Islands have long been considered nature's laboratories. By offering a study system with known area, isolation and age, research into island ecosystems has pioneered our understanding of how species assemblages develop and persist over time in relation to abiotic factors. Yet a consistent eco-evolutionary understanding of how biotic interactions -- such as mutualistic interactions between plants, pollinators and seed-dispersers -- operate over contemporary and historical timeframes remains elusive. Here, first we identify challenges that hamper progress in understanding and predicting mutualistic interactions, such as spatio-temporal sampling resolution and the ability to make solid inferences on species' evolutionary trajectories and histories. We then propose how modern molecular approaches provide solutions to these long-standing challenges, opening the possibility of using genomics to advance island research to obtain a novel eco-evolutionary understanding of plant-animal mutualisms. Notably, by outlining empirically testable hypotheses and illustrating how molecular approaches can address long-standing questions when combined with island theory, we provide a roadmap for transforming island mutualism research into a predictive eco-evolutionary science. The need for transformative progress in this field is ever more pressing because already vulnerable island ecosystems continuously become more threatened by anthropogenic impacts, and mainland ecosystems are increasingly being fragmented into habitat islands.

Freshwater habitats are under increasing pressure from numerous anthropogenic forces, including the introduction of alien species capable of altering ecosystems and threatening native species. Although alien species themself are likely to experience loss of genetic diversity when colonising novel environments, some manage to become invasive, suggesting that other factors might facilitate their adaptive capacity. Using a hologenomic approach, we elucidate population genomic trends, the gut microbiota composition and genome-environment-microbiota interaction in the endemic and endangered Spanish toothcarp (Aphanius iberus) and the highly invasive Eastern mosquitofish (Gambusia holbrooki). We found clear genetic signatures of captive breeding in the populations of A. iberus, while G. holbrooki are characterised by an overall low level of heterozygosity and likely signs of multiple introductions. Gut microbial communities of the two species differed significantly across locations, but no sign of increased microbial plasticity was detected in G. holbrooki. However, we report that the genetic profile of each fish was able to explain a considerable part of the microbiota variation measured across individuals. Using shotgun metagenomics, we observed an overall high functional capacity of the microbiota in both species, but we identified no significant differences in the functional capacity between them. The role of the gut microbiota in invasive species and conservation warrant further research using direct comparisons or controlled mesocosm setups, but based on the results of the current study, the gut microbiota of invasive species

The gut microbiomes that associate with animals can represent labile units of cooperating and competing microbes. This lability, sometimes referred to as metagenomic plasticity, has been posited to have an important role as an additional axis of hosts’ phenotypic plasticity. However, whether and how metagenomic plasticity varies across hosts with different ecological and evolutionary features remains unclear. To address this, we utilised faecal-derived genome-resolved metagenomics and compared how the taxonomic, phylogenetic and functional microbial dynamics varied across a series of disturbances in two mammal species; namely, the insectivorous-specialist, Crocidura russula (N = 29) and the omnivorous-generalist Apodemus sylvaticus (N = 22). Although faecal microbial diversity of both species remained stable, compositional dynamics differed significantly. C. russula exhibited substantially higher variability and directionality of microbial responses, with higher predictability associated with each disturbance, compared to A. sylvaticus. Predictions of functional traits using joint-species distribution modelling supported these observations. C. russula showed strong functional response to perturbations, with marked directional variation of various metabolic functions. In contrast, the significantly higher functional diversity and redundancy of the A. sylvaticus microbiome likely buffered its functional response to perturbations, which remained more constant across time. Our results indicate that the intrinsic properties (e.g., diversity, redundancy) of gut microbiomes associated with animals with different biological attributes shape the taxonomic, phylogenetic, and functional response to environmental stressors. This level of plasticity might affect the capacity of animal hosts to acclimate and adapt to changing environments.

Natural predators of arthropods provide an important ecosystem service by preying on crop-damaging species. However, measuring the positive impact of natural pest predators is still challenging. We present a framework to estimate the pest-consumption pressure by natural predators across space and project it onto geographical maps. We use DNA metabarcoding and species distribution modelling to integrate predator density estimations with their energetic requirements and direct pest predation, which yields a comprehensive measure of pest-consumption pressure per time-frame and area. We showcase it on a European bat assemblage, and show that bats consume a variety of pests whose predation pressure varies throughout space. We also report that the impact of different predators depends on spatial scale, and that pest predation pressure is negatively correlated with agricultural intensity. Our framework can be used to estimate broad-scale effects of natural predators on pest arthropods as well as to design research and management strategies.