M., & Xavier, J. B. (2012). Animal behavior and the microbiome. Science, 338, 198199. Filzmoser, P., Hron, K., & Reimann, C. (2010). The bivariate statistical analysis of environmental (compositional) data. Science of the Total Environment, 408, 42304238. Fontaine, S. S., & Kohl, K. D. (2020). Optimal integration between host physiology and functions of the gut microbiome. Philosophical Transactions of the Royal Society B: Biological Sciences, 375, 20190594. Forejt, J. (1996). Hybrid sterility in the mouse. Trends in Genetics, 12, 412417. Forejt, J., & Ivnyi, P. (1974). Genetic studies on male sterility of hybrids between laboratory and wild mice (Mus musculus L.). Genetics Research, 24, 189206. Friesen, M. L., Porter, S. S., Stark, S. C., von Wettberg, E. J., Sachs, J. L., & Martinez-Romero, E. (2011). Microbially mediated plant functional traits. Annual Review of Ecology, Evolution, and Systematics, 42, 2346. Fronk, D. C., & Sachs, J. L. (2022). Symbiotic organs: The nexus of host microbe evolution. Trends in Ecology & Evolution, 37, 599610. Gaona, O., Cerqueda-Garca, D., Martnez-Martnez, D., Moya, A., & Falcn, L. I. (2016). Microbiome of the sexual scent organ of Leptonycteris yerbabuenae. Report no. 2167-9843 (PeerJ Preprints). Goldschmidt, R. (1933). Some aspects of evolution. Science, 78, 539547. Goldschmidt, R. (1940). The material basis of evolution. Yale University Press. Good, J. M., Handel, M. A., & Nachman, M. W. (2008). Asymmetry and polymorphism of hybrid male sterility during the early stages of speciation in house mice. Evolution: International Journal of Organic Evolution, 62, 5065. Greene, L. K., Williams, C. V., Junge, R. E., Mahefarisoa, K. L., Rajaonarivelo, T., Rakotondrainibe, H., O'Connell, T. M., & Drea, C. M. (2020). A role for gut microbiota in host niche differentiation. The ISME Journal, 14, 16751687. Grinberg, M., Levin, R., Neuman, H., Ziv, O., Turjeman, S., Gamliel, G., Nosenko, R., & Koren, O. (2022). Antibiotics increase aggression behavior and aggression-related pheromones and receptors in Drosophila melanogaster. Iscience, 25, 104371. Haji, D., Vailionis, J., Stukel, M., Gordon, E., Lemmon, E. M., Lemmon, A. R., & Simon, C. (2022). Lack of host phylogenetic structure in the gut bacterial communities of New Zealand cicadas and their interspecific hybrids. Scientific Reports, 12, 20559. Harris, R. N., James, T. Y., Lauer, A., Simon, M. A., & Patel, A. (2006). Amphibian pathogen Batrachochytrium dendrobatidis is inhibited by the cutaneous bacteria of amphibian species. EcoHealth, 3, 5356. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 526 Harrison, X. A., Price, S. J., Hopkins, K., Leung, W. T., Sergeant, C., & Garner, T. W. (2019). Diversity-stability dynamics of the amphibian skin microbiome and susceptibility to a lethal viral pathogen. Frontiers in Microbiology, 10, 2883. He, J., Li, J., & Zheng, Y. (2013). Thoughts on the microbial diversity- stability relationship in soil ecosystems. Biodiversity Science, 21, 411420. Heard, S. B., & Hauser, D. L. (1995). Key evolutionary innovations and their ecological mechanisms. Historical Biology, 10, 151173. Heys, C., Fisher, A., Dewhurst, A., Lewis, Z., & Liz, A. (2021). Exposure to foreign gut microbiota can facilitate rapid dietary shifts. Scientific Reports, 11, 16791. Hovick, S. M., & Whitney, K. D. (2014). Hybridisation is associated with increased fecundity and size in invasive taxa: Meta-analytic support for the hybridisation-invasion hypothesis. Ecology Letters, 17, 14641477. Hu, Y., Sanders, J. G., ukasik, P., D'Amelio, C. L., Millar, J. S., Vann, D. R., Lan, Y., Newton, J. A., Schotanus, M., & Kronauer, D. J. (2018). Herbivorous turtle ants obtain essential nutrients from a conserved nitrogen-recycling gut microbiome. Nature Communications, 9, 964. Ives, A. R., & Carpenter, S. R. (2007). Stability and diversity of ecosystems. Science, 317, 5862. Jaccard, P. (1908). Nouvelles recherches sur la distribution florale. Bulletin de la Societe Vaudoise des Sciences Naturelles, 44, 223270. Jackson, J. F. (1973). The phenetics and ecology of a narrow hybrid zone. Evolution, 27, 5868. Janz, N. (2011). Ehrlich and Raven revisited: Mechanisms underlying codiversification of plants and enemies. Annual Review of Ecology, Evolution, and Systematics, 42, 7189. Janzen, D. H. (1980). When is it coevolution? Evolution, 34(3), 611612. https://doi.org/10.1111/j.1558-5646.1980.tb04849.x Jia, Y., Jin, S., Hu, K., Geng, L., Han, C., Kang, R., Pang, Y., Ling, E., Tan, E. K., & Pan, Y. (2021). Gut microbiome modulates drosophila aggression through octopamine signaling. Nature Communications, 12, 2698. Jimnez, R. R., & Sommer, S. (2017). The amphibian microbiome: Natural range of variation, pathogenic dysbiosis, and role in conservation. Biodiversity and Conservation, 26, 763786. Jin Song, S., Woodhams, D. C., Martino, C., Allaband, C., Mu, A., Javorschi-Miller-Montgomery, S., Suchodolski, J. S., & Knight, R. (2019). Engineering the microbiome for animal health and conservation. Experimental Biology and Medicine, 244, 494504. Jing, T.-Z., Qi, F.-H., & Wang, Z.-Y. (2020). Most dominant roles of insect gut bacteria: Digestion, detoxification, or essential nutrient provision? Microbiome, 8, 120. Jost, L., Choa, A., & Chazdon, R. L. (2011). Compositional similarity and (beta) diversity. In A. E. Magurran & B. J. Mc Gill (Eds.), Biological diversity: Frontiers in measurement and assessment (pp. 6684). Oxford University Press. Ju, J.-F., Bing, X.-L ., Zhao, D.-S., Guo, Y., Xi, Z., Hoffmann, A. A., Zhang, K.-J., Huang, H.-J., Gong, J.-T., & Zhang, X. (2020). Wolbachia supplement biotin and riboflavin to enhance reproduction in planthoppers. The ISME Journal, 14, 676687. Kamada, N., Chen, G. Y., Inohara, N., & Nez, G. (2013). Control of pathogens and pathobionts by the gut microbiota. Nature Immunology, 14, 685690. Kearney, M. (2005). Hybridization, glaciation and geographical parthenogenesis. Trends in Ecology & Evolution, 20, 495502. Kirchoff, N. S., Udell, M. A., & Sharpton, T. J. (2019). The gut microbiome correlates with conspecific aggression in a small population of rescued dogs (Canis familiaris). PeerJ, 7, e6103. Kohl, K. D., Weiss, R. B., Cox, J., Dale, C., & Denise Dearing, M. (2014). Gut microbes of mammalian herbivores facilitate intake of plant toxins. Ecology Letters, 17, 12381246. Kolodny, O., Callahan, B. J., & Douglas, A. E. (2020). The role of the microbiome in host evolution. Philosophical Transactions of the Royal Society B: Biological Sciences, 375, 20190588. | 527 Kratochwil, C. F., & Meyer, A. (2015). Closing the genotypephenotype gap: Emerging technologies for evolutionary genetics in ecological model vertebrate systems. BioEssays, 37, 213226. Kueneman, J. G., Parfrey, L. W., Woodhams, D. C., Archer, H. M., Knight, R., & McKenzie, V. J. (2014). The amphibian skin-associated microbiome across species, space and life history stages. Molecular Ecology, 23, 12381250. Lafarga-De la Cruz, F., Nez-Acua, G., & Gallardo-Escrate, C. (2013). Hybridization between Haliotis rufescens and Haliotis discus hannai: Evaluation of fertilization, larval development, growth and thermal tolerance. Aquaculture Research, 44, 12061220. Larouche, O., Hodge, J. R., Alencar, L. R., Camper, B., Adams, D. S., Zapfe, K., Friedman, S. T., Wainwright, P. C., & Price, S. A. (2020). Do key innovations unlock diversification? A case-study on the morphological and ecological impact of pharyngognathy in acanthomorph fishes. Current Zoology, 66, 575588. Lee, C.-R ., Wang, B., Mojica, J. P., Mandkov, T., Prasad, K. V., Goicoechea, J. L., Perera, N., Hellsten, U., Hundley, H. N., & Johnson, J. (2017). Young inversion with multiple linked QTLs under selection in a hybrid zone. Nature Ecology & Evolution, 1, 0119. Legendre, P. (2014). Interpreting the replacement and richness difference components of beta diversity. Global Ecology and Biogeography, 23, 13241334. Ley, R. E., Hamady, M., Lozupone, C., Turnbaugh, P. J., Ramey, R. R., Bircher, J. S., Schlegel, M. L., Tucker, T. A., Schrenzel, M. D., & Knight, R. (2008). Evolution of mammals and their gut microbes. Science, 320, 16471651. Li, M., Wang, B., Zhang, M., Rantalainen, M., Wang, S., Zhou, H., Zhang, Y., Shen, J., Pang, X., & Zhang, M. (2008). Symbiotic gut microbes modulate human metabolic phenotypes. Proceedings of the National Academy of Sciences of the United States of America, 105, 21172122. Li, W., Liu, J., Tan, H., Yang, C., Ren, L., Liu, Q., Wang, S., Hu, F., Xiao, J., & Zhao, R. (2018). Genetic effects on the gut microbiota assemblages of hybrid fish from parents with different feeding habits. Frontiers in Microbiology, 9, 2972. Li, Z., Wright, A. D. G., Si, H., Wang, X., Qian, W., Zhang, Z., & Li, G. (2016). Changes in the rumen microbiome and metabolites reveal the effect of host genetics on hybrid crosses. Environmental Microbiology Reports, 8, 10161023. Maebe, K., Vereecken, N. J., Piot, N., Revert, S., Cejas, D., Michez, D., Vandamme, P., & Smagghe, G. (2021). The holobiont as a key to the adaptation and conservation of wild bees in the anthropocene. Frontiers in Ecology and Evolution, 9, 781470. Magurran, A. E., & McGill, B. J. (2010). Biological diversity: Frontiers in measurement and assessment. OUP Oxford. Mallet, J. (2007). Hybrid speciation. Nature, 446, 279283. Malukiewicz, J., Cartwright, R. A., Dergam, J. A., Igayara, C. S., Kessler, S., Moreira, S. B., Nash, L. T., Nicola, P. A., Pereira, L. C., & Pissinati, A. (2019). The effects of host taxon, hybridization, and environment on the gut microbiome of callithrix marmosets. BioRxiv, 708255. https://doi.org/10.1101/708255 Margulis, L., & Fester, R. (1991). Symbiosis as a source of evolutionary innovation: Speciation and morphogenesis. MIT Press. Martins, N., Pearson, G. A., Gouveia, L., Tavares, A. I., Serrao, E. A., & Bartsch, I. (2019). Hybrid vigour for thermal tolerance in hybrids between the allopatric kelps Laminaria digitata and L. pallida (Laminariales, Phaeophyceae) with contrasting thermal affinities. European Journal of Phycology, 54, 548561. Mayneris-Perxachs, J., Bolick, D. T., Leng, J., Medlock, G. L., Kolling, G. L., Papin, J. A., Swann, J. R., & Guerrant, R. L. (2016). Protein-and zinc-deficient diets modulate the murine microbiome and metabolic phenotype. The American Journal of Clinical Nutrition, 104, 12531262. McCann, K. S. (2000). The diversitystability debate. Nature, 405, 228233. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License CAMPER et al. | McMurdie, P. J., & Holmes, S. (2013). phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE, 8, e61217. Meadows, B. A. (2022). Detecting signatures of discordant population structure and local adaptation in a codiversifying host-obligate endosymbiont mutualism. Texas Tech University. Mrot, C., Debat, V., Le Poul, Y., Merrill, R. M., Naisbit, R. E., Tholance, A., Jiggins, C. D., & Joron, M. (2020). Hybridization and transgressive exploration of colour pattern and wing morphology in Heliconius butterflies. Journal of Evolutionary Biology, 33, 942956. Miller, A. K., Westlake, C. S., Cross, K. L., Leigh, B. A., & Bordenstein, S. R. (2021). The microbiome impacts host hybridization and speciation. PLoS Biology, 19, e3001417. Moeller, A. H., & Sanders, J. G. (2020). Roles of the gut microbiota in the adaptive evolution of mammalian species. Philosophical Transactions of the Royal Society B: Biological Sciences, 375, 20190597. Moran, N. A., & Sloan, D. B. (2015). The hologenome concept: Helpful or hollow? PLoS Biology, 13, e1002311. Morris, M. R., Batra, P., & Ryan, M. J. (1992). Male-male competition and access to females in the swordtail Xiphophorus nigrensis. Copeia, 1992, 980986. Muller, H. J. (1942). Isolating mechanisms, evolution, and temperature. Biology Symposium, 6, 71125. Muoz, M., & Bodensteiner, B. (2019). Janzen's hypothesis meets the Bogert effect: Connecting climate variation, thermoregulatory behavior, and rates of physiological evolution. Integrative Organismal Biology, 1, oby002. Neufeld, K., Kang, N., Bienenstock, J., & Foster, J. A. (2011). Reduced anxiety-like behavior and central neurochemical change in germ- free mice. Neurogastroenterology and Motility, 23, 255264; e119. Nielsen, D. P., Harrison, J. G., Byer, N. W., Faske, T. M., Parchman, T. L., Simison, W. B., & Matocq, M. D. (2023). The gut microbiome reflects ancestry despite dietary shifts across a hybrid zone. Ecology Letters, 26, 6375. Nieuwdorp, M., Gilijamse, P. W., Pai, N., & Kaplan, L. M. (2014). Role of the microbiome in energy regulation and metabolism. Gastroenterology, 146, 15251533. Nishida, A. H., & Ochman, H. (2021). Captivity and the co-diversification of great ape microbiomes. Nature Communications, 12, 5632. Nobs, S. P., Tuganbaev, T., & Elinav, E. (2019). Microbiome diurnal rhythmicity and its impact on host physiology and disease risk. EMBO Reports, 20, e47129. O'Brien, A. M., Laurich, J., & Frederickson, M. E. (2022). Having the right microbiome matters for host trait expression and the strength of mutualism between duckweeds and microbes. bioRxiv, 2022.2002. 2010.479958. O'Brien, A. M., Sawers, R. J., Strauss, S. Y., & Ross-Ibarra, J. (2019). Adaptive phenotypic divergence in an annual grass differs across biotic contexts. Evolution, 73, 22302246. Ochman, H., Worobey, M., Kuo, C.-H., Ndjango, J.-B. N., Peeters, M., Hahn, B. H., & Hugenholtz, P. (2010). Evolutionary relationships of wild hominids recapitulated by gut microbial communities. PLoS Biology, 8, e1000546. Opstal, E. J. V., & Bordenstein, S. R. (2015). Rethinking heritability of the microbiome. Science, 349, 11721173. Orr, H. A. (1995). The population genetics of speciation: The evolution of hybrid incompatibilities. Genetics, 139, 18051813. Orr, H. A., & Turelli, M. (2001). The evolution of postzygotic isolation: Accumulating Dobzhansky-Muller incompatibilities. Evolution, 55, 10851094. Pala, I., & Coelho, M. M. (2005). Contrasting views over a hybrid complex: Between speciation and evolutionary dead-end. Gene, 347, 283294. Patton, A. H., Margres, M. J., Epstein, B., Eastman, J., Harmon, L. J., & Storfer, A. (2020). Hybridizing salamanders experience accelerated diversification. Scientific Reports, 10, 112. CAMPER et al. Pereira, R. J., Barreto, F. S., & Burton, R. S. (2014). Ecological novelty by hybridization: Experimental evidence for increased thermal tolerance by transgressive segregation in Tigriopus californicus. Evolution, 68, 204215. Petchey, O. L., & Gaston, K. J. (2002). Functional diversity (FD), species richness and community composition. Ecology Letters, 5, 402411. Pfennig, K. S. (2021). Biased hybridization and its impact on adaptive introgression. Trends in Ecology & Evolution, 36, 488497. Phillips, C. D., Phelan, G., Dowd, S. E., McDONOUGH, M. M., Ferguson, A. W., Delton Hanson, J., Siles, L., Ordez-Garza, N., San Francisco, M., & Baker, R. J. (2012). Microbiome analysis among bats describes influences of host phylogeny, life history, physiology and geography. Molecular Ecology, 21, 26172627. Pimentel, M., & Lembo, A. (2020). Microbiome and its role in irritable bowel syndrome. Digestive Diseases and Sciences, 65, 829839. Postler, T. S., & Ghosh, S. (2017). Understanding the holobiont: How microbial metabolites affect human health and shape the immune system. Cell Metabolism, 26, 110130. Quinn, T. P., Erb, I., Gloor, G., Notredame, C., Richardson, M. F., & Crowley, T. M. (2019). A field guide for the compositional analysis of any-omics data. GigaScience, 8, giz107. Rebollar, E. A., Antwis, R. E., Becker, M. H., Belden, L. K., Bletz, M. C., Brucker, R. M., Harrison, X. A., Hughey, M. C., Kueneman, J. G., & Loudon, A. H. (2016). Using omics and integrated multi-omics approaches to guide probiotic selection to mitigate chytridiomycosis and other emerging infectious diseases. Frontiers in Microbiology, 7, 68. Rebollar, E. A., Martnez-Ugalde, E., & Orta, A. H. (2020). The amphibian skin microbiome and its protective role against chytridiomycosis. Herpetologica, 76, 167177. Redford, K. H., Segre, J. A., Salafsky, N., del Rio, C. M., & McAloose, D. (2012). Conservation and the microbiome. Conservation Biology: The Journal of the Society for Conservation Biology, 26, 195197. Reed, K. M., & Sites, J. W., Jr. (1995). Female fecundity in a hybrid zone between two chromosome races of the Sceloporus grammicus complex (Sauria, Phrynosomatidae). Evolution, 49, 6169. Ren, C. G., Liu, Z. Y., Wang, X. L., & Qin, S. (2022). The seaweed holobiont: From microecology to biotechnological applications. Microbial Biotechnology, 15, 738754. Robbins, T. R., Pruitt, J. N., Straub, L. E., Mccoy, E. D., & Mushinsky, H. R. (2010). Transgressive aggression in Sceloporus hybrids confers fitness through advantages in male agonistic encounters. Journal of Animal Ecology, 79, 137147. Robbins, T. R., Walker, L. E., Gorospe, K. D., Karl, S. A., Schrey, A. W., McCoy, E. D., & Mushinsky, H. R. (2014). Rise and fall of a hybrid zone: Implications for the roles of aggression, mate choice, and secondary succession. Journal of Heredity, 105, 226236. Ross, B. (19831994). The joy of painting. Sage, R. D., Heyneman, D., Lim, K.-C ., & Wilson, A. C. (1986). Wormy mice in a hybrid zone. Nature, 324, 6063. Sampson, T. R., & Mazmanian, S. K. (2015). Control of brain development, function, and behavior by the microbiome. Cell Host & Microbe, 17, 565576. Sanders, J. G., Powell, S., Kronauer, D. J., Vasconcelos, H. L., Frederickson, M. E., & Pierce, N. E. (2014). Stability and phylogenetic correlation in gut microbiota: Lessons from ants and apes. Molecular Ecology, 23, 12681283. Scheelings, T. F., Moore, R. J., Van, T. T. H., Klaassen, M., & Reina, R. D. (2020). Microbial symbiosis and coevolution of an entire clade of ancient vertebrates: The gut microbiota of sea turtles and its relationship to their phylogenetic history. Animal Microbiome, 2, 112. Scher, J. U., & Abramson, S. B. (2011). The microbiome and rheumatoid arthritis. Nature Reviews Rheumatology, 7, 569578. Seehausen, O. (2004). Hybridization and adaptive radiation. Trends in Ecology & Evolution, 19, 198207. 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 528 Sevellec, M., Laporte, M., Bernatchez, A., Derome, N., & Bernatchez, L. (2019). Evidence for host effect on the intestinal microbiota of whitefish (Coregonus sp.) species pairs and their hybrids. Ecology and Evolution, 9, 1176211774. Shade, A., & Stopnisek, N. (2019). Abundance-occupancy distributions to prioritize plant core microbiome membership. Current Opinion in Microbiology, 49, 5058. Sharon, G., Segal, D., Ringo, J. M., Hefetz, A., Zilber-Rosenberg, I., & Rosenberg, E. (2010). Commensal bacteria play a role in mating preference of Drosophila melanogaster. Proceedings of the National Academy of Sciences of the United States of America, 107, 2005120056. Sorensen, T. (1948). A method of establishing groups of equal amplitude in plant sociology based on similarity of species content and its application to analyses of the vegetation on Danish commons. Biologiske Skrifter, 5, 134. Suzuki, T. A., Fitzstevens, J. L., Schmidt, V. T., Enav, H., Huus, K. E., Mbong Ngwese, M., Griehammer, A., Pfleiderer, A., Adegbite, B. R., & Zinsou, J. F. (2022). Codiversification of gut microbiota with humans. Science, 377, 13281332. Tams, J., Podani, J., & Csontos, P. (2001). An extension of presence/ absence coefficients to abundance data: A new look at absence. Journal of Vegetation Science, 12, 401410. Theien, G. (2006). The proper place of hopeful monsters in evolutionary biology. Theory in Biosciences, 124, 349369. Theien, G. (2009). Saltational evolution: Hopeful monsters are here to stay. Theory in Biosciences, 128, 4351. Thompson, J. N. (1989). Concepts of coevolution. Trends in Ecology & Evolution, 4, 179183. Thompson, J. N. (1994). The coevolutionary process. University of Chicago Press. Thompson, J. N. (2005). The geographic mosaic of coevolution. University of Chicago Press. Todesco, M., Pascual, M. A., Owens, G. L., Ostevik, K. L., Moyers, B. T., Hbner, S., Heredia, S. M., Hahn, M. A., Caseys, C., & Bock, D. G. (2016). Hybridization and extinction. Evolutionary Applications, 9, 892908. Trevelline, B. K., Fontaine, S. S., Hartup, B. K., & Kohl, K. D. (2019). Conservation biology needs a microbial renaissance: A call for the consideration of host-associated microbiota in wildlife management practices. Proceedings of the Royal Society B: Biological Sciences, 286, 20182448. Tripp, E. A., & Manos, P. S. (2008). Is floral specialization an evolutionary dead-end? Pollination system transitions in Ruellia (Acanthaceae). Evolution, 62, 17121737. Turner, L. M., Schwahn, D. J., & Harr, B. (2012). Reduced male fertility is common but highly variable in form and severity in a natural house mouse hybrid zone. Evolution: International Journal of Organic Evolution, 66, 443458. Ubeda, C., Djukovic, A., & Isaac, S. (2017). Roles of the intestinal microbiota in pathogen protection. Clinical & Translational Immunology, 6, e128. Wagg, C., Dudenhffer, J. H., Widmer, F., & Van Der Heijden, M. G. (2018). Linking diversity, synchrony and stability in soil microbial communities. Functional Ecology, 32, 12801292. Wagner, M. R., Roberts, J. H., Balint-Kurti, P., & Holland, J. B. (2020). Heterosis of leaf and rhizosphere microbiomes in field-grown maize. New Phytologist, 228, 10551069. Walker, D. M., Hill, A. J., Albecker, M. A., McCoy, M. W., Grisnik, M., Romer, A., Grajal-Puche, A., Camp, C., Kelehear, C., & Wooten, J. (2019). Variation in the slimy salamander (Plethodon spp.) skin and gut-microbial assemblages is explained by geographic distance and host affinity. Microbial Ecology, 79, 113. Walls, S. C. (2009). The role of climate in the dynamics of a hybrid zone in Appalachian salamanders. Global Change Biology, 15, 19031910. 529 Walter, J., Martnez, I., & Rose, D. J. (2013). Holobiont nutrition: Considering the role of the gastrointestinal microbiota in the health benefits of whole grains. Gut Microbes, 4, 340346. Walters, A. W., Hughes, R. C., Call, T. B., Walker, C. J., Wilcox, H., Petersen, S. C., Rudman, S. M., Newell, P. D., Douglas, A. E., & Schmidt, P. S. (2020). The microbiota influences the Drosophila melanogaster life history strategy. Molecular Ecology, 29, 639653. Wang, J., Kalyan, S., Steck, N., Turner, L. M., Harr, B., Knzel, S., Vallier, M., Hsler, R., Franke, A., & Oberg, H.-H . (2015). Analysis of intestinal microbiota in hybrid house mice reveals evolutionary divergence in a vertebrate hologenome. Nature Communications, 6, 110. West, A. G., Waite, D. W., Deines, P., Bourne, D. G., Digby, A., McKenzie, V. J., & Taylor, M. W. (2019). The microbiome in threatened species conservation. Biological Conservation, 229, 8598. Woodhams, D. C., Bletz, M., Kueneman, J., & McKenzie, V. (2016). Managing amphibian disease with skin microbiota. Trends in Microbiology, 24, 161164. Woodhouse, M. R., Cannon, E. K., Portwood, J. L., Harper, L. C., Gardiner, J. M., Schaeffer, M. L., & Andorf, C. M. (2021). A pan-genomic approach to genome databases using maize as a model system. BMC Plant Biology, 21, 110. Xifra, G., Moreno-Navarrete, J. M., & Fernndez-Real, J. M. (2019). The microbiota and energy balance. In P. Sbraccia & N. Finer (Eds.), Obesity. Endocrinology (pp. 109126). Springer. Zackular, J. P., Baxter, N. T., Iverson, K. D., Sadler, W. D., Petrosino, J. F., Chen, G. Y., & Schloss, P. D. (2013). The gut microbiome modulates colon tumorigenesis. MBio, 4, e00692-13. Zhang, Z., Chen, J., Li, L., Tao, M., Zhang, C., Qin, Q., Xiao, J., Liu, Y., & Liu, S. (2014). Research advances in animal distant hybridization. Science China Life Sciences, 57, 889902. Zheng, H., Nishida, A., Kwong, W. K., Koch, H., Engel, P., Steele, M. I., & Moran, N. A. (2016). Metabolism of toxic sugars by strains of the bee gut symbiont Gilliamella apicola. MBio, 7, e01326-16. Zhu, L., Wang, J., & Bahrndorff, S. (2021). The wildlife gut microbiome and its implication for conservation biology. Frontiers in Microbiology, 12, 1617. Zilber-Rosenberg, I., & Rosenberg, E. (2008). Role of microorganisms in the evolution of animals and plants: The hologenome theory of evolution. FEMS Microbiology Reviews, 32, 723735. S U P P O R T I N G I N FO R M AT I O N Additional supporting information can be found online in the Supporting Information section at the end of this article. Appendix S1: Monotonicity, density invariance and replication invariance. Appendix S2: System pre-analysis. Appendix S3: Example datasets. Appendix S4: Additional figures and tables. How to cite this article: Camper, B. T., Laughlin, Z., Malagon, D., Denton, R., & Bewick, S. (2024). A conceptual framework for host-associated microbiomes of hybrid organisms. Methods in Ecology and Evolution, 15, 511529. https://doi. org/10.1111/2041-210X.14279 2041210x, 2024, 3, Downloaded from https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.14279 by Marian University, Wiley Online Library on [29/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License | CAMPER et al. ...