Modular Pattern Research Articles

Adaptive radiation despite conserved modularity patterns in San Salvador Island Cyprinodon pupfishes and their hybrids

Abstract Adaptive radiations are striking examples of rapid speciation along ecological lines. In adaptive radiations, fast rates of lineage diversification often pair with rapid rates of morphological diversification. Such diversification has often been documented through the lens of ecological drivers, overlooking the intrinsic structural constraints that may also have a key role in configuring patterns of trait diversification. Covariation within and between traits has been hypothesized to govern the axes of trait evolution, either by increasing the degree of covariation between traits (i.e. integration), which promotes morphological coordination, or by strengthening the degree of covariation within traits (i.e. modularity), which allows organisms to explore novel trait combinations and different regions of morphospace. Here, we study the modularity of the skull within an adaptive radiation of pupfishes that is endemic to San Salvador Island, Bahamas. This radiation exhibits divergent craniofacial morphologies, including generalist, snail-eating specialist, and scale-eating specialist species. We assessed morphological disparity, integration strength, and modularity patterns across the sympatric San Salvador Island pupfish radiation, lab-reared hybrids, and closely related outgroup species. Our findings revealed an unexpected uniformity in the pattern of modularity across diverse species, supporting a five-module functional hypothesis comprising the oral jaw, pharyngeal jaw, neurocranium, hyoid apparatus, and hyomandibula. Despite this conserved modularity pattern, all species exhibited weak but significantly varying strengths of overall between-module integration and significant disparity across all cranial regions. Our results suggest rapid morphological diversification can occur even with conserved patterns of modularity. We propose that broadscale patterns of modularity are more conserved while between-module associations are more evolvable between species.

Patterns of Sexual Dimorphism, Static Allometry, and Modularity in the Skull of the Ethiopian White-Footed Mouse, Stenocephalemys albipes (Rodentia: Muridae)

Patterns of Sexual Dimorphism, Static Allometry, and Modularity in the Skull of the Ethiopian White-Footed Mouse, Stenocephalemys albipes (Rodentia: Muridae)

Fish trophic network in karst streams from Brazilian Pantanal headwaters

Abstract Trophic interactions have been a long-standing field of interest in ecology, helping to understand the relationships between organisms and how ecosystems function. In this study, we describe the trophic relationships of fish from karst environments in headwater streams of the upper Paraguai River basin. We analyzed the stomach contents of 81 fish species from the Serra da Bodoquena, calculated the metrics associated with the trophic network, incorporating the body size component in the analyses, and evaluated the participation of each species in network/module connectivity. The analyzed community trophic organization was based mainly on autochthonous items which were the most consumed items for about 30% of fish species. The trophic network showed a modular pattern without nestedness or specialization. However, nestedness was significant within each module, demonstrating a hierarchical compound topology (i.e., species with few connections interacted with subsets of the pairs of more connected species within each module). We also found a relationship between network connectivity and fish body size, in which small species tend to connect modules through generalist feeding strategies. Thus, we demonstrated a still little-known role of small species in fish trophic networks, and how trophic segregation occurs in a highly diverse community from Pantanal headwater streams.

MUSIC CREATIVITY INTERVENTION FOR OLDER ADULTS WITH AND WITHOUT MCI: COGNITIVE BENEFITS AND NEURAL MECHANISMS

Abstract Alzheimer’s and related dementias are debilitating conditions, and at this time, no cure has been identified. Older age is associated with cognitive decline and dementia. Developing and testing innovative non-pharmacological interventions to delay the loss of functional ability and understand the neurobiological mechanisms underlying the intervention’s efficacy is imperative. We conducted a semi-randomized clinical trial evaluating the effects of a novel music creativity curriculum on cognitive and neurological outcomes. Eighty older adults with and without a mild cognitive impairment (MCI) diagnosis were randomized to a 6-week music creativity intervention or a 6-week no-treatment control condition. Before and after the six weeks, both groups participated in a series of self-report and cognitive assessments; they also participated in resting-state fMRI scans to quantify whole-brain network patterns of modularity and flexibility. Those randomized to the music group exhibited improved overall cognitive function and working memory relative to the control group (assessed using standard clinical neuropsychological assessments). The association between being randomized to the music group and increased cognitive function was more robust among those who exhibited greater neural flexibility at baseline. The study shows that music creativity interventions may positively affect cognition, especially among those whose neural network structure is more dynamic or flexible. This study is innovative as it is among the first to identify possible neural mechanisms underlying why music creativity interventions confer a more significant cognitive benefit for some older adults than others.

Age‐related Modularity Changes of the Structural Brain Network: The Framingham Study

AbstractBackgroundWhile studies have consistently shown that brain network efficiency decreases with age, data on age‐related reconfiguration of network modules are scarce. We assessed structural brain networks from healthy adults of the Framingham Study to explore age‐related modularity differences.MethodStructural brain networks of 2,903 healthy adults (mean age 55, SD 0.96, range 25‐95, 53% female) were obtained from DTI probabilistic tractography. Connectivity matrices were constructed from 72 cortical and subcortical regions. First, we calculated efficiency to check the validity of our data. Second, we constructed a multilayer network with each layer representing one age decile and calculated modularity for each layer using a robust implementation of the Louvain community‐detection algorithm (Blondel et al., 2008). Third, we compared modular structures between each layer and a reference group (youngest) using the variation of information (VI), an index measure of distance between two modular patterns (e.g. modularity at different ages). Finally, we calculated the participation coefficient, a measure of modular node integration, and the node flexibility, a measure of module allegiance change and assessed their relationships with age.ResultEfficiency decreased with age (b = ‐0.34, SE 0.02, p<0.001), being stable until approximately 50 and declining linearly thereafter (see Figure 1, panel a). Modularity showed a u‐shaped trajectory with most integration at the fifth decile of age (see Figure1, panel b). VI index revealed that modular reconfiguration was greatest between the youngest and the oldest age decile (0.43). Participation coefficient was highest at the fifth decile (see Figure2, panel c) and in frontal nodes (see Figure2, panel a), where it also exhibited high variability. Frontal variability followed a complex pattern (Figure2, panel b). Conversely, central and temporal‐occipital nodes showed less participation and variability (see Figure2, panels a and b), and the lowest flexibility (see Figure3).ConclusionBrain network modules reconfigure with aging. This reconfiguration is non‐uniform. While midline‐near modules, including the cingulate and parieto‐occipital cortices, show relative stability, frontal modules seemed to reconfigure considerably, as seen by increased average flexibility (change in modular allegiance) and less integration with advancing age, as seen by declining participation coefficients.

Age‐related Modularity Changes of the Structural Brain Network: The Framingham Study

AbstractBackgroundWhile studies have consistently shown that brain network efficiency decreases with age, data on age‐related reconfiguration of network modules are scarce. We assessed structural brain networks from healthy adults of the Framingham Study to explore age‐related modularity differences.MethodStructural brain networks of 2,903 healthy adults (mean age 55, SD 0.96, range 25‐95, 53% female) were obtained from DTI probabilistic tractography. Connectivity matrices were constructed from 72 cortical and subcortical regions. First, we calculated efficiency to check the validity of our data. Second, we constructed a multilayer network with each layer representing one age decile and calculated modularity for each layer using a robust implementation of the Louvain community‐detection algorithm (Blondel et al., 2008). Third, we compared modular structures between each layer and a reference group (youngest) using the variation of information (VI), an index measure of distance between two modular patterns (e.g. modularity at different ages). Finally, we calculated the participation coefficient, a measure of modular node integration, and the node flexibility, a measure of module allegiance change and assessed their relationships with age.ResultEfficiency decreased with age (b = ‐0.34, SE 0.02, p<0.001), being stable until approximately 50 and declining linearly thereafter (see Figure 1, panel a). Modularity showed a u‐shaped trajectory with most integration at the fifth decile of age (see Figure1, panel b). VI index revealed that modular reconfiguration was greatest between the youngest and the oldest age decile (0.43). Participation coefficient was highest at the fifth decile (see Figure2, panel c) and in frontal nodes (see Figure2, panel a), where it also exhibited high variability. Frontal variability followed a complex pattern (Figure2, panel b). Conversely, central and temporal‐occipital nodes showed less participation and variability (see Figure2, panels a and b), and the lowest flexibility (see Figure3).ConclusionBrain network modules reconfigure with aging. This reconfiguration is non‐uniform. While midline‐near modules, including the cingulate and parieto‐occipital cortices, show relative stability, frontal modules seemed to reconfigure considerably, as seen by increased average flexibility (change in modular allegiance) and less integration with advancing age, as seen by declining participation coefficients.

Evaluating modularity in the hominine skull related to feeding biomechanics.

Modular architecture of traits in complex organisms can be important for morphological evolution at micro- and sometimes macroevolutionary scales as it may influence the tempo and direction of changes to groups of traits that are essential for particular functions, including food acquisition and processing. We tested several distinct hypotheses about craniofacial modularity in the hominine skull in relation to feeding biomechanics. First, we formulated hypothesized functional modules for craniofacial traits reflecting specific demands of feeding biomechanics (e.g., masseter leverage/gape or tooth crown mechanics) in Homo sapiens, Pan troglodytes, and Gorilla gorilla. Then, the pattern and strength of modular signal was quantified by the covariance ratio coefficient and compared across groups using covariance ratio effect size. Hierarchical clustering analysis was then conducted to examine whether a priori-defined functional modules correspond to empirically recovered clusters. There was statistical support for most a priori-defined functional modules in the cranium and half of the functional modules in the mandible. Modularity signal was similar in the cranium and mandible, and across the three taxa. Despite a similar strength of modularity, the empirically recovered clusters do not map perfectly onto our priori functional modules, indicating that further work is needed to refine our hypothesized functional modules. The results suggest that modular structure of traits in association with feeding biomechanics were mostly shared with humans and the two African apes. Thus, conserved patterns of functional modularity may have facilitated evolutionary changes to the skull during human evolution.

The impact of the land-to-sea transition on evolutionary integration and modularity of the pinniped backbone

In this study, we investigate how the terrestrial-aquatic transition influenced patterns of axial integration and modularity in response to the secondary adaptation to a marine lifestyle. We use 3D geometric morphometrics to quantify shape covariation among presacral vertebrae in pinnipeds (Carnivora; Pinnipedia) and to compare with patterns of axial integration and modularity in their close terrestrial relatives. Our results indicate that the vertebral column of pinnipeds has experienced a decrease in the strength of integration among all presacral vertebrae when compared to terrestrial carnivores (=fissipeds). However, separate integration analyses among the speciose Otariidae (i.e., sea lions and fur seals) and Phocidae (i.e., true seals) also suggests the presence of different axial organizations in these two groups of crown pinnipeds. While phocids present a set of integrated “thoracic” vertebrae, the presacral vertebrae of otariids are characterized by the absence of any set of vertebrae with high integration. We hypothesize that these differences could be linked to their specific modes of aquatic locomotion –i.e., pelvic vs pectoral oscillation. Our results provide evidence that the vertebral column of pinnipeds has been reorganized from the pattern observed in fissipeds but is more complex than a simple “homogenization” of the modular pattern of their close terrestrial relatives.

Comparative analysis of group information-guided independent component analysis and independent vector analysis for assessing brain functional network characteristics in autism spectrum disorder.

Group information-guided independent component analysis (GIG-ICA) and independent vector analysis (IVA) are two methods that improve estimation of subject-specific independent components in neuroimaging studies. These methods have shown better performance than traditional group independent component analysis (GICA) with respect to intersubject variability (ISV). In this study, we compared the patterns of community structure, spatial variance, and prediction performance of GIG-ICA and IVA-GL, respectively. The dataset was obtained from the publicly available Autism Brain Imaging Data Exchange (ABIDE) database, comprising 75 healthy controls (HC) and 102 Autism Spectrum Disorder (ASD) participants. The greedy rule was used to match components from IVA-GL and GIG-ICA in order to compare the similarities between the two methods. Robust correspondence was observed between the two methods the following networks: cerebellum network (CRN; |r| = 0.7813), default mode network (DMN; |r| = 0.7263), self-reference network (SRN; |r| = 0.7818), ventral attention network (VAN; |r| = 0.7574), and visual network (VSN; |r| = 0.7503). Additionally, the Sensorimotor Network demonstrated the highest similarity between IVA-GL and GIG-ICA (SOM: |r| = 0.8125). Our findings revealed a significant difference in the number of modules identified by the two methods (HC: p < 0.001; ASD: p < 0.001). GIG-ICA identified significant differences in FNC between HC and ASD compared to IVA-GL. However, in correlation analysis, IVA-GL identified a statistically negative correlation between FNC of ASD and the social total subscore of the classic Autism Diagnostic Observation Schedule (ADOS: pi = -0.26, p = 0.0489). Moreover, both methods demonstrated similar prediction performances on age within specific networks, as indicated by GIG-ICA-CRN (R2 = 0.91, RMSE = 3.05) and IVA-VAN (R2 = 0.87, RMSE = 3.21). In summary, IVA-GL demonstrated lower modularity, suggesting greater sensitivity in estimating networks with higher intersubject variability. The improved age prediction of cerebellar-attention networks underscores their importance in the developmental progression of ASD. Overall, IVA-GL may be appropriate for investigating disorders with greater variability, while GIG-ICA identifies functional networks with distinct modularity patterns.

The oldest known clones of Salix herbacea growing in the Northern Apennines, Italy are at least 2000 years old.

Dominant in many ecosystems around the world, clonal plants can reach considerable ages and sizes. Due to their modular growth patterns, individual clonal plants (genets) can consist of many subunits (ramets). Since single ramets do not reflect the actual age of genets, the ratio between genet size (radius) and longitudinal annual growth rate (LAGR) of living ramets is often used to approximate the age of clonal plants. However, information on how the LAGR changes along ramets and how LAGR variability may affect age estimates of genets is still limited. We assessed the variability of LAGR based on wood-section position along the ramets and on the duration of the growing season on three genetically distinct genets of Salix herbacea growing in the Northern Apennines (Italy). We compared genet ages estimated by dividing genet radius by the LAGRs of its ramets. LAGR increased significantly from the stem apex to the root collar; indicating that ramet growth rate decreased with time. Furthermore, a difference of ca. 2 weeks in the onset of the growing period did not impact LAGR. Considering the high LAGR variability, we estimated that the three genets started to grow between ~2100 and ~7000 years ago, which makes them the oldest known clones of S. herbacea even considering the most conservative age estimate. Our findings indicate that analyzing ramets at the root collar provides an integrative measurement of their overall LAGR, which is crucial for estimating the age of genets.

Lower jaw modularity in the African clawed frog (Xenopus laevis) and fire salamander (Salamandra salamandra gigliolii)

Abstract Modularity describes the degree to which the components of complex phenotypes vary semi-autonomously due to developmental, genetic and functional correlations. This is a key feature underlying the potential for evolvability, as it can allow individual components to respond to different selective pressures semi-independently. The vertebrate lower jaw has become a model anatomical system for understanding modularity, but to date most of this work has focused on the mandible of mammals and other amniotes. In contrast, modularity in the mandible of lissamphibians has been less well studied. Here, we used geometric morphometrics to quantify the static (intraspecific) modularity patterns in Xenopus laevis and Salamandra salamandra gigliolii. We tested developmental and functional hypotheses of modularity and demonstrate that both species exhibit significant modularity. Functional modularity was supported in both Xenopus and Salamandra. Allometry has a small yet significant impact on lower jaw shape in both taxa and sex has a significant effect on shape in Xenopus. The high lower jaw modularity in both species observed here, combined with the well-established modularity of the amphibian cranium, suggests that modularity is a ubiquitous feature of the tetrapod head.

Design and Fabrication of an Easily Assembled Anode Supported Microtubular SOFC Stack

Microtubular solid oxide fuel cells have shown many advantages that make them suitable for mobile and portable applications. In this study, a microtubular SOFC stack consisting of 8 cells was designed and fabricated using a metallic current collector and interconnect, enabling a pluggable modular design pattern. To avoid the difficulty of high-temperature sealing, the cold sealing method was adopted, where the temperature at the sealing position was under 130°C when the stack was operated at 700°C. The open circuit voltage (OCV) reached over 4 V while the power output reached over 20 W with a 2 (in series)×4 (in parallel) array. The stack module can be further connected in series or in parallel to satisfy larger power requirements.

The Type I Interferon Signature Reflects Multiple Phenotypic and Activity Measures in Dermatomyositis.

The type 1 interferon (IFN) pathway is up-regulated in dermatomyositis (DM). We sought to define how organ-specific disease activity as well as autoantibodies and other clinical factors are independently associated with systemic type I IFN activity in adult patients with DM. RNA sequencing was performed on 355 whole blood samples collected from 202 well-phenotyped DM patients followed up during the course of their clinical care. A previously defined 13-gene type I IFN score was modeled as a function of demographic, serologic, and clinical variables using both cross-sectional and longitudinal data. The pattern of type I IFN-driven transcriptional response was stereotyped across samples with a sequential modular activation pattern strikingly similar to systemic lupus erythematosus. The median type I IFN score was higher or lower in patients with anti-melanoma differentiation-associated protein 5 (anti-MDA-5) or anti-Mi-2 antibodies, respectively, compared to patients without these antibodies. Absolute type I IFN score was independently associated with muscle and skin disease activity, interstitial lung disease, and anti-MDA-5 antibodies. Changes in the type I IFN score over time were significantly associated with changes in skin or muscle disease activity. Stratified analysis accounting for heterogeneity in organ involvement and antibody class revealed high correlation between changes in the type I IFN score and skin disease activity (Spearman's ρ=0.84-0.95). The type I IFN score is independently associated with skin and muscle disease activity as well as certain clinical and serologic features in DM. Accounting for the effect of muscle disease and anti-MDA-5 status revealed that the type I IFN score is strongly correlated with skin disease activity, providing support for type I IFN blockade as a therapeutic strategy for DM.

EasyRec: An Easy-to-Use, Extendable and Efficient Framework for Building Industrial Recommendation Systems

We present EasyRec, an easy-to-use, extendable and efficient recommendation framework for building industrial recommendation systems. Our EasyRec framework is superior in the following aspects:first, EasyRec adopts a modular and pluggable design pattern to reduce the efforts to build custom models; second, EasyRec implements hyper-parameter optimization and feature selection algorithms to improve model performance automatically; third, EasyRec applies online learning to adapt to the ever-changing data distribution. The code is released: https://github.com/alibaba/EasyRec.

Relationship between flight muscle dimorphism and wing morphometry in Triatoma infestans (Klug, 1834) (Hemiptera, Reduviidae, Triatominae)

Triatoma infestans, the primary vector of Chagas disease in southern South America, is reported as a dimorphic species for flight muscle development and presents a marked phenotypic plasticity in traits related to flight. In triatomines, studying the relationship between flight muscle development and wing variation is important, since it involves flight capacity and dispersion. Therefore, we examined the association between flight muscle dimorphism and morphological wing variation in T. infestans individuals. We hypothesized that if the underdeveloped flight muscle were associated with a decrease in wing developmental stability, canalization and morphological aspect of wings, and with a loss of wing modularity, then wing variation would be associated with flight muscle dimorphism. Wing size and shape were characterized using geometric morphometrics. Patterns of canalization, developmental instability, and modularity were inferred from fluctuating asymmetry analysis of wing shape. Morphological aspects of wings were estimated using maximum wing length and wing area. The results showed that there are differences in wing shape but not in wing size among groups. Canalization and developmental stability were positively associated in individuals with muscle development and negatively associated in individuals with underdeveloped flight muscles. The covariance structure of variation among individuals was different between groups but not for fluctuating asymmetry, both for females and males. Modularity hypothesis related to the wing subdivision into two parts, a stiff part and a membranous one, showed significant association for the variation among individuals and fluctuating asymmetry variation in females with flight muscles. Wing aspect showed that females with flight muscles exhibited more elongated and thinner wings than those without flight muscles. The results obtained in this study reveal an association between flight muscle development and forewing variation in T. infestans. Females with flight muscles have certain differential characteristics in their wings that would be related to an enhanced flight performance.

Elevational variation of spider and insect communities in the Swedish mountains

AbstractMountain topography gives rise to often dramatic climate‐driven elevation gradients in primary productivity, which can generate substantial biodiversity variation. Therefore, mountain areas may be particularly useful for evaluating the ecological consequences of climate change. Arthropods are the most diverse animal phylum, which play important roles in most ecosystems. However, despite their ecological importance, we have limited information on how arthropods vary along elevation gradients. We investigated how taxonomic richness, taxonomic composition, and spatial structuring of spider and insect communities varied along elevation gradients and among three geographic locations in a mountain region of northern Sweden. The locations provided a latitude gradient spanning approximately 3° (from 62° N to 65° N), but were otherwise selected to contain similar environmental characteristics. Taxonomic richness of both spiders and insects declined monotonically with increasing elevation, and there were limited differences between the geographic locations in such declines. Taxonomic composition varied with elevation for both taxonomic groups, but also differed among the three sites. Linyphiid spiders were more widely distributed along the elevation gradients than other spider taxa, whereas a broad taxonomic range of insects occurred over almost all elevations. We observed nested as well as modular spatial distributions of both spider and insect communities along the elevation gradients. While the modular patterns suggest that species turnover has generated distinct communities at different elevations, some generalist species were still widespread throughout large parts of the gradients. Our results point to smaller differences among geographic locations than among taxonomic groups in how taxonomic richness and community structuring varied with elevation. We interpret these results as support for taxonomically specific adaptations to environmental conditions being important for structuring arthropod communities. We also suggest that climate‐driven changes to arthropod communities in mountain environments may be regulated by two not mutually exclusive processes, one in which generalist species may become more dominant and shift their ranges upward and one in which high‐elevation specialists may go extinct because of increasingly fragmented habitats.

Design and Fabrication of an Easily Assembled Anode Supported Microtubular SOFC Stack

Anode-supported micro-tubular solid oxide fuel cell has shown many advantages which make it suitable for portable devices. In this study, we developed a micro-tubular SOFC stack and module using a metallic current collector and interconnector which enabling pluggable modular design pattern. In order to ensure the sealing performance, we adopted the low-temperature sealing method and the temperature at the sealing position is under 130℃ while the stack is operated at 750℃.The stack was made of 8 micro-tubular cells, which were connected in series and parallel, in a 2×4 array. The open circuit voltage (OCV) was above 4.22V, thus the single cell’s OCV value could reach about 1.06V. The stack module can be further connected in series or in parallel to satisfy larger power requirements.

Characterization of altered molecular mechanisms in Parkinson's disease through cell type-resolved multiomics analyses.

Parkinson's disease (PD) is a progressive neurodegenerative disorder. However, cell type-dependent transcriptional regulatory programs responsible for PD pathogenesis remain elusive. Here, we establish transcriptomic and epigenomic landscapes of the substantia nigra by profiling 113,207 nuclei obtained from healthy controls and patients with PD. Our multiomics data integration provides cell type annotation of 128,724 cis-regulatory elements (cREs) and uncovers cell type-specific dysregulations in cREs with a strong transcriptional influence on genes implicated in PD. The establishment of high-resolution three-dimensional chromatin contact maps identifies 656 target genes of dysregulated cREs and genetic risk loci, uncovering both potential and known PD risk genes. Notably, these candidate genes exhibit modular gene expression patterns with unique molecular signatures in distinct cell types, highlighting altered molecular mechanisms in dopaminergic neurons and glial cells including oligodendrocytes and microglia. Together, our single-cell transcriptome and epigenome reveal cell type-specific disruption in transcriptional regulations related to PD.

Modular ontology modeling

Reusing ontologies for new purposes, or adapting them to new use-cases, is frequently difficult. In our experiences, we have found this to be the case for several reasons: (i) differing representational granularity in ontologies and in use-cases, (ii) lacking conceptual clarity in potentially reusable ontologies, (iii) lack and difficulty of adherence to good modeling principles, and (iv) a lack of reuse emphasis and process support available in ontology engineering tooling. In order to address these concerns, we have developed the Modular Ontology Modeling (MOMo) methodology, and its supporting tooling infrastructure, CoModIDE (the Comprehensive Modular Ontology IDE – “commodity”). MOMo builds on the established eXtreme Design methodology, and like it emphasizes modular development and design pattern reuse; but crucially adds the extensive use of graphical schema diagrams, and tooling that support them, as vehicles for knowledge elicitation from experts. In this paper, we present the MOMo workflow in detail, and describe several useful resources for executing it. In particular, we provide a thorough and rigorous evaluation of CoModIDE in its role of supporting the MOMo methodology’s graphical modeling paradigm. We find that CoModIDE significantly improves approachability of such a paradigm, and that it displays a high usability.

Contrasting nation-wide citizen science and expert collected data on hummingbird–plant interactions

Citizen science has the potential to increase the efficiency of scientific data collection. However, such initiatives often focus on unique taxa for each record, not necessarily involving interspecific interactions. Moreover, whether openly available unstructured citizen science data can contribute to better understand ecological patterns is still not well understood. Here, we identify hummingbird-plant interactions recorded by amateur birdwatchers in the most popular online platform in Brazil, Wikiaves. Then, we evaluated how this information can benefit our understanding of interactions in a large Tropical country by comparing with data generated by experts. We also constructed a nation-wide meta-network to identify the structural roles of hummingbirds and plants. In total, 3210 interactions were compiled, with better hummingbirds and geographic coverage of citizen data in relation to expert data. The interaction network showed a modular pattern, and some plant species found as most frequently interacting here were similar to those found by experts. Nevertheless, when comparing the plant partners for hummingbirds featured in both expert and citizen data, the proportion of plants in common were generally low (usually less than 40%), indicating that amateur birdwatchers are mostly recording interactions not captured by scientists. Finally, as in other cases of compilation of interaction data, we found that sampling intensity (here, number of photographs) is a strong driver of interaction records, highlighting the unique challenge of separating biologically meaningful patterns from sampling artifacts in citizen science data. Our study illustrates the richness of citizen-gathered biodiversity data available in a megadiverse country, which show great potential to complement expert collected data.