Bridging muscle and bone: shared genetic signatures of osteosarcopenia revealed by reverse Gene Ontology lookup

Bridging muscle and bone: shared genetic signatures of osteosarcopenia revealed by reverse Gene Ontology lookup | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Bridging muscle and bone: shared genetic signatures of osteosarcopenia revealed by reverse Gene Ontology lookup Simon Žetko, Lan Vukolić, Aljoša Škorjanc, Vladimir Smrkolj, Nejc Umek This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8000419/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Osteosarcopenia, the coexistence of osteoporosis and sarcopenia, is an emerging geriatric syndrome that markedly increases frailty, fracture risk, and disability. Although traditionally considered distinct conditions, growing evidence indicates that both share convergent molecular pathways linking bone and muscle degeneration. To elucidate their shared genetic background, we applied GOReverseLookup, a Gene Ontology–based reverse lookup tool, using 367 curated GO terms representing key disease mechanisms. The analysis, integrating ortholog data from four vertebrate species, identified 37 genes significantly enriched for annotations related to both disorders (FDR < 5 × 10⁻⁸). Among these, 14 genes had no prior association with either condition, and five (RBCK1, TERF2IP, LGALS9, ZBTB7A, RPS3) showed moderate expression in both bone and muscle. KEGG enrichment revealed 15 pathways, prominently NF-κB, TNF, osteoclast differentiation, necroptosis, and cytosolic DNA-sensing signaling, highlighting an inflammatory and immune-regulatory convergence. These findings identify novel, biologically plausible candidates and pathways underlying osteosarcopenia, and validate GOReverseLookup as a transparent approach for cross-phenotype gene discovery. The integration of GO-based inference with expression and pathway analysis provides a reproducible framework for exploring multi-tissue syndromes. Health sciences/Biomarkers Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Biological sciences/Genetics Health sciences/Medical research GOReverseLookup Osteosarcopenia Gene discovery Muscle-bone crosstalk NF-κB signaling Necroptosis Cytosolic DNA sensing Figures Figure 1 Figure 2 Introduction Osteosarcopenia is an emerging geriatric syndrome and an increasingly recognized global health concern. A recent meta-analysis estimated its pooled prevalence at approximately 21% ( 1 , 2 ). The condition reflects the coexistence of two age-related disorders, osteoporosis and sarcopenia, each contributing to frailty and fracture risk. Osteoporosis is characterized by reduced bone mass and microarchitectural deterioration, resulting in skeletal fragility and a heightened susceptibility to fractures ( 3 ). Sarcopenia, by contrast, involves the progressive loss of skeletal muscle mass and strength, which underlies physical disability, diminished quality of life, and increased mortality ( 4 ). Although traditionally studied as distinct entities, accumulating evidence indicates substantial overlap in their pathogenesis and risk factors including aging, hormonal decline, nutritional deficiency, and sedentary behavior ( 2 , 5 ). The concurrent manifestation of both conditions synergistically exacerbates the risk of falls, fractures, and long-term disability ( 5 ). At the molecular level, osteoporosis and sarcopenia arise from distinct yet parallel disruptions of tissue homeostasis. In osteoporosis, bone resorption exceeds bone formation, leading to a progressive decline in bone mass and an increased risk of fracture ( 2 ). The balance between osteoclast-mediated resorption and osteoblast-driven formation is governed by several key signalling axes, including the receptor activator of nuclear factor κB ligand/receptor activator of nuclear factor κB/osteoprotegerin (RANKL/RANK/OPG) system, Wnt/β-catenin, and transforming growth factor-β/bone morphogenetic protein (TGF-β/BMP) pathways ( 6 ). Additional factors, such as estrogen deficiency in postmenopausal women and chronic inflammation, further exacerbate the imbalance in bone remodeling, contributing to a loss in bone density ( 7 ). In parallel, sarcopenia is driven by impaired muscle protein synthesis and enhanced proteolytic activity. Age-related changes in muscle architecture, including the selective atrophy of type II fibers and reduced satellite cell regenerative capacity, further compromise muscle maintenance and repair ( 8 ). The pathological overlap between osteoporosis and sarcopenia arises from the interdependent signaling networks that link bone and muscle homeostasis. Both tissues originate from mesenchymal stem cells and remain closely interlinked through paracrine, endocrine and mechanical stimuli ( 5 , 9 ). Myokines, such as myostatin, insulin-like growth factor 1 (IGF-1), irisin, and fibroblast growth factor 2 (FGF-2) have been implicated in bone remodeling ( 10 ), whereas osteokines including osteocalcin and TGF-β influence muscle metabolism and contractile function ( 11 ). The Wnt signalling pathway, essential for bone formation, also plays a pivotal role in muscle regeneration ( 12 ). Chronic low-grade inflammation contributes to the atrophy of both tissues, as pro-inflammatory signals accelerate osteoclastogenesis and proteolysis ( 10 ). Oxidative stress and hormonal dysregulation further amplify these pathological processes ( 13 ). The reciprocal communication between muscle and bone appears essential for maintaining musculoskeletal function, disruptions in this signaling can contribute to osteosarcopenia. These insights underscore the need for therapeutic strategies that target skeletal fragility and muscle wasting simultaneously rather than in isolation ( 14 ). Identifying genes that are shared drivers of osteosarcopenia is essential for the development of targeted therapies. Achieving this goal typically relies on computational approaches, each of which introduces its own methodological biases and technical limitations. Existing computational pipelines for gene discovery span variant-centric, expression-centric and ontology-centric paradigms. Genome-wide association studies (GWAS) have identified thousands of loci associated with complex traits, however, most effect sizes are modest, and replication demands very large cohorts ( 15 ). Transcriptome-wide association studies (TWAS) improve functional inference but can inflate false-positive rates when genetic confounding is not properly accounted ( 16 ). Phenome-wide association studies (PheWAS) invert the GWAS logic by scanning across clinical phenotypes, yet remain susceptible to phenotype misclassification and a substantial multiple-testing burden ( 17 ). Convergent functional genomics (CFG) and related evidence integration frameworks aim to prioritize loci by aggregating heterogeneous data, but their outputs can be dominated by well-studied pathways and literature bias ( 18 ). Over the past five years, representation-learning approaches including hypergraph embeddings, protein-interaction graph neural networks and ontology vectorizations such as Onto2Vec and SmuDGE, have captured latent structure across GO, OMIM and HPO to predict disease genes, yet at the cost of “black-box” interpretability and continued reliance on existing annotations ( 19 , 20 ). Collectively, these methods either lack interpretability, demand extremely large datasets, or over-prioritize well-trodden pathways, leaving emerging, biologically plausible genes unexplored. A recently developed tool GOReverseLookup provides a complementary route. Instead of starting from variants or embeddings, researchers curate GO terms that define a phenotype or broader state of interest and the tool then evaluates which genes are statistically over- or under-represented in that user-defined state. This work presents the first disease-focused application of GOReverseLookup, demonstrating its utility as a broadly applicable framework for gene discovery that generates transparent, biologically contextual, and interpretable candidate lists well suited to early-stage hypothesis generation ( 21 ). Accordingly, the aim of this study was to discover new genes implicated in pathogenesis of both osteoporosis and sarcopenia. Materials and Methods Identification of candidate genes using the GOReverseLookup tool Candidate genes were identified using GOReverseLookup, a computational tool that prioritizes genes based on their statistical association with a curated set of Gene Ontology (GO) terms that collectively define a biological phenotype. The tool's workflow consists of three main stages: input curation, annotation retrieval, and statistical analysis, with only the input curation performed by the researcher and all subsequent steps handled automatically by GOReverseLookup. First, two states of interest (SOIs) were defined based on the research objective: “osteoporosis” and “sarcopenia.” For each SOI, relevant Gene Ontology (GO) terms were manually curated and manually annotated as either stimulatory or inhibitory with respect to the SOI’s pathology. Key analytical parameters were then specified, including the depth of indirect annotations, p -value cutoff threshold, filtering by evidence codes, ortholog selection, and the choice of a multiple testing correction method. Next, the tool retrieved all corresponding gene annotations from the GO database based on the selected GO terms and parameters. When cross-species analysis is specified, ortholog mappings (e.g., from Ensembl or gProfiler) are integrated, and the complete set of annotations is filtered according to the pre-selected evidence codes. Finally, GOReverseLookup performed statistical analysis to quantify the association between each gene and the defined SOI. For every gene a 2 x 2 contingency table was constructed, and enrichment was assessed using Fisher’s exact test. The resulting p -values were corrected for multiple comparisons (e.g. false discovery rate), and genes meeting the significance threshold were reported as a prioritized candidate list. Curation of GO Terms A curated list of GO terms representing the biological processes underlying osteoporosis and sarcopenia was generated following a systematic workflow adapted from Škorjanc et al. ( 21 ). The curation process began with an extensive literature review to delineate key pathophysiology for each condition. Each mechanistic group was then translated into a representative keyword (e.g., "osteoclast," "satellite cell," "estrogen") for targeted searches within the QuickGO database, restricted to the biological process sub-ontology. Relevant GO terms were manually compiled using QuickGO's basket feature, with a consistent preference for specific child terms (e.g., "positive regulation of osteoclast proliferation") to leverage the tool's ability to map to parent annotations. The resulting list was annotated and structured for analysis, then independently reviewed by two researchers experienced in bone and muscle biology, yielding a final set of 367 GO terms representing the major biological processes implicated in osteoporosis and sarcopenia. The complete and categorized list of 367 GO terms is provided in Supplementary Information 1. Reverse Enrichment Analysis Reverse enrichment analysis was performed using GOReverseLookup (v1.0.56; Python 3.11) on June 27, 2025. The analysis targeted Homo sapiens , incorporating ortholog data from Danio rerio , Rattus norvegicus , Mus musculus , and Xenopus tropicalis . Key parameters included the use of indirect annotations up to three parent levels and the exclusion of electronically inferred annotations (IEA). A stringent significance threshold was set at a p -value < 5 × 10⁻⁸, and the resulting p -values were adjusted for multiple testing using the Benjamini–Hochberg false discovery rate (FDR-BH) procedure. To ensure reproducibility, all input files are made readily available at ( https://github.com/GenomeDev/GOReverseLookup ). We used the Gene Ontology OBO (format-version 1.2; data-version releases/2025-03-16), the Homo sapiens UniProt-GOA annotation set (GO Annotations File version 2.2; date-generated 2024-06-19), and GO metadata via the GO API (release 2025-06-26). Orthologs to human (NCBITaxon:9606) were mapped from mouse (MGI; 10090), rat (RGD; 10116), zebrafish (ZFIN; 7955), and tropical clawed frog (Xenbase; 8364). Systematic Characterization of Candidate Genes Candidate genes identified by GOReverseLookup were further characterized through a systematic three-step validation and functional annotation process. 1. Literature Validation and Disease Association A comprehensive literature search was performed in PubMed and Google Scholar to identify existing evidence linking each gene to the diseases of interest. Search queries combined the gene symbol with the keywords "osteoporosis," "sarcopenia," and "osteosarcopenia." In parallel, gene-disease associations were quantified using the DisGeNET database version 25.1.1. For each gene, we recorded the evidence-weighted Gene-Disease Association (GDA) score, which aggregates contributions from expert-curated human sources (0.30 each), animal-model databases (0.10 each) and PubMed text-mining hits (small, cumulative weights). Scores range from 0 to 1, with values greater than 0.30 indicating support from at least one expert-curated human source (22). The primary objective of this step was to determine whether a gene had been previously associated with either osteoporosis, sarcopenia, or their co-manifestation (osteosarcopenia) in the scientific literature or curated databases. 2. Tissue-Specific Expression Gene expression levels in disease-relevant human tissues were assessed using the Expression Atlas (EMBL-EBI). For each gene, the highest recorded expression level was extracted across muscle-related tissues ("muscle of arm," "muscle of leg," "skeletal muscle," "skeletal muscle of trunk," and "skeletal muscle tissue") and bone tissue ("cervical vertebrae"). Expression levels were categorized by transcripts per million (TPM) as high (> 1000), medium (> 11), low (> 0.5), or below cutoff (≤ 0.5 TPM). It should be noted that bone tissue datasets in Expression Atlas are limited in diversity and coverage (23). 3. Pathway Enrichment Analysis To identify enriched biological pathways, KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment pathway analysis was performed using ShinyGO v0.82 ( https://bioinformatics.sdstate.edu/go/ ). The background reference gene set was defined as all human protein-coding genes annotated in the KEGG database. Pathways were ranked according to fold enrichment and considered significantly enriched at a Benjamini–Hochberg adjusted p -value (FDR) < 5 × 10⁻ 5 . Results Identified and Categorized Candidate Genes Our analysis identified 37 genes significantly associated with both osteoporosis and sarcopenia simultaneously (FDR < 5 × 10⁻⁸). To prioritize these candidates for further follow-up analysis, we classified them according to prior literature evidence and tissue-specific expression patterns in bone and skeletal muscle (Table 1 and Figure 2). Comprehensive statistical outputs are provided in Supplementary Information 2. The analysis highlighted 14 genes with no previously reported association to either disease. Among these, five genes ( RBCK1, TERF2IP, LGALS9, ZBTB7A, and RPS3 ) exhibited at least moderate (medium) expression in both bone and muscle and no existing disease association score (GDA ≈ 0). The remaining nine genes ( PYCARD, MALT1, BCL2L11, PRKCQ, CARD9, IL18R, ANKRD42, BTK, NOD2 ) displayed detectable but lower (low) expression in both tissues. In total, nine genes (TNFSF11, STAT3, IL1B, AGER, CLU, SPHK, IL18, TP53, FOXO3) were previously described in the literature as being associated with both osteoporosis and sarcopenia and have a non-zero GDA score for these conditions. Notably, two of these genes (TNFSF11, STAT3) have been directly implicated in osteosarcopenia (24,25). The remaining 14 significant genes showed either evidence linked to only one of the two conditions or expression levels below the defined cutoff expression level in one or both tissue types. KEGG pathway enrichment KEGG enrichment of the 37 osteosarcopenia-associated genes yielded 15 significantly overrepresented pathways that exceeded the predefined significance threshold (FDR < 5 × 10⁻⁵) and showed a fold-enrichment (FE) higher than 25. To emphasize biological mechanisms rather than incidental annotation overlap, only non-infectious pathways are presented in Table 2. The complete list, including infection-related pathways (e.g., Malaria , Shigellosis , Tuberculosis ), is provided in Supplementary Information 3. Pathways are ranked by FE, the effect-size metric recommended by ShinyGO for assessing biological relevance. The most enriched was the NF-κB signalling pathway (FE = 57.9), followed by the cytosolic DNA-sensing pathway (FE = 47.8) and the NOD-like-receptor signalling pathway (FE = 39.5). Additional enriched pathways included prolactin signalling (FE = 34.4), C-type lectin receptor signalling (FE = 28.9), osteoclast differentiation (FE = 28.7), TNF signalling (FE = 26.9), necroptosis (FE = 26.5) and lipid and atherosclerosis (FE = 25.3). Discussion This study represents the first integrative application of reverse Gene Ontology analysis to a complex musculoskeletal phenotype. Our analysis identified 37 genes associated with osteosarcopenia, systematically categorized by prior disease associations, expression across bone and skeletal muscle, and their pathway context derived from KEGG enrichment. This integrative framework recovered well-established mediators, serving as an internal validation of the approach, while simultaneously uncovering underexplored candidates that may reveal novel mechanisms underlying the coordinated regulation of bone and muscle homeostasis. Incorporating tissue-specific expression further refines gene prioritization, as genes expressed in both bone and muscle are more likely to influence shared metabolic and inflammatory signaling axes that maintain musculoskeletal integrity. However, genes with relatively low expression yet high functional relevance, such as TNFSF11 and IL1B , emphasize the importance of interpreting expression data in a mechanistic context rather than relying on quantitative thresholds alone. Genes without prior association to either osteoporosis or sarcopenia represent particularly valuable, underexplored candidates that may illuminate biological mechanisms not yet captured by current literature. Five such genes, RBCK1, TERF2IP, LGALS9, ZBTB7A , and RPS3 , showed moderate expression in both bone and skeletal muscle, suggesting a possible contribution to the shared molecular and inflammatory networks underpinning osteosarcopenia. RBCK1 encodes HOIL-1, a catalytic component of the linear ubiquitin chain assembly complex (LUBAC), one of the three subunits responsible for M-1–linked ubiquitination (26). LUBAC regulates multiple immune signaling cascades, particularly through activation of the NF-κB pathway (27), a mechanism consistently enriched in our KEGG analysis. Beyond inflammation, HOIL-1 also modulates apoptosis and necroptosis (28), two processes essential for tissue remodeling in both muscle and bone. Mutations in RBCK1 cause RBCK1-associated polyglucosan body myopathy-1 (PGBM1), a glycogen storage disorder that manifests with skeletal muscle weakness and cardiomyopathy (29), further supporting its potential relevance to musculoskeletal pathology. Similarly, LGALS9 encodes Galectin-9, a β-galactoside-binding lectin with broad immunomodulatory functions (30). In bone, Galectin-9 suppresses osteoclastogenesis through the TIM-3 receptor and promotes osteoblast differentiation (31,32), thereby contributing to balanced bone turnover. Galectin-9 also restrains the NLRP3 inflammasome by binding to NLRP3 and inducing its p62-dependent autophagic degradation (33), a process consistent with the enrichment of innate immune and cytosolic DNA-sensing pathways identified in this study. Elevated Galectin-9 levels have further been reported as a biomarker of frailty (34), reinforcing its potential involvement in systemic processes coupling bone and muscle loss. The TERF2IP gene encodes Rap1 (Repressor/Activator Protein-1), a shelterin complex component that safeguards telomeric DNA integrity (35). In addition to its structural telomere role, Rap1 acts as an adaptor within the IκB kinase (IKK) complex, amplifying canonical NF-κB signaling and driving pro-inflammatory gene expression (36). Phosphorylation-dependent nuclear export of Rap1, or telomere dysfunction, triggers a DNA-damage response and cellular senescence, accompanied by a senescence-associated secretory phenotype (SASP) enriched in cytokines such as IL-6 and TNF-α (37). These cytokines are well-known mediators of osteoclast activation and muscle proteolysis, providing a plausible mechanistic bridge between telomere stress, inflammation, and osteosarcopenic remodeling. ZBTB7A encodes the leukemia/lymphoma-related factor (LRF, also known as Pokémon), a zinc-finger transcription factor that orchestrates lineage-specific differentiation (38). Within the osteoclast lineage (originally cloned as OCZF), ZBTB7A exerts stage-specific effects: repressing early RANKL-driven commitment by limiting NFATc1 induction, while in mature osteoclasts supporting resorptive activity and cell survival (38,39). In skeletal muscle, in-vitro studies indicate that suppression of ZBTB7A by miR-182-5p facilitates myogenic differentiation (40), implying a context-dependent regulatory function relevant to both tissue types. Finally, RPS3 encodes ribosomal protein S3, a component of the 40S small ribosomal subunit with multiple extra-ribosomal roles, including regulation of apoptosis, DNA repair, and transcription (41). RPS3 also acts as a noncanonical subunit of the NF-κB complex, directly modulating the transcription of specific target genes (42,43). As NF-κB activity governs both osteoclastogenesis and catabolic inflammation in aging muscle, RPS3 may represent a nodal element linking these processes. Moreover, ribosomal biogenesis is a key determinant of protein synthesis capacity; its impairment in aging muscle contributes to anabolic resistance, although data on RPS3 expression in sarcopenic tissue remain limited (44). Conversely, two genes already linked to osteosarcopenia and nine additional genes associated with both osteoporosis and sarcopenia provide internal validation, demonstrating that our approach successfully recovers established disease biology. Importantly, the presence of prior evidence does not reduce their relevance within the combined osteosarcopenic phenotype but rather highlights convergent mechanisms across the bone–muscle axis. Among these, TNFSF11 is the most translationally advanced example. Its encoded ligand, RANKL, is a key driver of osteoclast differentiation and the established pharmacological target of denosumab therapy. While several small and heterogeneous clinical studies have suggested potential improvements in muscle performance following denosumab treatment (45), a recent randomized controlled trial reported no significant difference in muscle health between denosumab and placebo groups (46). Conversely, STAT3 emerges from transcriptomic analyses as a shared molecular hub for both postmenopausal osteoporosis and sarcopenia (25). Although this points to STAT3 as a promising therapeutic target for dual bone–muscle interventions, experimental validation across both tissue systems remains limited. Two other genes, FOXO3 and AGER , though less explored in the combined phenotype, have been independently implicated in each disorder. FOXO3 is a central regulator of cellular aging and oxidative stress (47). In skeletal muscle, it acts downstream of the IGF-1/PI3K/AKT pathway (48), where nuclear accumulation of FOXO3 correlates with muscle atrophy and sarcopenic phenotype (49). In bone, reduced FOXO3 expression is characteristic of osteoporotic patients, and FOXO3 deficiency exacerbates oxidative damage and bone loss in aged mice (50). AGER , encoding the receptor for advanced glycation end products (RAGE), is similarly positioned at the intersection of inflammation, metabolism, and aging (51). Activation of RAGE by AGEs stimulates NF-κB signaling, thereby promoting catabolic processes and muscle cell loss (51,52). Osteoporosis has also been proposed to partially mediate the association between circulating AGEs and sarcopenia (51). In bone, RAGE signaling suppresses osteoblast proliferation and enhances osteoclastogenesis, driving the osteoporotic phenotype (53). Collectively, these established genes underscore the validity of our analytical framework and its ability to rediscover biologically confirmed mediators while positioning them within the broader osteosarcopenic context. KEGG pathway enrichment revealed a convergence of both established and emerging biological mechanisms underlying osteosarcopenia. The overrepresentation of NF-κB, TNF, and osteoclast-differentiation pathways aligns with well-characterized processes in bone remodeling, where RANKL–RANK signaling serves as a central therapeutic axis (54,55). These same pathways are mechanistically linked to sarcopenia through TNF/NF-κB–mediated muscle proteolysis (56–58) and elevated inflammatory signaling observed in frail and low–muscle-mass older adults (59). Collectively, these findings reinforce inflammation-driven molecular crosstalk as a shared denominator in bone and muscle wasting. Notably, necroptosis, a regulated form of cell death mediated primarily by receptor-interacting protein kinases RIPK1 and RIPK3, together with the pseudokinase MLKL (60), emerged as an additional enriched pathway. Necroptosis has been identified as a promising therapeutic target for the prevention of osteoporosis, with preclinical data demonstrating its modulation across osteoclasts, osteoblasts, and osteocytes (60,61). Parallel evidence from Duchenne muscular dystrophy and inflammatory myopathies shows that genetic ablation of RIPK3/MLKL or pharmacological inhibition of RIPK1 reduces myofiber necrosis and improves muscle performance in murine models (62). These converging observations suggest that regulated necrosis may represent a mechanistic bridge between bone and muscle pathology. Another intriguing finding was the enrichment of the cGAS–STING DNA-sensing pathway, recently recognized as a key regulator of sterile inflammation in musculoskeletal tissues (63). Cytoplasmic accumulation of damaged DNA triggers type I interferon and proinflammatory cytokine cascades, linking DNA damage responses with chronic inflammation and tissue remodeling in both bone and muscle. This supports the hypothesis that impaired cellular quality control and innate immune activation jointly contribute to osteosarcopenic decline. Although several infection-related KEGG terms (e.g., Malaria , Shigellosis , Tuberculosis , Legionellosis ) also surpassed the enrichment threshold, these associations were primarily driven by broad innate immunity genes, including IL1B , IL18 , MYD88 , PYCARD , RIPK1 , TRAF2 , and NOD2 , reflecting generalized inflammasome and NF-κB signaling rather than pathogen-specific mechanisms. Their co-enrichment further underscores the central role of innate immune pathways in the systemic regulation of musculoskeletal homeostasis. A major strength of GOReverseLookup lies in its ability to let researchers explicitly define states of interest by curating GO terms that reflect disease-relevant mechanisms, thereby producing transparent, auditable, and biologically interpretable gene lists. In contrast to black-box artificial intelligence models, GOReverseLookup offers full methodological transparency. Users can inspect the queried processes, inclusion criteria, and resulting statistical associations. The inclusion of ortholog-aware querying (via Ensembl and UniProtKB) across more than 240 species broadens the discovery space and helps surface evolutionarily conserved or understudied candidate genes while preserving reproducibility. Importantly, the tool performs particularly well for composite or multifactorial phenotypes, such as osteosarcopenia, where intersecting GO term sets capture biologically meaningful overlaps between related conditions. Nevertheless, several methodological limitations should be acknowledged. The robustness of GOReverseLookup outcomes critically depends on the careful selection and curation of GO terms, as well as on the chosen analytical parameters. Suboptimal input curation may yield biologically irrelevant or misleading associations. While cross-species ortholog integration expands candidate coverage, it can also increase the risk of false positives. Furthermore, the accuracy of the results is inherently constrained by the completeness of existing GO annotations, insufficient or outdated annotation data may lead to false negatives. The availability of tissue-specific expression datasets also presents a limitation, particularly for bone (23), in contrast to the richer datasets available for muscle. In conclusion, this study presents the first disease-focused application of GOReverseLookup, a recently developed, non-machine-learning bioinformatics tool, to systematically identify genes implicated in the pathogenesis of osteosarcopenia. The discovery of established mediators such as TNFSF11 and STAT3 validates the robustness of the approach, while the identification of 14 previously understudied genes ( RBCK1, TERF2IP, LGALS9, ZBTB7A, RPS3, PYCARD, MALT1, BCL2L11, PRKCQ, CARD9, IL18R1, ANKRD42, BTK , and NOD2 ), five of which exhibit at least moderate expression in both bone and muscle, reveals novel candidates for further exploration. KEGG enrichment analysis highlighted both established and emerging pathways, including regulated necrosis and cytosolic DNA sensing, emphasizing the interplay of inflammatory and stress-response processes in osteosarcopenia. Collectively, these findings demonstrate that GOReverseLookup can effectively uncover biologically plausible gene candidates in complex, multi-tissue syndromes. By integrating GOReverseLookup with tissue-expression stratification, KEGG pathway mapping, and DisGeNET-based gene–disease association scoring, we provide a reproducible and transparent framework for gene discovery that can be extended to other multifactorial conditions beyond osteosarcopenia. The identification of shared pathways provides mechanistic targets for future studies aimed at modulating bone–muscle crosstalk and mitigating osteosarcopenic decline. Abbreviations ASC Apoptosis-associated speck-like protein containing a CARD BH Benjamini–Hochberg procedure Bim Bcl-2 interacting mediator of cell death BMP Bone morphogenetic protein CARD Caspase activation and recruitment domain CFG Convergent functional genomics cpdm Chronic proliferative dermatitis FDR False discovery rate FE Fold enrichment FGF-2 Fibroblast growth factor 2 GDA Gene-disease association GO Gene Ontology GWAS Genome-wide association study IEA Inferred from electronic annotation IKK IκB kinase complex KEGG Kyoto Encyclopedia of Genes and Genomes LRF Leukemia/lymphoma-related factor LUBAC Linear ubiquitin chain assembly complex MALT1 Mucosa-associated lymphoid tissue lymphoma translocation protein 1 MLKL Mixed-lineage kinase domain-like pseudokinase NFAT Nuclear factor of activated T cells NF-κB Nuclear factor kappa-light-chain-enhancer of activated B cells NLRP3 NLR family pyrin domain containing 3 OPG Osteoprotegerin PGBM1 RBCK1-associated polyglucosan body myopathy-1 PKC-θ Protein kinase C theta RANK Receptor activator of NF-κB (encoded by TNFRSF11A) RANKL Receptor activator of NF-κB ligand (encoded by TNFSF11) RBR RING-between-RING (E3 ubiquitin ligase family) RIPK Receptor-interacting protein kinase SASP Senescence-associated secretory phenotype SOI State of interest STAT3 Signal transducer and activator of transcription 3 TPM Transcripts per million TWAS Transcriptome-wide association study Wnt Wingless/Int-1 signaling pathway Declarations Competing interests The authors declare that they have no competing interests. Author contributions statement S.Ž. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. L.V. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. A.Š. contributed to conceptualization, methodology, investigation and edited the manuscript. V.S. contributed to conceptualization, methodology and reviewed and edited the manuscript. N.U. contributed to conceptualization, investigation, project administration, supervision, funding acquisition and reviewed and edited the manuscript. All authors read and approved the final version of the manuscript. Funding This research was funded by Slovenian Research and Innovation Agency (Grant No. J3- 4527 and P3-0043) and Foundation of ing. Lenarčič Milan Scholarship. Author Contribution S.Ž. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. L.V. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. A.Š. contributed to conceptualization, methodology, investigation and edited the manuscript. V.S. contributed to conceptualization, methodology and reviewed and edited the manuscript. N.U. contributed to conceptualization, investigation, project administration, supervision, funding acquisition and reviewed and edited the manuscript. All authors read and approved the final version of the manuscript. Acknowledgement The authors would like to thank Dr. Chiedozie K. Ugwoke for manuscript proofreading. Data Availability All data supporting the conclusions of this article are provided in the main text and the Supplementary Information. References Kirk, B., Zanker, J. & Duque, G. Osteosarcopenia: epidemiology, diagnosis and treatment — facts and numbers. J. Cachexia Sarcopenia Muscle . 11 , 609 (2020). Huang, T. et al. Prevalence and risk factors of osteosarcopenia: a systematic review and meta-analysis. BMC Geriatr. 23 , 369 (2023). Compston, J. E., McClung, M. R., Leslie, W. D. & Osteoporosis Lancet 393 , 364–376 (2019). Cruz-Gentoft, A. J. et al. Sarcopenia: European consensus on definition and diagnosis. Age Ageing . 39 , 412–423 (2010). Clynes, M. A., Gregson, C. L., Bruyère, O., Cooper, C. & Dennison, E. M. Osteosarcopenia: where osteoporosis and sarcopenia collide. Rheumatol. (Oxford) . 60 , 529–537 (2021). Noh, J. Y., Yang, Y. & Jung, H. Molecular mechanisms and emerging therapeutics for osteoporosis. Int. J. Mol. Sci. 21 , 7623 (2020). Chen, Q., Lou, H. & Peng, C. Recent advances in the genetic association between osteoporotic fracture and sarcopenia. Aging Pathobiol Ther. 3 , 2–9 (2021). Cho, M. R., Lee, S. & Song, S. K. A review of sarcopenia pathophysiology, diagnosis, treatment and future direction. J. Korean Med. Sci. 37 , e146 (2022). Fagundes Belchior, G., Kirk, B., Pereira da Silva, E. A. & Duque, G. Osteosarcopenia: beyond age-related muscle and bone loss. Eur. Geriatr. Med. 11 , 715–724 (2020). Michaud, M. et al. Proinflammatory cytokines, aging and age-related diseases. J. Am. Med. Dir. Assoc. 14 , 877–882 (2013). Zhao, Z., Yan, K., Guan, Q., Guo, Q. & Zhao, C. Mechanism and physical activities in bone–skeletal muscle crosstalk. Front. Endocrinol. 14 , 1287972 (2024). Polesskaya, A., Seale, P. & Rudnicki, M. A. Wnt signaling induces the myogenic specification of resident CD45⁺ adult stem cells during muscle regeneration. Cell 113 , 841–852 (2003). Manolagas, S. C. From estrogen-centric to aging and oxidative stress: a revised perspective of the pathogenesis of osteoporosis. Endocr. Rev. 31 , 266–300 (2010). Sheng, R. et al. Muscle–bone crosstalk via endocrine signals and potential targets for osteosarcopenia-related fracture. J. Orthop. Transl . 43 , 36–46 (2023). Manolio, T. A. et al. Finding the missing heritability of complex diseases. Nature 461 , 747–753 (2009). Wainberg, M. et al. Opportunities and challenges for transcriptome-wide association studies. Nat. Genet. 51 , 592–599 (2019). Denny, J. C., Bastarache, L. & Roden, D. M. Phenome-wide association studies as a tool to advance precision medicine. Annu. Rev. Genomics Hum. Genet. 17 , 353–373 (2016). Niculescu, A. B. & Le-Niculescu, H. Convergent functional genomics: what we have learned and can learn about genes, pathways and mechanisms. Neuropsychopharmacology 35 , 355–356 (2010). Smaili, F. Z., Gao, X. & Hoehndorf, R. Onto2Vec: joint vector-based representation of biological entities and their ontology-based annotations. Bioinformatics 34 , i52–i60 (2018). Alshahrani, M. & Hoehndorf, R. Semantic disease gene embeddings (SmuDGE): phenotype-based disease gene prioritization without phenotypes. Bioinformatics 34 , i901–i907 (2018). Škorjanc, A., Smrkolj, V. & Umek, N. GOReverseLookup: a gene ontology reverse lookup tool. Comput. Biol. Med. 191 , 110185 (2025). Tables Table 1. Categorization of genes based on prior literature evidence and expression levels. Previously linked to both osteoporosis and sarcopenia No prior links; expressed in bone and muscle (≥ LOW) Other significant genes TNFSF11 RBCK1 * NOD1 * STAT3 * TERF2IP * PLCG2 IL1B LGALS9 * RIPK1 AGER ZBTB7A * RTKN2 CLU * RPS3 * RIPK2 * SPHK1 PYCARD TNFRSF11A IL18 MALT1 MYD88 TP53 BCL2L11 SOSTDC1 FOXO3 * PRKCQ EIF2AK2 CARD9 CD40LG IL18R1 AIM2 ANKRD42 TRAF2 BTK RIPK4 NOD2 CD36 Table note : Genes are stratified by (i) existing literature support for concomitant involvement in osteoporosis and sarcopenia, (ii) absence of prior disease association plus expression in both bone and muscle (low or medium), and (iii) all other significant genes. Genes in bold denote those previously linked to osteosarcopenia, whereas genes marked with an asterisk (*) indicate at least medium tissue expression in both tissues. Table 2. KEGG pathways significantly enriched among the 37 osteosarcopenia-associated genes. Pathways FE FDR Genes NF-kappa B signaling pathway 57.9 4.9×10⁻ 14 MALT1, IL1B, MYD88, PLCG2, PRKCQ, BTK, TRAF2, TNFSF11, RIPK1, TNFRSF11A Cytosolic DNA-sensing pathway 47.8 7.6×10⁻ 7 PYCARD, IL1B, IL18, RIPK1, AIM2 NOD-like receptor signaling pathway 39.5 7.7×10⁻ 15 NOD1, RBCK1, PYCARD, IL1B, IL18, MYD88, NOD2, CARD9, TRAF2, RIPK1, RIPK2, AIM2 Inflammatory bowel disease 37.1 3.1×10⁻ 5 IL1B, IL18, NOD2, STAT3 Prolactin signaling pathway 34.4 3.9×10⁻ 5 FOXO3, STAT3, TNFSF11, TNFRSF11A C-type lectin receptor signaling pathway 28.9 8.7×10⁻ 6 MALT1, PYCARD, IL1B, PLCG2, CARD9 Osteoclast differentiation 28.7 7.6×10⁻ 7 IL1B, PLCG2, BTK, TRAF2, TNFSF11, TNFRSF11A TNF signaling pathway 26.9 1.2×10⁻ 5 IL1B, NOD2, TRAF2, RIPK1, IL18R1 Necroptosis 26.5 1.3×10⁻ 7 RBCK1, PYCARD, IL1B, EIF2AK2, STAT3, TRAF2, RIPK1 Lipid and atherosclerosis 25.3 1.7×10⁻ 9 PYCARD, IL1B, IL18, MYD88, STAT3, TP53, TRAF2, CD36, CD40LG Table note : FE, fold enrichment; FDR, false discovery rate. Pathways were identified with ShinyGO v0.82 using all human protein-coding genes in KEGG as background. Results are ordered by FE and limited to pathways with FE > 25 and a Benjamini–Hochberg false-discovery rate (FDR) < 5 × 10⁻⁵. Infection-related pathways that also met the significance threshold (e.g. Malaria , Shigellosis , Tuberculosis ) are presented in Supplementary Information 3. Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformation1.xlsx SupplementaryInformation2.docx SupplementaryInformation3.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8000419","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":539058940,"identity":"17628029-624d-43ad-8145-4bf5035a625d","order_by":0,"name":"Simon Žetko","email":"","orcid":"","institution":"University of Ljubljana","correspondingAuthor":false,"prefix":"","firstName":"Simon","middleName":"","lastName":"Žetko","suffix":""},{"id":539058941,"identity":"f1af4352-2bfc-41df-af93-7a5b0db2c64d","order_by":1,"name":"Lan Vukolić","email":"","orcid":"","institution":"University of Ljubljana","correspondingAuthor":false,"prefix":"","firstName":"Lan","middleName":"","lastName":"Vukolić","suffix":""},{"id":539058942,"identity":"307c073e-cdff-4f8c-838c-b87b12e406d1","order_by":2,"name":"Aljoša Škorjanc","email":"","orcid":"","institution":"University of Ljubljana","correspondingAuthor":false,"prefix":"","firstName":"Aljoša","middleName":"","lastName":"Škorjanc","suffix":""},{"id":539058943,"identity":"6a9a51f4-cdde-45d9-97b2-934e133e51fd","order_by":3,"name":"Vladimir Smrkolj","email":"","orcid":"","institution":"University of Ljubljana","correspondingAuthor":false,"prefix":"","firstName":"Vladimir","middleName":"","lastName":"Smrkolj","suffix":""},{"id":539058944,"identity":"56e444b3-2eac-47d3-bdc6-8c41ebd9a520","order_by":4,"name":"Nejc Umek","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIie2RoQoCQRCGZzFsWfvKib7CiUFMvsocgk0wWb106R5Ai+9g0Xqw4cppFk7wLBbLxgsGx1UxiKtRcL+wCzN8zD8MgMPxm6B5JQ9B0y+AXyv+N4pI2NQolc/KDSnRKECKnQ4PD5pFy0ZtdgpCFu3qpBSsHL1XunHSlyzK2543XJByNMEqwhLM3+IASAnm3nBVsEiJniLFtou/L4wymdfW1ynKTGGldQoYBT1ZfSpgDZZhX+Imb81i2gU3N0VZlTQLtB7nTZlSMD1WPeBxcCjP7xW6Hd6PCY+rCkxsAgB/6b9WHA6H47+5AL3xTimCGQkVAAAAAElFTkSuQmCC","orcid":"","institution":"University of Ljubljana","correspondingAuthor":true,"prefix":"","firstName":"Nejc","middleName":"","lastName":"Umek","suffix":""}],"badges":[],"createdAt":"2025-10-31 16:53:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8000419/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8000419/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95103673,"identity":"83c4480f-0b0b-4a00-91b2-52cdd7bf7422","added_by":"auto","created_at":"2025-11-04 10:24:38","extension":"png","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":838145,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/4dde397624353983df28055c.png"},{"id":95103769,"identity":"8cf4f275-e5f5-436d-aa9e-9e3853dace77","added_by":"auto","created_at":"2025-11-04 10:24:50","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":132469,"visible":true,"origin":"","legend":"","description":"","filename":"ZetkoMS.docx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/90dccdcdbba5b812e59de5bf.docx"},{"id":95103688,"identity":"b06ca034-9aba-4d9b-b3a8-31d7d2dc4692","added_by":"auto","created_at":"2025-11-04 10:24:42","extension":"png","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5059073,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/5c7c79b9caf08c0e0fea9ccb.png"},{"id":95103724,"identity":"8c421e7e-991b-4779-894e-0fbe8a843084","added_by":"auto","created_at":"2025-11-04 10:24:48","extension":"json","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6946,"visible":true,"origin":"","legend":"","description":"","filename":"b2e0c0f8af4442c486a17d1ac6399ed0.json","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/22837724846cc5c50b713c76.json"},{"id":95103655,"identity":"683c2d49-70a4-4c67-8dd2-dc9fc55db52e","added_by":"auto","created_at":"2025-11-04 10:24:33","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":24242,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/3279040d490ddc03e93b6d7a.xlsx"},{"id":95103686,"identity":"0111e22b-792c-4073-b751-8d275fefec01","added_by":"auto","created_at":"2025-11-04 10:24:41","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":59615,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/7588ebb84e5c83346d2b74b4.docx"},{"id":95103662,"identity":"ace40a32-4ffc-4106-b60d-b7bd6a57e15e","added_by":"auto","created_at":"2025-11-04 10:24:35","extension":"docx","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21687,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation3.docx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/11452a7dfdf9ebe445b894c2.docx"},{"id":95103732,"identity":"85d2357c-bd12-45b4-9e1e-79feb94dc9f1","added_by":"auto","created_at":"2025-11-04 10:24:48","extension":"xml","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87591,"visible":true,"origin":"","legend":"","description":"","filename":"b2e0c0f8af4442c486a17d1ac6399ed01enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/3d0ddc08bc6ba02e552da3c6.xml"},{"id":95103702,"identity":"9c7a3f3e-9ad6-4d20-8eec-c12e828bef12","added_by":"auto","created_at":"2025-11-04 10:24:45","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":838145,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/31198c4ec5f49aaeac16073f.png"},{"id":95103680,"identity":"8823a143-e492-49ad-953d-f07529b88c89","added_by":"auto","created_at":"2025-11-04 10:24:40","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5059073,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/6504af0bcc62e4141514d47f.png"},{"id":95103668,"identity":"d21b0365-8b5d-410b-b8c1-1b291b456a90","added_by":"auto","created_at":"2025-11-04 10:24:37","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":268614,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/9188f47dc526b5a0c9daa5b2.png"},{"id":95103641,"identity":"894991fe-04e1-4f74-b8d6-e57800183626","added_by":"auto","created_at":"2025-11-04 10:24:31","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":517222,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/0c4b1cdfa83cc85d6c95e4c8.png"},{"id":95103659,"identity":"67feadf6-11f2-4fab-873a-d88ea2a53ed3","added_by":"auto","created_at":"2025-11-04 10:24:34","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":86808,"visible":true,"origin":"","legend":"","description":"","filename":"b2e0c0f8af4442c486a17d1ac6399ed01structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/d432167ee155576888974b80.xml"},{"id":95103712,"identity":"a8cb3a80-3035-488d-bb3e-53b508ed92cc","added_by":"auto","created_at":"2025-11-04 10:24:46","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":96029,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/311dff978afe5b5922380aa3.html"},{"id":95103657,"identity":"41a008c8-de1c-450c-a30b-3da4546e7e3f","added_by":"auto","created_at":"2025-11-04 10:24:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":838145,"visible":true,"origin":"","legend":"\u003cp\u003eMethods overview. A curated list of Gene Ontology (GO) terms for the states of interest (osteoporosis and sarcopenia) was used to run the GOReverseLookup analysis. Genes identified by this analysis were systematically characterized through (1) literature validation and (4) disease association (DisGeNET GDA scores), (2) tissue-specific expression profiling (Expression Atlas), and (3) pathway enrichment analysis (KEGG via ShinyGO). Created with BioRender.com.\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/21063d98948277418b531d7e.png"},{"id":95103716,"identity":"394a85fa-7bf0-480f-ae96-a65ce875d84a","added_by":"auto","created_at":"2025-11-04 10:24:47","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":5059073,"visible":true,"origin":"","legend":"\u003cp\u003e37 candidate genes categorized by prior literature association and expression levels in disease-relevant tissues. The y-axis denotes literature evidence: None (no prior association), O or S (associated with osteoporosis or sarcopenia), and O and S (associated with both conditions). The x-axis stratifies expression levels in bone and muscle into three groups: Below cutoff, Low, and Medium. Created with BioRender.com.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/b2c501a06be49004b7db9863.png"},{"id":95822450,"identity":"8a794dd5-c198-4d23-89d0-662a215cc71d","added_by":"auto","created_at":"2025-11-13 10:49:15","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6156045,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/d792cd7a-2b68-42d0-935d-43975029bd20.pdf"},{"id":95103739,"identity":"87b865b8-5043-4030-aefd-ead338ffcf54","added_by":"auto","created_at":"2025-11-04 10:24:49","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":24242,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/adb0041efa810cb6314261f3.xlsx"},{"id":95103671,"identity":"233414cb-19a3-4d2e-a9a4-17a4ad023c2e","added_by":"auto","created_at":"2025-11-04 10:24:38","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":59615,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation2.docx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/e6a0bc12e2e8c2b34721328c.docx"},{"id":95103734,"identity":"ea608ae8-1d6a-4695-b48e-499a86bab48f","added_by":"auto","created_at":"2025-11-04 10:24:49","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":21687,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation3.docx","url":"https://assets-eu.researchsquare.com/files/rs-8000419/v1/b5452eef605ea41a4a34b627.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Bridging muscle and bone: shared genetic signatures of osteosarcopenia revealed by reverse Gene Ontology lookup","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOsteosarcopenia is an emerging geriatric syndrome and an increasingly recognized global health concern. A recent meta-analysis estimated its pooled prevalence at approximately 21% (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). The condition reflects the coexistence of two age-related disorders, osteoporosis and sarcopenia, each contributing to frailty and fracture risk. Osteoporosis is characterized by reduced bone mass and microarchitectural deterioration, resulting in skeletal fragility and a heightened susceptibility to fractures (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e). Sarcopenia, by contrast, involves the progressive loss of skeletal muscle mass and strength, which underlies physical disability, diminished quality of life, and increased mortality (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e). Although traditionally studied as distinct entities, accumulating evidence indicates substantial overlap in their pathogenesis and risk factors including aging, hormonal decline, nutritional deficiency, and sedentary behavior (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e). The concurrent manifestation of both conditions synergistically exacerbates the risk of falls, fractures, and long-term disability (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eAt the molecular level, osteoporosis and sarcopenia arise from distinct yet parallel disruptions of tissue homeostasis. In osteoporosis, bone resorption exceeds bone formation, leading to a progressive decline in bone mass and an increased risk of fracture (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e). The balance between osteoclast-mediated resorption and osteoblast-driven formation is governed by several key signalling axes, including the receptor activator of nuclear factor κB ligand/receptor activator of nuclear factor κB/osteoprotegerin (RANKL/RANK/OPG) system, Wnt/β-catenin, and transforming growth factor-β/bone morphogenetic protein (TGF-β/BMP) pathways (\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e). Additional factors, such as estrogen deficiency in postmenopausal women and chronic inflammation, further exacerbate the imbalance in bone remodeling, contributing to a loss in bone density (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e). In parallel, sarcopenia is driven by impaired muscle protein synthesis and enhanced proteolytic activity. Age-related changes in muscle architecture, including the selective atrophy of type II fibers and reduced satellite cell regenerative capacity, further compromise muscle maintenance and repair (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThe pathological overlap between osteoporosis and sarcopenia arises from the interdependent signaling networks that link bone and muscle homeostasis. Both tissues originate from mesenchymal stem cells and remain closely interlinked through paracrine, endocrine and mechanical stimuli (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e). Myokines, such as myostatin, insulin-like growth factor 1 (IGF-1), irisin, and fibroblast growth factor 2 (FGF-2) have been implicated in bone remodeling (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e), whereas osteokines including osteocalcin and TGF-β influence muscle metabolism and contractile function (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e). The Wnt signalling pathway, essential for bone formation, also plays a pivotal role in muscle regeneration (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e). Chronic low-grade inflammation contributes to the atrophy of both tissues, as pro-inflammatory signals accelerate osteoclastogenesis and proteolysis (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e). Oxidative stress and hormonal dysregulation further amplify these pathological processes (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e). The reciprocal communication between muscle and bone appears essential for maintaining musculoskeletal function, disruptions in this signaling can contribute to osteosarcopenia. These insights underscore the need for therapeutic strategies that target skeletal fragility and muscle wasting simultaneously rather than in isolation (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIdentifying genes that are shared drivers of osteosarcopenia is essential for the development of targeted therapies. Achieving this goal typically relies on computational approaches, each of which introduces its own methodological biases and technical limitations. Existing computational pipelines for gene discovery span variant-centric, expression-centric and ontology-centric paradigms. Genome-wide association studies (GWAS) have identified thousands of loci associated with complex traits, however, most effect sizes are modest, and replication demands very large cohorts (\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e). Transcriptome-wide association studies (TWAS) improve functional inference but can inflate false-positive rates when genetic confounding is not properly accounted (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e). Phenome-wide association studies (PheWAS) invert the GWAS logic by scanning across clinical phenotypes, yet remain susceptible to phenotype misclassification and a substantial multiple-testing burden (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e). Convergent functional genomics (CFG) and related evidence integration frameworks aim to prioritize loci by aggregating heterogeneous data, but their outputs can be dominated by well-studied pathways and literature bias (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e). Over the past five years, representation-learning approaches including hypergraph embeddings, protein-interaction graph neural networks and ontology vectorizations such as Onto2Vec and SmuDGE, have captured latent structure across GO, OMIM and HPO to predict disease genes, yet at the cost of \u0026ldquo;black-box\u0026rdquo; interpretability and continued reliance on existing annotations (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e). Collectively, these methods either lack interpretability, demand extremely large datasets, or over-prioritize well-trodden pathways, leaving emerging, biologically plausible genes unexplored. A recently developed tool GOReverseLookup provides a complementary route. Instead of starting from variants or embeddings, researchers curate GO terms that define a phenotype or broader state of interest and the tool then evaluates which genes are statistically over- or under-represented in that user-defined state.\u003c/p\u003e\u003cp\u003eThis work presents the first disease-focused application of GOReverseLookup, demonstrating its utility as a broadly applicable framework for gene discovery that generates transparent, biologically contextual, and interpretable candidate lists well suited to early-stage hypothesis generation (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). Accordingly, the aim of this study was to discover new genes implicated in pathogenesis of both osteoporosis and sarcopenia.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eIdentification of candidate genes using the GOReverseLookup tool\u003c/h2\u003e\u003cp\u003eCandidate genes were identified using GOReverseLookup, a computational tool that prioritizes genes based on their statistical association with a curated set of Gene Ontology (GO) terms that collectively define a biological phenotype. The tool's workflow consists of three main stages: input curation, annotation retrieval, and statistical analysis, with only the input curation performed by the researcher and all subsequent steps handled automatically by GOReverseLookup.\u003c/p\u003e\u003cp\u003eFirst, two states of interest (SOIs) were defined based on the research objective: \u0026ldquo;osteoporosis\u0026rdquo; and \u0026ldquo;sarcopenia.\u0026rdquo; For each SOI, relevant Gene Ontology (GO) terms were manually curated and manually annotated as either stimulatory or inhibitory with respect to the SOI\u0026rsquo;s pathology. Key analytical parameters were then specified, including the depth of indirect annotations, \u003cem\u003ep\u003c/em\u003e-value cutoff threshold, filtering by evidence codes, ortholog selection, and the choice of a multiple testing correction method.\u003c/p\u003e\u003cp\u003eNext, the tool retrieved all corresponding gene annotations from the GO database based on the selected GO terms and parameters. When cross-species analysis is specified, ortholog mappings (e.g., from Ensembl or gProfiler) are integrated, and the complete set of annotations is filtered according to the pre-selected evidence codes.\u003c/p\u003e\u003cp\u003eFinally, GOReverseLookup performed statistical analysis to quantify the association between each gene and the defined SOI. For every gene a 2 x 2 contingency table was constructed, and enrichment was assessed using Fisher\u0026rsquo;s exact test. The resulting \u003cem\u003ep\u003c/em\u003e-values were corrected for multiple comparisons (e.g. false discovery rate), and genes meeting the significance threshold were reported as a prioritized candidate list.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eCuration of GO Terms\u003c/h3\u003e\n\u003cp\u003eA curated list of GO terms representing the biological processes underlying osteoporosis and sarcopenia was generated following a systematic workflow adapted from Škorjanc et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e). The curation process began with an extensive literature review to delineate key pathophysiology for each condition. Each mechanistic group was then translated into a representative keyword (e.g., \"osteoclast,\" \"satellite cell,\" \"estrogen\") for targeted searches within the QuickGO database, restricted to the biological process sub-ontology. Relevant GO terms were manually compiled using QuickGO's basket feature, with a consistent preference for specific child terms (e.g., \"positive regulation of osteoclast proliferation\") to leverage the tool's ability to map to parent annotations. The resulting list was annotated and structured for analysis, then independently reviewed by two researchers experienced in bone and muscle biology, yielding a final set of 367 GO terms representing the major biological processes implicated in osteoporosis and sarcopenia. The complete and categorized list of 367 GO terms is provided in Supplementary Information 1.\u003c/p\u003e\n\u003ch3\u003eReverse Enrichment Analysis\u003c/h3\u003e\n\u003cp\u003eReverse enrichment analysis was performed using GOReverseLookup (v1.0.56; Python 3.11) on June 27, 2025. The analysis targeted \u003cem\u003eHomo sapiens\u003c/em\u003e, incorporating ortholog data from \u003cem\u003eDanio rerio\u003c/em\u003e, \u003cem\u003eRattus norvegicus\u003c/em\u003e, \u003cem\u003eMus musculus\u003c/em\u003e, and \u003cem\u003eXenopus tropicalis\u003c/em\u003e. Key parameters included the use of indirect annotations up to three parent levels and the exclusion of electronically inferred annotations (IEA). A stringent significance threshold was set at a \u003cem\u003ep\u003c/em\u003e-value\u0026thinsp;\u0026lt;\u0026thinsp;5 \u0026times; 10⁻⁸, and the resulting \u003cem\u003ep\u003c/em\u003e-values were adjusted for multiple testing using the Benjamini\u0026ndash;Hochberg false discovery rate (FDR-BH) procedure.\u003c/p\u003e\u003cp\u003eTo ensure reproducibility, all input files are made readily available at (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/GenomeDev/GOReverseLookup\u003c/span\u003e\u003cspan address=\"https://github.com/GenomeDev/GOReverseLookup\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). We used the Gene Ontology OBO (format-version 1.2; data-version releases/2025-03-16), the \u003cem\u003eHomo sapiens\u003c/em\u003e UniProt-GOA annotation set (GO Annotations File version 2.2; date-generated 2024-06-19), and GO metadata via the GO API (release 2025-06-26). Orthologs to human (NCBITaxon:9606) were mapped from mouse (MGI; 10090), rat (RGD; 10116), zebrafish (ZFIN; 7955), and tropical clawed frog (Xenbase; 8364).\u003c/p\u003e\n\u003ch3\u003eSystematic Characterization of Candidate Genes\u003c/h3\u003e\n\u003cp\u003eCandidate genes identified by GOReverseLookup were further characterized through a systematic three-step validation and functional annotation process.\u003c/p\u003e\u003cp\u003e1. Literature Validation and Disease Association\u003c/p\u003e\u003cp\u003eA comprehensive literature search was performed in PubMed and Google Scholar to identify existing evidence linking each gene to the diseases of interest. Search queries combined the gene symbol with the keywords \"osteoporosis,\" \"sarcopenia,\" and \"osteosarcopenia.\" In parallel, gene-disease associations were quantified using the DisGeNET database version 25.1.1. For each gene, we recorded the evidence-weighted Gene-Disease Association (GDA) score, which aggregates contributions from expert-curated human sources (0.30 each), animal-model databases (0.10 each) and PubMed text-mining hits (small, cumulative weights). Scores range from 0 to 1, with values greater than 0.30 indicating support from at least one expert-curated human source (22). The primary objective of this step was to determine whether a gene had been previously associated with either osteoporosis, sarcopenia, or their co-manifestation (osteosarcopenia) in the scientific literature or curated databases.\u003c/p\u003e\u003cp\u003e2. Tissue-Specific Expression\u003c/p\u003e\u003cp\u003eGene expression levels in disease-relevant human tissues were assessed using the Expression Atlas (EMBL-EBI). For each gene, the highest recorded expression level was extracted across muscle-related tissues (\"muscle of arm,\" \"muscle of leg,\" \"skeletal muscle,\" \"skeletal muscle of trunk,\" and \"skeletal muscle tissue\") and bone tissue (\"cervical vertebrae\"). Expression levels were categorized by transcripts per million (TPM) as high (\u0026gt;\u0026thinsp;1000), medium (\u0026gt;\u0026thinsp;11), low (\u0026gt;\u0026thinsp;0.5), or below cutoff (\u0026le;\u0026thinsp;0.5 TPM). It should be noted that bone tissue datasets in Expression Atlas are limited in diversity and coverage (23).\u003c/p\u003e\u003cp\u003e3. Pathway Enrichment Analysis\u003c/p\u003e\u003cp\u003eTo identify enriched biological pathways, KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment pathway analysis was performed using ShinyGO v0.82 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://bioinformatics.sdstate.edu/go/\u003c/span\u003e\u003cspan address=\"https://bioinformatics.sdstate.edu/go/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The background reference gene set was defined as all human protein-coding genes annotated in the KEGG database. Pathways were ranked according to fold enrichment and considered significantly enriched at a Benjamini\u0026ndash;Hochberg adjusted \u003cem\u003ep\u003c/em\u003e-value (FDR)\u0026thinsp;\u0026lt;\u0026thinsp;5 \u0026times; 10⁻\u003csup\u003e5\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eIdentified and Categorized Candidate Genes\u003c/h2\u003e\n\u003cp\u003eOur analysis identified 37 genes significantly associated with both osteoporosis and sarcopenia simultaneously (FDR \u0026lt; 5 \u0026times; 10⁻⁸). To prioritize these candidates for further follow-up analysis, we classified them according to prior literature evidence and tissue-specific expression patterns in bone and skeletal muscle (Table 1 and Figure 2). Comprehensive statistical outputs are provided in Supplementary Information 2.\u003c/p\u003e\n\u003cp\u003eThe analysis highlighted 14 genes with no previously reported association to either disease. Among these, five genes (\u003cem\u003eRBCK1, TERF2IP, LGALS9, ZBTB7A,\u0026nbsp;\u003c/em\u003eand\u003cem\u003e\u0026nbsp;RPS3\u003c/em\u003e) exhibited at least moderate (medium) expression\u0026nbsp;in both bone and muscle and no existing disease association score (GDA \u0026asymp; 0). The remaining nine genes (\u003cem\u003ePYCARD, MALT1, BCL2L11, PRKCQ,\u003c/em\u003e \u003cem\u003eCARD9, IL18R, ANKRD42, BTK, NOD2\u003c/em\u003e) displayed detectable but lower (low) expression\u0026nbsp;in both tissues.\u003c/p\u003e\n\u003cp\u003eIn total, nine genes \u003cem\u003e(TNFSF11, STAT3,\u0026nbsp;\u003c/em\u003e\u003cem\u003eIL1B, AGER, CLU, SPHK, IL18, TP53, FOXO3)\u003c/em\u003e were previously described in the literature as being associated with both osteoporosis and sarcopenia and have a non-zero GDA score for these conditions. Notably, two of these genes \u003cem\u003e(TNFSF11, STAT3)\u003c/em\u003e have been directly implicated in osteosarcopenia\u0026nbsp;(24,25).\u003c/p\u003e\n\u003cp\u003eThe remaining 14 significant genes showed either evidence linked to only one of the two conditions or expression levels below the defined cutoff expression level in one or both tissue types.\u003c/p\u003e\n\u003ch2\u003eKEGG pathway enrichment\u003c/h2\u003e\n\u003cp\u003eKEGG enrichment of the 37 osteosarcopenia-associated genes yielded 15 significantly overrepresented pathways that exceeded the predefined significance threshold (FDR \u0026lt; 5 \u0026times; 10⁻⁵) and showed a fold-enrichment (FE) higher than 25.\u0026nbsp; To emphasize biological mechanisms rather than incidental annotation overlap, only non-infectious pathways are presented in Table 2. The complete list, including infection-related pathways (e.g., \u003cem\u003eMalaria\u003c/em\u003e, \u003cem\u003eShigellosis\u003c/em\u003e, \u003cem\u003eTuberculosis\u003c/em\u003e), is provided in Supplementary Information 3. \u0026nbsp;Pathways are ranked by FE, the effect-size metric recommended by ShinyGO for assessing biological relevance. The most enriched was the NF-\u0026kappa;B signalling pathway (FE = 57.9), followed by the cytosolic DNA-sensing pathway (FE = 47.8) and the NOD-like-receptor signalling pathway (FE = 39.5). \u0026nbsp;Additional enriched pathways included prolactin signalling (FE = 34.4), C-type lectin receptor signalling (FE = 28.9), osteoclast differentiation (FE = 28.7), TNF signalling (FE = 26.9), necroptosis (FE = 26.5) and lipid and atherosclerosis (FE = 25.3).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study represents the first integrative application of reverse Gene Ontology analysis to a complex musculoskeletal phenotype. Our analysis identified 37 genes associated with osteosarcopenia, systematically categorized by prior disease associations, expression across bone and skeletal muscle, and their pathway context derived from KEGG enrichment. This integrative framework recovered well-established mediators, serving as an internal validation of the approach, while simultaneously uncovering underexplored candidates that may reveal novel mechanisms underlying the coordinated regulation of bone and muscle homeostasis. Incorporating tissue-specific expression further refines gene prioritization, as genes expressed in both bone and muscle are more likely to influence shared metabolic and inflammatory signaling axes that maintain musculoskeletal integrity. However, genes with relatively low expression yet high functional relevance, such as \u003cem\u003eTNFSF11\u003c/em\u003e and \u003cem\u003eIL1B\u003c/em\u003e, emphasize the importance of interpreting expression data in a mechanistic context rather than relying on quantitative thresholds alone.\u003c/p\u003e\u003cp\u003eGenes without prior association to either osteoporosis or sarcopenia represent particularly valuable, underexplored candidates that may illuminate biological mechanisms not yet captured by current literature. Five such genes, \u003cem\u003eRBCK1, TERF2IP, LGALS9, ZBTB7A\u003c/em\u003e, and \u003cem\u003eRPS3\u003c/em\u003e, showed moderate expression in both bone and skeletal muscle, suggesting a possible contribution to the shared molecular and inflammatory networks underpinning osteosarcopenia.\u003c/p\u003e\u003cp\u003e\u003cem\u003eRBCK1\u003c/em\u003e encodes HOIL-1, a catalytic component of the linear ubiquitin chain assembly complex (LUBAC), one of the three subunits responsible for M-1\u0026ndash;linked ubiquitination (26). LUBAC regulates multiple immune signaling cascades, particularly through activation of the NF-κB pathway (27), a mechanism consistently enriched in our KEGG analysis. Beyond inflammation, HOIL-1 also modulates apoptosis and necroptosis (28), two processes essential for tissue remodeling in both muscle and bone. Mutations in \u003cem\u003eRBCK1\u003c/em\u003e cause RBCK1-associated polyglucosan body myopathy-1 (PGBM1), a glycogen storage disorder that manifests with skeletal muscle weakness and cardiomyopathy (29), further supporting its potential relevance to musculoskeletal pathology.\u003c/p\u003e\u003cp\u003eSimilarly, \u003cem\u003eLGALS9\u003c/em\u003e encodes Galectin-9, a β-galactoside-binding lectin with broad immunomodulatory functions (30). In bone, Galectin-9 suppresses osteoclastogenesis through the TIM-3 receptor and promotes osteoblast differentiation (31,32), thereby contributing to balanced bone turnover. Galectin-9 also restrains the NLRP3 inflammasome by binding to NLRP3 and inducing its p62-dependent autophagic degradation (33), a process consistent with the enrichment of innate immune and cytosolic DNA-sensing pathways identified in this study. Elevated Galectin-9 levels have further been reported as a biomarker of frailty (34), reinforcing its potential involvement in systemic processes coupling bone and muscle loss.\u003c/p\u003e\u003cp\u003eThe \u003cem\u003eTERF2IP\u003c/em\u003e gene encodes Rap1 (Repressor/Activator Protein-1), a shelterin complex component that safeguards telomeric DNA integrity (35). In addition to its structural telomere role, Rap1 acts as an adaptor within the IκB kinase (IKK) complex, amplifying canonical NF-κB signaling and driving pro-inflammatory gene expression (36). Phosphorylation-dependent nuclear export of Rap1, or telomere dysfunction, triggers a DNA-damage response and cellular senescence, accompanied by a senescence-associated secretory phenotype (SASP) enriched in cytokines such as IL-6 and TNF-α (37). These cytokines are well-known mediators of osteoclast activation and muscle proteolysis, providing a plausible mechanistic bridge between telomere stress, inflammation, and osteosarcopenic remodeling.\u003c/p\u003e\u003cp\u003e\u003cem\u003eZBTB7A\u003c/em\u003e encodes the leukemia/lymphoma-related factor (LRF, also known as Pok\u0026eacute;mon), a zinc-finger transcription factor that orchestrates lineage-specific differentiation (38). Within the osteoclast lineage (originally cloned as OCZF), ZBTB7A exerts stage-specific effects: repressing early RANKL-driven commitment by limiting NFATc1 induction, while in mature osteoclasts supporting resorptive activity and cell survival (38,39). In skeletal muscle, in-vitro studies indicate that suppression of \u003cem\u003eZBTB7A\u003c/em\u003e by miR-182-5p facilitates myogenic differentiation (40), implying a context-dependent regulatory function relevant to both tissue types.\u003c/p\u003e\u003cp\u003eFinally, \u003cem\u003eRPS3\u003c/em\u003e encodes ribosomal protein S3, a component of the 40S small ribosomal subunit with multiple extra-ribosomal roles, including regulation of apoptosis, DNA repair, and transcription (41). RPS3 also acts as a noncanonical subunit of the NF-κB complex, directly modulating the transcription of specific target genes (42,43). As NF-κB activity governs both osteoclastogenesis and catabolic inflammation in aging muscle, \u003cem\u003eRPS3\u003c/em\u003e may represent a nodal element linking these processes. Moreover, ribosomal biogenesis is a key determinant of protein synthesis capacity; its impairment in aging muscle contributes to anabolic resistance, although data on \u003cem\u003eRPS3\u003c/em\u003e expression in sarcopenic tissue remain limited (44).\u003c/p\u003e\u003cp\u003eConversely, two genes already linked to osteosarcopenia and nine additional genes associated with both osteoporosis and sarcopenia provide internal validation, demonstrating that our approach successfully recovers established disease biology. Importantly, the presence of prior evidence does not reduce their relevance within the combined osteosarcopenic phenotype but rather highlights convergent mechanisms across the bone\u0026ndash;muscle axis.\u003c/p\u003e\u003cp\u003eAmong these, \u003cem\u003eTNFSF11\u003c/em\u003e is the most translationally advanced example. Its encoded ligand, RANKL, is a key driver of osteoclast differentiation and the established pharmacological target of denosumab therapy. While several small and heterogeneous clinical studies have suggested potential improvements in muscle performance following denosumab treatment (45), a recent randomized controlled trial reported no significant difference in muscle health between denosumab and placebo groups (46). Conversely, \u003cem\u003eSTAT3\u003c/em\u003e emerges from transcriptomic analyses as a shared molecular hub for both postmenopausal osteoporosis and sarcopenia (25). Although this points to \u003cem\u003eSTAT3\u003c/em\u003e as a promising therapeutic target for dual bone\u0026ndash;muscle interventions, experimental validation across both tissue systems remains limited.\u003c/p\u003e\u003cp\u003eTwo other genes, \u003cem\u003eFOXO3\u003c/em\u003e and \u003cem\u003eAGER\u003c/em\u003e, though less explored in the combined phenotype, have been independently implicated in each disorder. FOXO3 is a central regulator of cellular aging and oxidative stress (47). In skeletal muscle, it acts downstream of the IGF-1/PI3K/AKT pathway (48), where nuclear accumulation of FOXO3 correlates with muscle atrophy and sarcopenic phenotype (49). In bone, reduced \u003cem\u003eFOXO3\u003c/em\u003e expression is characteristic of osteoporotic patients, and \u003cem\u003eFOXO3\u003c/em\u003e deficiency exacerbates oxidative damage and bone loss in aged mice (50).\u003c/p\u003e\u003cp\u003e\u003cem\u003eAGER\u003c/em\u003e, encoding the receptor for advanced glycation end products (RAGE), is similarly positioned at the intersection of inflammation, metabolism, and aging (51). Activation of RAGE by AGEs stimulates NF-κB signaling, thereby promoting catabolic processes and muscle cell loss (51,52). Osteoporosis has also been proposed to partially mediate the association between circulating AGEs and sarcopenia (51). In bone, RAGE signaling suppresses osteoblast proliferation and enhances osteoclastogenesis, driving the osteoporotic phenotype (53). Collectively, these established genes underscore the validity of our analytical framework and its ability to rediscover biologically confirmed mediators while positioning them within the broader osteosarcopenic context.\u003c/p\u003e\u003cp\u003eKEGG pathway enrichment revealed a convergence of both established and emerging biological mechanisms underlying osteosarcopenia. The overrepresentation of NF-κB, TNF, and osteoclast-differentiation pathways aligns with well-characterized processes in bone remodeling, where RANKL\u0026ndash;RANK signaling serves as a central therapeutic axis (54,55). These same pathways are mechanistically linked to sarcopenia through TNF/NF-κB\u0026ndash;mediated muscle proteolysis (56\u0026ndash;58) and elevated inflammatory signaling observed in frail and low\u0026ndash;muscle-mass older adults (59). Collectively, these findings reinforce inflammation-driven molecular crosstalk as a shared denominator in bone and muscle wasting.\u003c/p\u003e\u003cp\u003eNotably, necroptosis, a regulated form of cell death mediated primarily by receptor-interacting protein kinases RIPK1 and RIPK3, together with the pseudokinase MLKL (60), emerged as an additional enriched pathway. Necroptosis has been identified as a promising therapeutic target for the prevention of osteoporosis, with preclinical data demonstrating its modulation across osteoclasts, osteoblasts, and osteocytes (60,61). Parallel evidence from Duchenne muscular dystrophy and inflammatory myopathies shows that genetic ablation of RIPK3/MLKL or pharmacological inhibition of RIPK1 reduces myofiber necrosis and improves muscle performance in murine models (62). These converging observations suggest that regulated necrosis may represent a mechanistic bridge between bone and muscle pathology.\u003c/p\u003e\u003cp\u003eAnother intriguing finding was the enrichment of the cGAS\u0026ndash;STING DNA-sensing pathway, recently recognized as a key regulator of sterile inflammation in musculoskeletal tissues (63). Cytoplasmic accumulation of damaged DNA triggers type I interferon and proinflammatory cytokine cascades, linking DNA damage responses with chronic inflammation and tissue remodeling in both bone and muscle. This supports the hypothesis that impaired cellular quality control and innate immune activation jointly contribute to osteosarcopenic decline.\u003c/p\u003e\u003cp\u003eAlthough several infection-related KEGG terms (e.g., \u003cem\u003eMalaria\u003c/em\u003e, \u003cem\u003eShigellosis\u003c/em\u003e, \u003cem\u003eTuberculosis\u003c/em\u003e, \u003cem\u003eLegionellosis\u003c/em\u003e) also surpassed the enrichment threshold, these associations were primarily driven by broad innate immunity genes, including \u003cem\u003eIL1B\u003c/em\u003e, \u003cem\u003eIL18\u003c/em\u003e, \u003cem\u003eMYD88\u003c/em\u003e, \u003cem\u003ePYCARD\u003c/em\u003e, \u003cem\u003eRIPK1\u003c/em\u003e, \u003cem\u003eTRAF2\u003c/em\u003e, and \u003cem\u003eNOD2\u003c/em\u003e, reflecting generalized inflammasome and NF-κB signaling rather than pathogen-specific mechanisms. Their co-enrichment further underscores the central role of innate immune pathways in the systemic regulation of musculoskeletal homeostasis.\u003c/p\u003e\u003cp\u003eA major strength of GOReverseLookup lies in its ability to let researchers explicitly define states of interest by curating GO terms that reflect disease-relevant mechanisms, thereby producing transparent, auditable, and biologically interpretable gene lists. In contrast to black-box artificial intelligence models, GOReverseLookup offers full methodological transparency. Users can inspect the queried processes, inclusion criteria, and resulting statistical associations. The inclusion of ortholog-aware querying (via Ensembl and UniProtKB) across more than 240 species broadens the discovery space and helps surface evolutionarily conserved or understudied candidate genes while preserving reproducibility. Importantly, the tool performs particularly well for composite or multifactorial phenotypes, such as osteosarcopenia, where intersecting GO term sets capture biologically meaningful overlaps between related conditions.\u003c/p\u003e\u003cp\u003eNevertheless, several methodological limitations should be acknowledged. The robustness of GOReverseLookup outcomes critically depends on the careful selection and curation of GO terms, as well as on the chosen analytical parameters. Suboptimal input curation may yield biologically irrelevant or misleading associations. While cross-species ortholog integration expands candidate coverage, it can also increase the risk of false positives. Furthermore, the accuracy of the results is inherently constrained by the completeness of existing GO annotations, insufficient or outdated annotation data may lead to false negatives. The availability of tissue-specific expression datasets also presents a limitation, particularly for bone (23), in contrast to the richer datasets available for muscle.\u003c/p\u003e\u003cp\u003eIn conclusion, this study presents the first disease-focused application of GOReverseLookup, a recently developed, non-machine-learning bioinformatics tool, to systematically identify genes implicated in the pathogenesis of osteosarcopenia. The discovery of established mediators such as \u003cem\u003eTNFSF11\u003c/em\u003e and \u003cem\u003eSTAT3\u003c/em\u003e validates the robustness of the approach, while the identification of 14 previously understudied genes (\u003cem\u003eRBCK1, TERF2IP, LGALS9, ZBTB7A, RPS3, PYCARD, MALT1, BCL2L11, PRKCQ, CARD9, IL18R1, ANKRD42, BTK\u003c/em\u003e, and \u003cem\u003eNOD2\u003c/em\u003e), five of which exhibit at least moderate expression in both bone and muscle, reveals novel candidates for further exploration. KEGG enrichment analysis highlighted both established and emerging pathways, including regulated necrosis and cytosolic DNA sensing, emphasizing the interplay of inflammatory and stress-response processes in osteosarcopenia. Collectively, these findings demonstrate that GOReverseLookup can effectively uncover biologically plausible gene candidates in complex, multi-tissue syndromes. By integrating GOReverseLookup with tissue-expression stratification, KEGG pathway mapping, and DisGeNET-based gene\u0026ndash;disease association scoring, we provide a reproducible and transparent framework for gene discovery that can be extended to other multifactorial conditions beyond osteosarcopenia. The identification of shared pathways provides mechanistic targets for future studies aimed at modulating bone\u0026ndash;muscle crosstalk and mitigating osteosarcopenic decline.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eASC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eApoptosis-associated speck-like protein containing a CARD\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eBH\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eBenjamini\u0026ndash;Hochberg procedure\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eBim\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eBcl-2 interacting mediator of cell death\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eBMP\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eBone morphogenetic protein\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCARD\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eCaspase activation and recruitment domain\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eCFG\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eConvergent functional genomics\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ecpdm\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eChronic proliferative dermatitis\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eFDR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eFalse discovery rate\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eFE\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eFold enrichment\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eFGF-2\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eFibroblast growth factor 2\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGDA\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eGene-disease association\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGO\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eGene Ontology\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eGWAS\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eGenome-wide association study\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIEA\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eInferred from electronic annotation\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eIKK\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eIκB kinase complex\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eKEGG\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eKyoto Encyclopedia of Genes and Genomes\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLRF\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLeukemia/lymphoma-related factor\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eLUBAC\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eLinear ubiquitin chain assembly complex\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eMALT1\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eMucosa-associated lymphoid tissue lymphoma translocation protein 1\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eMLKL\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eMixed-lineage kinase domain-like pseudokinase\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNFAT\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNuclear factor of activated T cells\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNF-κB\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNuclear factor kappa-light-chain-enhancer of activated B cells\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eNLRP3\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eNLR family pyrin domain containing 3\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eOPG\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eOsteoprotegerin\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePGBM1\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eRBCK1-associated polyglucosan body myopathy-1\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003ePKC-θ\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eProtein kinase C theta\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eRANK\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eReceptor activator of NF-κB (encoded by TNFRSF11A)\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eRANKL\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eReceptor activator of NF-κB ligand (encoded by TNFSF11)\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eRBR\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eRING-between-RING (E3 ubiquitin ligase family)\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eRIPK\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eReceptor-interacting protein kinase\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSASP\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSenescence-associated secretory phenotype\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSOI\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eState of interest\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eSTAT3\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eSignal transducer and activator of transcription 3\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTPM\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eTranscripts per million\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eTWAS\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eTranscriptome-wide association study\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003cdiv class=\"DefinitionListEntry\"\u003e\u003cdiv class=\"Term\"\u003eWnt\u003c/div\u003e\u003cdiv class=\"Description\"\u003e\u003cp\u003eWingless/Int-1 signaling pathway\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003ch2\u003eAuthor contributions statement\u003c/h2\u003e\n\u003cp\u003eS.Ž. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. L.V. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. A.\u0026Scaron;. contributed to conceptualization, methodology, investigation and edited the manuscript. V.S. contributed to conceptualization, methodology and reviewed and edited the manuscript. N.U. contributed to conceptualization, investigation, project administration, supervision, funding acquisition and reviewed and edited the manuscript. All authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis research was funded by Slovenian Research and Innovation Agency (Grant No. J3- 4527 and P3-0043) and Foundation of ing. Lenarčič Milan Scholarship.\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eS.Ž. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. L.V. contributed to conceptualization, methodology, investigation, visualization, and writing of the original draft. A.\u0026Scaron;. contributed to conceptualization, methodology, investigation and edited the manuscript. V.S. contributed to conceptualization, methodology and reviewed and edited the manuscript. N.U. contributed to conceptualization, investigation, project administration, supervision, funding acquisition and reviewed and edited the manuscript. All authors read and approved the final version of the manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eThe authors would like to thank Dr. Chiedozie K. Ugwoke for manuscript proofreading.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eAll data supporting the conclusions of this article are provided in the main text and the Supplementary Information.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eKirk, B., Zanker, J. \u0026amp; Duque, G. Osteosarcopenia: epidemiology, diagnosis and treatment \u0026mdash; facts and numbers. \u003cem\u003eJ. Cachexia Sarcopenia Muscle\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, 609 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHuang, T. et al. Prevalence and risk factors of osteosarcopenia: a systematic review and meta-analysis. \u003cem\u003eBMC Geriatr.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e, 369 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCompston, J. E., McClung, M. R., Leslie, W. D. \u0026amp; Osteoporosis \u003cem\u003eLancet\u003c/em\u003e \u003cb\u003e393\u003c/b\u003e, 364\u0026ndash;376 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCruz-Gentoft, A. J. et al. Sarcopenia: European consensus on definition and diagnosis. \u003cem\u003eAge Ageing\u003c/em\u003e. \u003cb\u003e39\u003c/b\u003e, 412\u0026ndash;423 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eClynes, M. A., Gregson, C. L., Bruy\u0026egrave;re, O., Cooper, C. \u0026amp; Dennison, E. M. Osteosarcopenia: where osteoporosis and sarcopenia collide. \u003cem\u003eRheumatol. (Oxford)\u003c/em\u003e. \u003cb\u003e60\u003c/b\u003e, 529\u0026ndash;537 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNoh, J. Y., Yang, Y. \u0026amp; Jung, H. Molecular mechanisms and emerging therapeutics for osteoporosis. \u003cem\u003eInt. J. Mol. Sci.\u003c/em\u003e \u003cb\u003e21\u003c/b\u003e, 7623 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen, Q., Lou, H. \u0026amp; Peng, C. Recent advances in the genetic association between osteoporotic fracture and sarcopenia. \u003cem\u003eAging Pathobiol Ther.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 2\u0026ndash;9 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCho, M. R., Lee, S. \u0026amp; Song, S. K. A review of sarcopenia pathophysiology, diagnosis, treatment and future direction. \u003cem\u003eJ. Korean Med. Sci.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, e146 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFagundes Belchior, G., Kirk, B., Pereira da Silva, E. A. \u0026amp; Duque, G. Osteosarcopenia: beyond age-related muscle and bone loss. \u003cem\u003eEur. Geriatr. Med.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 715\u0026ndash;724 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMichaud, M. et al. Proinflammatory cytokines, aging and age-related diseases. \u003cem\u003eJ. Am. Med. Dir. Assoc.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 877\u0026ndash;882 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhao, Z., Yan, K., Guan, Q., Guo, Q. \u0026amp; Zhao, C. Mechanism and physical activities in bone\u0026ndash;skeletal muscle crosstalk. \u003cem\u003eFront. Endocrinol.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1287972 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePolesskaya, A., Seale, P. \u0026amp; Rudnicki, M. A. Wnt signaling induces the myogenic specification of resident CD45⁺ adult stem cells during muscle regeneration. \u003cem\u003eCell\u003c/em\u003e \u003cb\u003e113\u003c/b\u003e, 841\u0026ndash;852 (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eManolagas, S. C. From estrogen-centric to aging and oxidative stress: a revised perspective of the pathogenesis of osteoporosis. \u003cem\u003eEndocr. Rev.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 266\u0026ndash;300 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSheng, R. et al. Muscle\u0026ndash;bone crosstalk via endocrine signals and potential targets for osteosarcopenia-related fracture. \u003cem\u003eJ. Orthop. Transl\u003c/em\u003e. \u003cb\u003e43\u003c/b\u003e, 36\u0026ndash;46 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eManolio, T. A. et al. Finding the missing heritability of complex diseases. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e461\u003c/b\u003e, 747\u0026ndash;753 (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWainberg, M. et al. Opportunities and challenges for transcriptome-wide association studies. \u003cem\u003eNat. Genet.\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e, 592\u0026ndash;599 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDenny, J. C., Bastarache, L. \u0026amp; Roden, D. M. Phenome-wide association studies as a tool to advance precision medicine. \u003cem\u003eAnnu. Rev. Genomics Hum. Genet.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 353\u0026ndash;373 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNiculescu, A. B. \u0026amp; Le-Niculescu, H. Convergent functional genomics: what we have learned and can learn about genes, pathways and mechanisms. \u003cem\u003eNeuropsychopharmacology\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 355\u0026ndash;356 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSmaili, F. Z., Gao, X. \u0026amp; Hoehndorf, R. Onto2Vec: joint vector-based representation of biological entities and their ontology-based annotations. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, i52\u0026ndash;i60 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlshahrani, M. \u0026amp; Hoehndorf, R. Semantic disease gene embeddings (SmuDGE): phenotype-based disease gene prioritization without phenotypes. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, i901\u0026ndash;i907 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eŠkorjanc, A., Smrkolj, V. \u0026amp; Umek, N. GOReverseLookup: a gene ontology reverse lookup tool. \u003cem\u003eComput. Biol. Med.\u003c/em\u003e \u003cb\u003e191\u003c/b\u003e, 110185 (2025).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u003c/strong\u003e Categorization of genes based on prior literature evidence and expression levels.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"482\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 160px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePreviously linked to both osteoporosis and sarcopenia\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo prior links; expressed in bone and muscle (\u0026ge; LOW)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 161px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOther significant genes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTNFSF11\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eRBCK1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eNOD1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSTAT3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eTERF2IP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003ePLCG2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eIL1B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eLGALS9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eRIPK1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eAGER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eZBTB7A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eRTKN2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eCLU\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eRPS3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eRIPK2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eSPHK1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003ePYCARD\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eTNFRSF11A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eIL18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eMALT1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eMYD88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eTP53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eBCL2L11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eSOSTDC1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eFOXO3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003ePRKCQ\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eEIF2AK2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eCARD9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCD40LG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eIL18R1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eAIM2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eANKRD42\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eTRAF2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eBTK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eRIPK4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eNOD2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eCD36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 38px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable note\u003c/strong\u003e: Genes are stratified by (i) existing literature support for concomitant involvement in osteoporosis and sarcopenia, (ii) absence of prior disease association plus expression in both bone and muscle (low or medium), and (iii) all other significant genes. Genes in bold denote those previously linked to osteosarcopenia, whereas genes marked with an asterisk (*)\u0026nbsp;indicate at least medium tissue expression in both tissues.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2.\u0026nbsp;\u003c/strong\u003eKEGG pathways significantly enriched among the 37 osteosarcopenia-associated genes.\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"624\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePathways\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFE\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFDR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eGenes\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNF-kappa B signaling pathway\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e57.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e4.9\u0026times;10⁻\u003csup\u003e14\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eMALT1,\u0026nbsp;IL1B,\u0026nbsp;MYD88, PLCG2, PRKCQ, BTK,\u0026nbsp;TRAF2, TNFSF11,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRIPK1,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTNFRSF11A\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCytosolic DNA-sensing pathway\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e47.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e7.6\u0026times;10⁻\u003csup\u003e7\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003ePYCARD,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL18,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRIPK1,\u003cem\u003e\u0026nbsp;\u003c/em\u003eAIM2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNOD-like receptor signaling pathway\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e39.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e7.7\u0026times;10⁻\u003csup\u003e15\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eNOD1,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRBCK1,\u003cem\u003e\u0026nbsp;\u003c/em\u003ePYCARD,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL18,\u003cem\u003e\u0026nbsp;\u003c/em\u003eMYD88,\u003cem\u003e\u0026nbsp;\u003c/em\u003eNOD2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eCARD9,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTRAF2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRIPK1,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRIPK2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eAIM2\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInflammatory bowel disease\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e37.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e3.1\u0026times;10⁻\u003csup\u003e5\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL18,\u003cem\u003e\u0026nbsp;\u003c/em\u003eNOD2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eSTAT3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eProlactin signaling pathway\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e34.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e3.9\u0026times;10⁻\u003csup\u003e5\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eFOXO3,\u003cem\u003e\u0026nbsp;\u003c/em\u003eSTAT3,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTNFSF11,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTNFRSF11A\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eC-type lectin receptor signaling pathway\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e28.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e8.7\u0026times;10⁻\u003csup\u003e6\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eMALT1,\u003cem\u003e\u0026nbsp;\u003c/em\u003ePYCARD,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003ePLCG2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eCARD9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOsteoclast differentiation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e28.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e7.6\u0026times;10⁻\u003csup\u003e7\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003ePLCG2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eBTK,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTRAF2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTNFSF11,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTNFRSF11A\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTNF signaling pathway\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e26.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e1.2\u0026times;10⁻\u003csup\u003e5\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003eNOD2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTRAF2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRIPK1,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL18R1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNecroptosis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e26.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e1.3\u0026times;10⁻\u003csup\u003e7\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003eRBCK1,\u003cem\u003e\u0026nbsp;\u003c/em\u003ePYCARD,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003eEIF2AK2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eSTAT3,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTRAF2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eRIPK1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLipid and atherosclerosis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e25.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 76px;\"\u003e\n \u003cp\u003e1.7\u0026times;10⁻\u003csup\u003e9\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 284px;\"\u003e\n \u003cp\u003ePYCARD,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL1B,\u003cem\u003e\u0026nbsp;\u003c/em\u003eIL18,\u003cem\u003e\u0026nbsp;\u003c/em\u003eMYD88,\u003cem\u003e\u0026nbsp;\u003c/em\u003eSTAT3,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTP53,\u003cem\u003e\u0026nbsp;\u003c/em\u003eTRAF2,\u003cem\u003e\u0026nbsp;\u003c/em\u003eCD36,\u003cem\u003e\u0026nbsp;\u003c/em\u003eCD40LG\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eTable note\u003c/strong\u003e\u003cstrong\u003e:\u003c/strong\u003e FE, fold enrichment; FDR, false discovery rate. Pathways were identified with ShinyGO v0.82 using all human protein-coding genes in KEGG as background. Results are ordered by FE and limited to pathways with FE \u0026gt; 25 and a Benjamini\u0026ndash;Hochberg false-discovery rate (FDR) \u0026lt; 5 \u0026times; 10⁻⁵. Infection-related pathways that also met the significance threshold (e.g. \u003cem\u003eMalaria\u003c/em\u003e, \u003cem\u003eShigellosis\u003c/em\u003e, \u003cem\u003eTuberculosis\u003c/em\u003e) are presented in Supplementary Information 3.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"GOReverseLookup, Osteosarcopenia, Gene discovery, Muscle-bone crosstalk, NF-κB signaling, Necroptosis, Cytosolic DNA sensing","lastPublishedDoi":"10.21203/rs.3.rs-8000419/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8000419/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eOsteosarcopenia, the coexistence of osteoporosis and sarcopenia, is an emerging geriatric syndrome that markedly increases frailty, fracture risk, and disability. Although traditionally considered distinct conditions, growing evidence indicates that both share convergent molecular pathways linking bone and muscle degeneration. To elucidate their shared genetic background, we applied GOReverseLookup, a Gene Ontology\u0026ndash;based reverse lookup tool, using 367 curated GO terms representing key disease mechanisms. The analysis, integrating ortholog data from four vertebrate species, identified 37 genes significantly enriched for annotations related to both disorders (FDR\u0026thinsp;\u0026lt;\u0026thinsp;5 \u0026times; 10⁻⁸). Among these, 14 genes had no prior association with either condition, and five (RBCK1, TERF2IP, LGALS9, ZBTB7A, RPS3) showed moderate expression in both bone and muscle. KEGG enrichment revealed 15 pathways, prominently NF-κB, TNF, osteoclast differentiation, necroptosis, and cytosolic DNA-sensing signaling, highlighting an inflammatory and immune-regulatory convergence. These findings identify novel, biologically plausible candidates and pathways underlying osteosarcopenia, and validate GOReverseLookup as a transparent approach for cross-phenotype gene discovery. The integration of GO-based inference with expression and pathway analysis provides a reproducible framework for exploring multi-tissue syndromes.\u003c/p\u003e","manuscriptTitle":"Bridging muscle and bone: shared genetic signatures of osteosarcopenia revealed by reverse Gene Ontology lookup","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-04 10:23:55","doi":"10.21203/rs.3.rs-8000419/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5c9751b2-b341-47e3-b4e3-6db8217772dd","owner":[],"postedDate":"November 4th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":57325061,"name":"Health sciences/Biomarkers"},{"id":57325062,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":57325063,"name":"Health sciences/Diseases"},{"id":57325064,"name":"Biological sciences/Genetics"},{"id":57325065,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2025-11-13T10:47:52+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-04 10:23:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8000419","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8000419","identity":"rs-8000419","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}