Male pattern baldness is strongly heritable and linked to a variant in the androgen receptor gene on the X chromosome. That partly explains the 'blame your mother's father' folklore, but dozens of other genes also matter.
Androgenetic alopecia — commonly called male pattern baldness — is the most common form of hair loss, affecting a large share of men over a lifetime and progressing in a recognizable pattern: recession at the temples, thinning at the crown, or both, eventually meeting into a horseshoe-shaped fringe in more advanced cases. It happens because certain scalp hair follicles become progressively more sensitive to androgens (male sex hormones, especially a testosterone derivative called dihydrotestosterone), which shortens their growth cycle and makes each generation of hair thinner and shorter until follicles stop producing visible hair altogether.
Twin and family studies have consistently found that heredity plays a large role in who develops this pattern and how early — but "hereditary" does not mean "one single gene." The genetics turn out to be more interesting, and more distributed, than the old folk wisdom suggests.
One of the first genes convincingly linked to male pattern baldness was the androgen receptor gene, AR, which sits on the X chromosome. Because men (with one X and one Y chromosome) get their only X chromosome from their mother, a variant in this specific gene is necessarily inherited through the maternal line — which is very likely the biological seed of the popular saying that baldness "comes from your mother's father."
But the folklore oversimplifies the genetics in an important way. A mother herself carries two X chromosomes — one from her own mother, one from her own father — and she passes just one of those two X chromosomes to any given son, essentially at random. So even for the AR gene alone, a son has roughly a 50% chance of inheriting the maternal-grandfather copy and a 50% chance of inheriting the maternal-grandmother's X instead. And AR is only one contributor among many: large genetic studies have identified dozens of other locations across the genome — including regions on chromosomes 3 and 20 that are inherited in the ordinary way from either parent — that also influence baldness risk. The father's side of the family genuinely matters too, just through different genes.
Studies of early-onset baldness have found that genetic variation in and around the AR gene is a near-universal prerequisite for the trait to appear at all in the men studied — one influential analysis described AR-region variation as accounting for close to half of the attributable genetic contribution to early-onset cases specifically. But "prerequisite" is not the same as "sufficient": having the associated AR variant makes the trait possible, not guaranteed, and plenty of men without any particular risk version of AR still go bald, driven by other genetic and biological factors.
The largest genetic studies of male pattern baldness to date have identified over 60 separate regions of the genome associated with the trait, together explaining about 39% of phenotypic variance — roughly half of the ~80% heritability twin studies estimate — meaning most of the inherited variation in who goes bald, and how severely, still isn't pinned down to specific known genes. Some of the implicated genes and pathways are shared with other traits (hair growth signaling, cell-cycle and hormone-related genes), which is part of why researchers have also explored whether pattern baldness shares biology with a few other conditions — this is an area of active research and should not be read as baldness causing or predicting those conditions in any individual.
If baldness runs in your family — on either side — genetics is very likely part of the reason, but it is a many-gene story, not a single switch inherited neatly from one grandparent. It is a cosmetic and, for some people, psychologically significant condition, not a marker of overall health, and effective options (such as topical or oral medications, discussed with a clinician) exist for those who want to slow or treat it.
Locus and variant. rs6152 is a synonymous (silent) coding variant in exon 1 of the androgen receptor gene, AR (OMIM *313700), at Xq11-12 — historically genotyped as the presence or absence of a StuI restriction endonuclease site, and sometimes annotated as a G>A change at codon 211/213 (Glu, unchanged at the protein level). It lies between the CAG- and GGN-trinucleotide repeat tracts in AR exon 1 that themselves have been separately studied for association with AGA. Because rs6152 does not alter the AR protein sequence, it is generally interpreted as a marker in linkage disequilibrium with the true functional variant(s) in or near AR (candidate regulatory elements affecting AR expression level in scalp tissue), rather than as the causal change itself.
Key association studies. Ellis, Stebbing, and Harrap (J Invest Dermatol 2001; PMID 11231320) first reported that the StuI restriction site (rs6152) was present in 98.1% of young bald men (n=54) and 92.3% of older bald men (n=392), versus 76.6% of non-bald older controls (n=107) (p=0.0005 and p=0.000004, respectively), alongside an association with shorter AR CAG/GGN repeat lengths. Kucerova et al. (J Eur Acad Dermatol Venereol 2015; PMID 24665929), studying 309 Czech men with benign prostatic hyperplasia or prostate cancer, confirmed an association between rs6152 genotype and AGA grade (Hamilton-Norwood scale), reporting the G allele more frequent among men with higher-grade alopecia and higher PSA levels, though not associated with prostate cancer diagnosis itself.
X-linked inheritance and etiological fraction. Hillmer et al. (Am J Hum Genet 2005; PMID 15902657) demonstrated that genetic variability at the AR locus is the cardinal prerequisite for common early-onset AGA, with a calculated etiological (population-attributable) fraction of 0.46 in their German cohort, and proposed a polyglycine-encoding GGN repeat in AR exon 1 as the most plausible functional candidate. Because AR is X-chromosomal, the authors noted this underscores the importance of the maternal line — sons inherit their single X, and thus their AR allele, from their mother — while explicitly framing AR as necessary in most cases studied but not the sole determinant.
Genome-wide polygenic architecture. Heilmann-Heimbach et al. (Nat Commun 2017; PMID 28272467) conducted the largest GWAS meta-analysis of male-pattern baldness to date (10,846 early-onset cases and 11,672 controls across eight cohorts), identifying 63 genome-wide-significant loci (23 novel), together explaining approximately 39% of phenotypic variance; implicated genes and pathways included FGF5, IRF4, DKK2, melatonin signaling, and adipogenesis-related genes. The study also reported shared genetic architecture between MPB risk loci and several other traits/conditions (including prostate cancer and cardiovascular phenotypes), which the authors framed as evidence of overlapping biology rather than a causal or diagnostic relationship. Twin-based heritability estimates for the overall trait have been reported around 80% in some population studies, underscoring that known common variants — AR included — still leave much of the trait's heritability unexplained by named loci.
Differential considerations. AGA should be distinguished clinically from telogen effluvium (diffuse, typically reversible shedding, often triggered by illness, stress, or medication), alopecia areata (autoimmune, patchy hair loss with a distinct HLA-associated genetic architecture), and scarring alopecias. Diagnosis of AGA is clinical, based on pattern and progression (Hamilton-Norwood scale in men, Ludwig scale in women); genetic testing, including for rs6152, is not used as a diagnostic tool in clinical practice and has no established role in guiding treatment choice.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Androgenetic Alopecia (Male Pattern Baldness) comes down to these specific, well-studied positions — not a diagnosis.
Partly, but it's an oversimplification. One important gene, the androgen receptor (AR), sits on the X chromosome, which sons inherit only from their mother — and that X chromosome itself came from either the maternal grandmother or maternal grandfather, roughly 50/50. On top of that, more than 60 other genes located elsewhere in the genome — inherited from either side of the family in the usual way — also contribute to baldness risk.
Yes. While the X-linked AR gene comes only from your mother, dozens of other contributing genes are inherited from both parents in the ordinary way, and your father's own pattern and family history remain a meaningful (if imperfect) clue about your genetic risk.
No. Research suggests variation at the AR gene is a near-necessary ingredient for many cases of early-onset baldness, but it is not sufficient on its own — the trait also depends on other genes, age, and biology, and going bald is not an inevitability tied to any single genetic marker.
No — it's a cosmetic condition driven by hair follicles' sensitivity to normal androgen hormones, not a marker of overall health, hormone imbalance, or virility, despite persistent folk claims to the contrary.
Not reliably at present. Because more than 60 genetic regions are involved and together explain less than half of the trait's heritability, no current genetic test — including one that just looks at the AR gene — can precisely predict an individual's future pattern or severity of hair loss.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.