Bone Mineral Density and Your Genetics
Reviewed by the ExomeDNA science team. Last updated 2026-05-29.
This page contains general information only. For personal health decisions, consult a qualified clinician.
Bone mineral density (BMD) is a measure of how much calcium and other minerals are packed into a segment of bone — and your genetics influence where your personal set-point sits on that spectrum. Large-scale genome-wide association studies (GWAS) of heel BMD, one of the most extensively studied skeletal traits in human genetics, have now identified hundreds of loci that collectively explain a meaningful portion of lifelong fracture risk. Below: what those findings mean, which genes are involved, and what you can do with the result.
What is bone mineral density?
Heel BMD, measured by QUS or peripheral DXA, reflects the trabecular bone density of the calcaneus — the spongy, lattice-like inner bone that is metabolically active and remodels continuously throughout life. Because trabecular bone turns over faster than cortical bone, heel measurements are sensitive to metabolic changes and serve as a reliable predictor of osteoporotic fracture risk at other skeletal sites, including the hip and spine.
Osteoporosis — defined by a T-score of −2.5 or lower on DXA — affects approximately 10 million adults in the United States and is responsible for roughly 1.5 million fractures per year. Even osteopenia (T-score between −1.0 and −2.5) substantially elevates fracture probability over a lifetime. BMD declines naturally after peak bone mass (typically achieved in the late twenties) and accelerates in women following menopause due to the loss of estrogen's bone-protective effects.
Genetics determines approximately 50–80% of an individual's peak bone mass, making BMD one of the more heritable common traits in medicine. The ExomeDNA BMD result reflects polygenic signal derived from GWAS of heel bone mineral density — one of the largest phenotype-specific GWAS datasets in musculoskeletal genetics.
The genetics behind bone mineral density
The genetic architecture of heel BMD is highly polygenic, meaning hundreds of common variants each contribute a small effect. Two landmark GWAS published in 2017 and 2018 dramatically expanded the known genetic landscape of this trait.
Kemp et al. (2017, PMID 28869591) reported the identification of 153 new loci associated with heel BMD using data from the UK Biobank, bringing the total number of known loci to over 300 at that time. The study demonstrated that heel BMD has robust heritability detectable at genome-wide scale and that many associated loci map near genes with plausible roles in bone biology — including regulators of osteoblast differentiation, bone matrix production, and skeletal mechanosensing.
Kim SK (2018, PMID 30048462) extended this substantially, identifying 613 new loci associated with heel BMD, underscoring just how polygenic this trait is. Together these two studies established heel BMD GWAS as one of the most statistically powered in all of musculoskeletal genetics, with thousands of individual variants contributing to the polygenic signal.
Among the genes represented in ExomeDNA's authorized gene set for this trait, several reveal mechanistically interesting and non-obvious pathways.
ACHE encodes acetylcholinesterase, the enzyme that degrades the neurotransmitter acetylcholine (ACh) in synaptic clefts and tissue microenvironments. Its appearance in a bone density context surprises most people — but it reflects a genuinely important neuroendocrine pathway. Osteoblasts (bone-forming cells) and osteoclasts (bone-resorbing cells) both express cholinergic receptors, particularly muscarinic subtypes. Local cholinergic neurons in bone marrow and co-released ACh from sympathetic terminals activate muscarinic receptors on osteoblasts, stimulating bone formation. ACHE degrades this ACh signal in the bone marrow microenvironment. Variants that reduce ACHE activity allow ACh to accumulate around osteoblasts longer, amplifying the pro-formation signal and tilting the remodeling balance toward net bone gain. This neuroendocrine pathway is an active area of bone biology research.
ACKR3 (atypical chemokine receptor 3, also known as CXCR7) is a scavenger receptor for CXCL12 (also called SDF-1, stromal cell-derived factor 1). CXCL12 is arguably the most important chemokine in bone marrow biology: it is the primary homing signal attracting hematopoietic stem cells, mesenchymal stem cells, and osteogenic progenitor cells to the bone marrow niche. ACKR3 acts as a decoy receptor — it binds and internalizes CXCL12 without canonical signaling, regulating how much free CXCL12 is available. Because mesenchymal stem cells that home to bone marrow include the progenitor pool for osteoblasts, ACKR3 variants that alter CXCL12 availability affect the long-term supply of bone-forming cells..
AASS encodes alpha-aminoadipic semialdehyde synthase, which catalyzes the first two steps of lysine catabolism. Lysine is an essential amino acid that is critical for collagen biosynthesis — specifically, it provides the lysine residues that are hydroxylated by lysyl hydroxylase enzymes to form hydroxylysine, which then forms the cross-links that give collagen Type I its tensile strength. Type I collagen constitutes approximately 90% of the organic matrix of bone; without well-cross-linked collagen, bone mineral cannot be properly organized, and the resulting matrix is more brittle even if mineral content is normal.
ACTG2 (gamma-2 smooth muscle actin) and ABI2 (Abelson interactor 2) both participate in the actin cytoskeleton dynamics of bone cells. Osteoblasts must attach firmly to bone surfaces, polarize directionally, and secrete matrix in an organized fashion — all processes requiring dynamic actin remodeling. ABI2 scaffolds the Rac1-Arp2/3 actin polymerization pathway, which is essential for osteoblast and osteoclast morphology and motility. Beyond secretion, cytoskeletal organization is how bone cells sense mechanical load (mechanosensing): when bone is physically stressed by weight-bearing exercise, the cytoskeletal deformation in osteocytes and osteoblasts triggers signaling cascades that upregulate bone formation. ACTG2 and ABI2 variants that affect cytoskeletal dynamics in osteoblasts may therefore modulate the magnitude of bone's anabolic response to exercise.
ABO (ABO blood group) is a pleiotropic locus with associations across dozens of traits. Its presence in BMD GWAS reflects the broad regulatory reach of this chromosomal region rather than a direct mechanistic role in osteoblast biology. ABR (active BCR-related gene) contains a RhoGAP domain that regulates Rho GTPase signaling — another cytoskeletal pathway relevant to bone cell function.
What the research says
Research base: Robust.
Heel BMD is among the best-powered phenotypes in the GWAS literature for musculoskeletal disease. The evidence base supports the following quantified conclusions:
Study scale: The combined GWAS data underlying this trait includes analysis of over 400,000 individuals with heel BMD measurements from the UK Biobank and other cohorts, with over 300 genome-wide significant loci established by 2017 (Kemp et al., PMID 28869591) and over 600 additional loci identified by 2018 (Kim, PMID 30048462) — making heel BMD one of the highest-resolution polygenic architectures characterized in human medicine.
Fracture prediction: Polygenic scores derived from heel BMD GWAS loci predict osteoporotic fracture risk independently of clinical risk factors. Individuals in the bottom quintile of polygenic BMD score have meaningfully elevated fracture probability over a lifetime compared to those in the upper quintiles, with the relationship most pronounced for hip and vertebral fractures.
Heritability: Twin and family studies estimate BMD heritability at 50–80%. Common variants identified in GWAS explain a substantial portion of this, with the remainder attributable to rare variants, gene-environment interaction, and environmental factors not captured by genotyping.
Exercise interaction: Wang et al. (2019, PMID 31453325) demonstrated that genetic effects on BMD interact with physical activity. Individuals with higher polygenic BMD scores who are also physically active show additive or synergistic bone density gains compared to either factor alone. Conversely, physical inactivity blunts genetic advantage in bone — and weight-bearing activity meaningfully improves BMD even in lower-scoring individuals.
Collagen quality dimension: Research from the AASS locus highlights that GWAS signal for BMD may partly capture variation in bone matrix quality (collagen cross-linking architecture) rather than mineral density alone. Bone toughness — resistance to fracture under impact — depends on both the mineral component and the collagen scaffold. DXA-measured BMD does not directly capture collagen quality, meaning genetic risk may in some cases reflect bone fragility not fully visible on standard imaging.
How bone mineral density affects you
BMD is not a static number. It follows a predictable arc across the lifespan: bone accrual from childhood through the late twenties, a plateau through midlife, and progressive decline thereafter — with a sharper decline in women during the perimenopausal and early postmenopausal window driven by estrogen loss.
A higher BMD result means your genetics favor a higher bone mineral density set-point — associated with stronger bones, lower fracture risk over a lifetime, and greater resilience through the bone-loss years. Individuals with higher polygenic BMD scores tend to reach higher peak bone mass and, all else equal, maintain bone density above the clinical osteopenia threshold for longer.
A lower BMD result means your genetic set-point is lower, which does not mean osteoporosis is inevitable — but it does mean the margin for lifestyle-driven bone maintenance is smaller, and that early, consistent attention to bone-protective habits is more consequential for long-term skeletal health.
Sex differences are real and large. Women lose bone more rapidly than men after midlife (particularly in the first 5–10 years post-menopause), meaning a lower polygenic score in a woman may carry more absolute fracture risk than the same score in a man. Hormone replacement therapy substantially attenuates this loss; its appropriateness is a conversation for a clinician.
Secondary causes amplify genetic risk. Conditions that secondarily reduce BMD — prolonged corticosteroid use, rheumatoid arthritis, celiac disease, vitamin D deficiency, amenorrhea from low body weight — stack multiplicatively with genetic predisposition. A lower polygenic score combined with a secondary cause is a strong signal to pursue formal bone density evaluation.
Working with your bone mineral density result
Whether your result is higher or lower, the actions below directly influence where your BMD trajectory goes from here. The most effective strategies, in evidence-ranked order:
Weight-bearing aerobic exercise is the single most effective bone-building lifestyle intervention available. Walking, running, hiking, dancing, and court sports all apply ground-reaction forces through the skeleton that mechanically stimulate osteoblasts via the cytoskeletal mechanosensing pathway (ACTG2, ABI2). Aim for 150 minutes per week of moderate-intensity weight-bearing activity.
Resistance training adds periosteal bone loading — particularly at the hip, spine, and wrist — beyond what aerobic activity alone provides. Progressive overload with free weights, machines, or body weight (jumping, plyometrics) directly stimulates bone apposition at cortical surfaces. Two to three sessions per week targeting major muscle groups is the evidence-supported dose.
Calcium intake (1,000–1,200 mg/day) provides the raw mineral building material for hydroxyapatite crystal deposition in bone matrix. Food sources (dairy, fortified plant milks, leafy greens, canned fish with bones) are preferred over supplements for absorption and safety; supplementation is reasonable if dietary intake is inadequate.
Vitamin D (800–2,000 IU/day) is required for intestinal calcium absorption and for osteoblast function; deficiency substantially accelerates bone loss. Serum 25-OH vitamin D levels are easy to test and guide dosing more precisely than population targets.
Protein-adequate diet supports collagen synthesis — particularly the glycine, proline, and lysine amino acid content of Type I collagen (relevant to the AASS pathway). Dietary protein restriction common in caloric-restriction regimens accelerates bone loss; adequate protein (≥1.0 g/kg body weight) is the minimum target for bone health in older adults.
Baseline and longitudinal DEXA or heel ultrasound screening is the most actionable step for those with a lower polygenic result. Clinical guidelines set baseline DXA at age 65 for women and around 70 for men, earlier where clinical risk factors are present — prior fragility fracture, parental hip fracture, long-term glucocorticoid use, low body weight, smoking, or conditions such as celiac disease or rheumatoid arthritis. Screening timing is set on those clinical factors, not on a polygenic score.
Consult a clinician before beginning any new exercise program for those with existing bone fragility, balance concerns, or joint disease. Pharmacological bone protection (bisphosphonates, denosumab, romosozumab) is a clinician-prescribed conversation reserved for individuals who already meet clinical thresholds — not a lifestyle supplement.
ExomeDNA results are not intended to be used to start, stop, or change any medication, or to determine a dose. That is a clinical decision.
Related traits and genes
Bone mineral density does not exist in isolation — it connects to a network of related genetic and phenotypic signals across the ExomeDNA trait library.
ACHE appears in other ExomeDNA trait contexts involving autonomic nervous system regulation, including heart rate variability and resting heart rate — reflecting its role as the primary ACh-degrading enzyme across multiple physiological systems. The bone biology angle described here (cholinergic regulation of osteoblast activity) is distinct from its cardiac autonomic role.
ACKR3/CXCL12 biology overlaps with traits involving hematopoiesis and immune cell trafficking, since CXCL12 is the master chemokine for bone marrow stem cell homing across lineages.
AASS and amino acid metabolism genes connect to protein metabolism traits, muscle traits, and body composition traits — all of which share lysine and collagen-relevant pathways.
ACTG2 and actin cytoskeleton genes appear across musculoskeletal traits, including muscle fiber type, grip strength, and body composition — reflecting the shared cytoskeletal machinery in both muscle and bone cells.
Clinically adjacent traits: Vitamin D levels (calcium absorption, osteoblast function); Body mass index (lean mass and bone loading); Muscle strength (primary mechanical stimulus for bone apposition).
Frequently asked questions
What is heel bone mineral density and why is it measured at the heel?
The heel (calcaneus) contains abundant trabecular (spongy) bone that is metabolically active and remodels rapidly, making heel BMD sensitive to hormonal changes and a reliable predictor of fracture risk at the hip and spine. Quantitative ultrasound of the heel is faster and lower-cost than full-spine DXA, which is why the largest BMD GWAS datasets use heel measurements from cohorts like the UK Biobank.
Does a higher BMD polygenic score mean I cannot get osteoporosis?
No. A higher polygenic score shifts your probability toward stronger bones but does not make osteoporosis impossible. Genetics explains 50–80% of peak bone mass variation; diet, exercise, hormonal status, medications, and secondary conditions drive the rest. The score is a baseline, not a guarantee.
I have a lower BMD score — should I be worried?
A lower polygenic score is information, not a clinical finding. It means your genetic set-point for BMD is lower than average, which makes consistent bone-protective habits more consequential and early screening more worthwhile. Baseline DEXA is the standard measure of bone density; timing follows clinical guidelines.
What does ACHE (acetylcholinesterase) have to do with bone density?
Osteoblasts in bone marrow express muscarinic cholinergic receptors; acetylcholine released locally stimulates these receptors to promote bone formation. ACHE degrades this signal. Variants that reduce ACHE activity allow acetylcholine to accumulate longer around osteoblasts, prolonging the pro-formation signal and resulting in higher BMD.
How much can lifestyle change my BMD if my genetic score is lower?
Substantially. Physical activity modifies genetic effects on bone density. Weight-bearing exercise and resistance training can increase BMD by 1–3% annually in adults of all ages, including those with lower polygenic scores — meaningfully shifting trajectory over years.
When should I get my first bone density scan?
Standard guidelines recommend DXA at age 65 for women and 70 for men. Earlier screening is appropriate for: prior fragility fracture, parental hip fracture, long-term glucocorticoid use, low body weight, smoking, secondary conditions such as celiac disease or rheumatoid arthritis.
References
- Kemp JP et al. Identification of 153 new loci associated with heel bone mineral density and functional involvement of GPC5 in osteoporosis. Nat Genet. 2017;49(10):1468–1475. PMID 28869591
- Kim SK. Identification of 613 new loci associated with heel bone mineral density and a polygenic risk score for bone mineral density, osteoporosis and fracture. PLoS One. 2018;13(7):e0200785. PMID 30048462
- Wang H et al. Genotype-by-environment interactions inferred from genetic effects on phenotypic variability. Nat Commun. 2019;10(1):4input. PMID 31453325
ExomeDNA genetic results are for wellness and educational purposes only. Consult a clinician for personalized health guidance.
Additional research: bone density
The genetics behind bone density
Several genes represented in the authorized variant set for this trait point to distinct biological mechanisms that collectively shape bone architecture.
Among the most biologically compelling are AXIN1 and AXIN2. These genes encode scaffold proteins that are core components of the beta-catenin destruction complex, a molecular brake on Wnt signaling. The Wnt/beta-catenin pathway is one of the most critical regulators of bone formation known in human biology. When Wnt ligands bind their receptors, the destruction complex is inhibited, beta-catenin accumulates, and downstream target genes are activated — including those that drive osteoblast differentiation and survival. Osteoblasts are the bone-building cells responsible for laying down the mineralized matrix that gives bone its density.
AXIN1 and AXIN2 normally act as negative regulators: they facilitate the phosphorylation and degradation of beta-catenin, keeping Wnt signaling in check. Genetic variation in these genes that alters their scaffolding function can shift the equilibrium of this pathway, influencing how actively osteoblasts form new bone. AXIN2 variants in particular have been documented in human genetics studies in connection with skeletal and dental developmental differences, consistent with its role in mineralized tissue biology. When AXIN function is reduced, more beta-catenin escapes degradation, Wnt targets stay active longer, and bone-forming activity may be enhanced.
This Wnt pathway axis is the primary molecular narrative behind bone density genetics — and it is why AXIN1 and AXIN2 represent especially meaningful anchors in understanding the genetic architecture of this trait.
Beyond the Wnt axis, ADAMTSL1 contributes through a different mechanism: extracellular matrix (ECM) organization. ADAMTSL1 is a secreted protein with thrombospondin type 1 motifs — structural domains that interact with fibrillin microfibrils in the ECM. Bone mineral does not deposit in isolation; it nucleates within a collagen-fibrillin scaffold, and the quality of that scaffold influences how efficiently mineralization proceeds. ADAMTSL1's interactions with fibrillin suggest it plays a role in organizing the ECM environment in which bone mineral is laid down, making it relevant to bone architecture at a structural level.
AFF1, a transcriptional coactivator in the AF4/lymphoid nuclear protein family, appears in genome-wide association data across multiple skeletal and metabolic traits, reflecting its broad role in transcriptional regulation that may include genes involved in bone cell differentiation.
ABR, which encodes a protein with GTPase-activating protein (GAP) activity involved in Rho GTPase signaling, rounds out the signaling-pathway dimension of this trait's genetic basis. Rho GTPase pathways influence cytoskeletal dynamics in bone cells, relevant to osteoblast function and mechanosensing.
What the research says
Research base: Moderate.
The genetic associations underlying this trait were identified through structured analysis of UK Biobank population data, reported in Carey et al. (2024). The study applied Confirmatory Factor Analysis to distill coherent phenotypic factors from thousands of measured variables in the UK Biobank, a large prospective cohort of UK adults. Bone density emerged as one of the structured factors (Factor 19) identified through this phenotype distillation framework — meaning it represents a phenotype with consistent genetic and biological coherence across the population, not an arbitrary measurement.
This methodological approach reflects a more principled way of identifying biologically meaningful phenotypes from large-scale biobank data, moving beyond single-variable analysis toward structured representations of human health variation.
Stat block: Bone density is moderately heritable, with estimates from twin and family studies suggesting genetics accounts for roughly 50–85% of peak bone mass variation between individuals — one of the higher heritability estimates among common skeletal traits.
Stat block: The Wnt signaling pathway, anchored by genes including AXIN1 and AXIN2, has been implicated in bone mineral density across multiple independent population studies and represents one of the most pharmacologically targeted pathways in bone biology research.
Because this trait is grounded in a phenotype-distillation study rather than a classical single-endpoint GWAS, specific effect sizes and odds ratios for individual variants are not reported here. The confidence tier reflects the quality of the underlying phenotype framework and the biological coherence of the associated genes.
References (bone density)
- Carey CE, et al. (2024). Principled distillation of UK Biobank phenotype data reveals underlying structure in human variation. Nature Human Behaviour. PMID: 38965376.
Additional research: bone density score
The genetics behind bone density score
Research base: Moderate.
A 2023 genome-wide association study published in Cell Genomics (PMID 38116116) analyzed data spanning three large population biobanks — Taiwan Biobank, Biobank Japan, and UK Biobank — to map genetic loci associated with bone density T-score alongside 35 other quantitative traits. The trans-ancestry design of that study is scientifically meaningful: genetic associations that replicate across East Asian and European population samples carry greater credibility than findings anchored to a single ancestry group, because they are less likely to reflect population-specific linkage patterns. The study identified hundreds of novel loci across the 36 traits examined, including associations with bone density.
Several genes within the ExomeDNA-authorized panel are relevant to understanding what that genetic architecture looks like at a biological level.
CARTPT — a neuropeptide with a bone connection. The most distinctive gene in this panel is CARTPT, which encodes cocaine- and amphetamine-regulated transcript prepropeptide. This neuropeptide is primarily known for its role in the hypothalamus, where it participates in energy homeostasis and appetite regulation. What makes CARTPT remarkable in the context of bone biology is that it is also expressed in bone tissue itself, where it has been shown to influence osteoblast activity and the bone remodeling cycle through both central hypothalamic signaling pathways and direct peripheral mechanisms. This neuropeptide-bone axis represents a genuinely novel intersection between neurological and skeletal systems — a reminder that bone density is regulated by biology that extends well beyond calcium metabolism alone.
BCKDHB — amino acid catabolism and collagen scaffolding. The gene BCKDHB encodes the beta-subunit of the branched-chain alpha-keto acid dehydrogenase complex, a mitochondrial enzyme system responsible for catabolizing the branched-chain amino acids leucine, isoleucine, and valine. The relevance to bone becomes clear when considering that collagen — the primary structural protein of the bone matrix — requires a continuous supply of amino acid precursors. Efficiency in branched-chain amino acid catabolism may influence the substrate pool available for building and maintaining the collagen scaffolding upon which bone mineral is deposited.
ASS1 — arginine, nitric oxide, and bone remodeling. ASS1 encodes argininosuccinate synthase 1, an enzyme involved in arginine biosynthesis. Arginine serves as the substrate for nitric oxide synthase enzymes, which produce nitric oxide (NO). Nitric oxide functions as a signaling molecule in bone tissue, modulating the relative activity of osteoblasts and osteoclasts. The ASS1 pathway thus connects amino acid metabolism to a signaling axis with direct implications for bone remodeling balance.
CCDC170 — a genomic locus with repeated skeletal associations. CCDC170, a coiled-coil domain containing protein located at chromosome 6q25.1, has appeared across multiple bone-related genomic studies. Its precise molecular function in bone tissue has not been fully characterized, but its recurrence in skeletal phenotype analyses positions it as a genomic region of interest.
AFF1 is a member of the AF4/lymphoid nuclear protein family functioning as a transcriptional coactivator. Its appearance in bone density GWAS contexts is consistent with its broader role in regulating gene expression programs across tissue types.
stat-block The trans-ancestry design of Chen et al. 2023 (PMID 38116116) tested associations across East Asian and European populations simultaneously. Replication across ancestry groups strengthens confidence that identified loci reflect genuine biological relationships rather than ancestry-specific statistical artifacts.
What the research says
The genomic landscape of bone density T-score is one of the more extensively studied quantitative phenotypes in human genetics. Large-scale biobank efforts have progressively expanded the known set of associated loci, revealing that bone density is influenced by hundreds of genetic variants acting across a wide range of biological pathways — skeletal development, endocrine signaling, connective tissue biology, and now, neuropeptide signaling via pathways such as those involving CARTPT.
The Chen 2023 analysis (PMID 38116116) contributed to this literature by applying a trans-ancestry framework that increases statistical power and improves the generalizability of findings. Identifying associations that hold across Taiwan Biobank (East Asian ancestry), Biobank Japan (Japanese ancestry), and UK Biobank (primarily European ancestry) participants provides a cross-population foundation that earlier single-biobank studies could not offer.
It is important to understand what genetic associations in this context do and do not represent. These are population-level statistical relationships between genetic variants and measured bone density values. They illuminate biological pathways and mechanisms relevant to bone metabolism. They do not assign a fixed skeletal fate to any individual. Bone density across a lifetime is shaped by the interplay of genetic predisposition with environmental exposures, physical activity, dietary patterns, hormonal status, and other modifiable factors.
stat-block Bone density is a polygenic trait, meaning that many genetic variants each contribute small effects. No single gene determines an individual's bone density outcome. The value of genetic data lies in understanding which biological pathways are relevant — not in predicting a single outcome.
References (bone density score)
- Chen CY et al. (2023). Analysis across Taiwan Biobank, Biobank Japan, and UK Biobank identifies hundreds of novel loci for 36 quantitative traits. Cell Genomics. DOI: 10.1016/j.xgen.2023.100436. PMID: 38116116.