Basic Neurosciences · Paper A
Genetic loci of neuropsychiatric disorders
Several neuropsychiatric disorders map to one specific, well-characterised chromosome, and knowing the exact chromosome helps distinguish similar-looking conditions from each other. For example, imagine two forms of early-onset Alzheimer's disease caused by mutations on two different chromosomes. In practice, knowing which specific chromosome and gene is involved is exactly what these questions test.
Huntington's disease results from a CAG repeat expansion in the HTT gene on chromosome 4, inherited as autosomal dominant, with anticipation more marked through paternal transmission. Early-onset Alzheimer's disease has three genetic causes. APP sits on chromosome 21, also present as an extra copy in Down syndrome, explaining that condition's near-universal early Alzheimer's changes. PSEN1, on chromosome 14, is the commonest familial cause. PSEN2, on chromosome 1, is the rarest. APOE on chromosome 19 is a risk-modifying, not causative, gene for the more common late-onset sporadic form. Other important loci include FMR1 on the X chromosome for fragile X syndrome, and MAPT on chromosome 17 for familial frontotemporal dementia. The 22q11.2 deletion causes DiGeorge syndrome and raises the risk of schizophreniaA long-term mental illness involving episodes of psychosis such as hallucinations and delusions.Read more → (a psychotic illness involving loss of touch with reality). Take a patient with Down syndrome who develops early cognitive decline in their 40s. That pattern reflects the extra APP gene copy on chromosome 21. For Parkinson's disease, LRRK2 (leucine-rich repeat kinase 2) is the single gene most commonly implicated in autosomal dominant familial Parkinson's disease, ahead of SNCA (alpha-synuclein), which was the first Parkinson's gene identified but accounts for fewer familial cases. PARK2 (parkin) mutations instead cause autosomal recessive, typically early-onset Parkinson's disease. Genetic linkage studies that map a disorder to a locus are reported as a LOD score (logarithm of the odds). This is the log10 of the ratio of the likelihood that two loci are linked at a given recombination fraction, versus the likelihood that they assort independently (unlinked). By convention a LOD score above +3 is taken as significant statistical evidence FOR linkage, with odds of at least 1000:1 in favour. A LOD score below -2 is taken as evidence AGAINST linkage at that distance; scores in between are inconclusive and need a larger family or sample to resolve.
Genetic loci of neuropsychiatric disorders describes where in the genome specific psychiatric and neurological conditions are located, and how the scientific approach to finding these loci (locations on a chromosome) has changed radically over time.
A locus is simply a specific physical location on a chromosome. For a small number of neuropsychiatric conditions, one locus, or a handful of loci, is enough to cause the disease outright (a "single-gene" or Mendelian condition). For most common psychiatric disorders, by contrast, risk comes from thousands of loci acting together, each contributing only a tiny amount, a genuinely different genetic architecture that the exam expects you to be able to tell apart.
For example, imagine two forms of early-onset Alzheimer's disease, each caused by a mutation on a different chromosome: that single-gene model reflects an older approach that mostly applies to rare, highly heritable conditions, not to common psychiatric illness. Earlier candidate-gene studies, for example the mid-2000s work on NRG1 (neuregulin 1) and dysbindin as supposed risk genes, tested one pre-selected gene at a time based on a biological hypothesis. Many of these early findings failed to replicate in larger, better-designed studies, teaching the field an important caveat: a single small study linking one gene to one disorder should be interpreted cautiously until replicated at scale.
GWAS (genome-wide association studies) scan the entire genome for common genetic variants statistically associated with a disorder, without needing to guess the gene in advance. This approach, powered by very large sample sizes often coordinated through international collaborations such as the Psychiatric Genomics Consortium, has replaced most single-candidate-gene work for common disorders such as , bipolar disorder, and major depression. The consistent finding across GWAS is that these disorders are highly polygenic: risk comes from the combined, additive effect of many hundreds or thousands of common variants, each contributing only a tiny individual effect, rather than from one dominant disease-causing gene. Because testing millions of variants at once creates enormous risk of false positives, a GWAS finding is only accepted as real once it clears a very strict, pre-agreed statistical threshold, far stricter than an ordinary p<0.05, and has replicated in an independent sample.
Alongside the polygenic, GWAS-era picture for common disorders, a smaller group of neuropsychiatric conditions still follow classical single-gene, Mendelian inheritance, and these loci remain directly clinically relevant. Huntington's disease results from a CAG repeat expansion in the HTT gene on chromosome 4, inherited as autosomal dominant, with anticipation (earlier onset in successive generations) more marked through paternal transmission. Early-onset Alzheimer's disease has three recognised genetic causes: APP on chromosome 21, also present as an extra copy in Down syndrome, explaining that condition's near-universal early Alzheimer's changes; PSEN1 on chromosome 14, the commonest familial cause; and PSEN2 on chromosome 1, the rarest. Other clinically important loci include FMR1 on the X chromosome for fragile X syndrome, MAPT on chromosome 17 for familial frontotemporal dementia, and the 22q11.2 deletion, which causes DiGeorge syndrome and substantially raises the risk of .
APOE on chromosome 19 is easy to confuse with APP, PSEN1, and PSEN2, because all four get mentioned in the same breath about Alzheimer's disease, but APOE is only a risk-modifying gene for the far more common late-onset sporadic form, behaving very differently from the fully penetrant, causative genes above it. A second, subtler confusion is mixing up a heritability estimate with an individual risk prediction. Consider a heritability estimate for , which twin and family studies consistently place as substantially genetically influenced at the population level; that figure describes variation across a whole population, not the probability that any single individual will develop the disorder, and it says nothing about which specific genes are responsible in any one person. A high heritability figure does not mean a specific person's illness is "mostly genetic" in some deterministic sense; environment and gene-environment interaction still matter substantially. A third look-alike pair is the failed candidate-gene finding versus a genuine, replicated GWAS locus: both get reported as "the gene for" a disorder in older material, but only the replicated GWAS finding has survived independent confirmation.
For the classical single-gene conditions, diagnosis relies on targeted genetic testing, for example CAG repeat length analysis for suspected Huntington's disease, or specific mutation screening for PSEN1 in a patient with a strong three-generation family history of early-onset dementia. This kind of targeted, deterministic genetic testing is only appropriate for the classical single-gene conditions; it has no equivalent for common polygenic psychiatric disorders like , where no single test can confirm or exclude the diagnosis genetically: a polygenic risk score, which sums a person's carried risk variants weighted by their individual small , remains primarily a research tool and is not validated for individual clinical diagnosis or prediction.
Management for confirmed single-gene conditions centres on genetic counselling for the patient and at-risk relatives, delivered through recognised clinical genetics services, alongside condition-specific treatment, for example symptomatic management of chorea in Huntington's disease, or cholinesterase inhibitor medication trialled in early-onset Alzheimer's disease. That sits under ordinary NICE dementia guidance, while GeneReviews-style clinical genetics summaries make the separate counselling point: once a familial pathogenic variant is known, predictive testing and reproductive counselling can be offered to relatives, but this is very different from using a polygenic score to diagnose . For the common polygenic disorders, there is no genetic "management" in this targeted sense; a polygenic risk score cannot currently be used to guide an individual patient's treatment, and management follows the disorder's usual clinical pathway rather than its genetic architecture.
If a question asks you which gene is "responsible" for Alzheimer's disease and offers APOE as an option alongside APP, PSEN1, or PSEN2, remember APOE only modifies risk for late-onset sporadic disease; it does not cause the disease the way the other three do in early-onset familial disease. This APOE-versus-the-rest distinction is one of the most reliably tested points in the whole topic.
A handful of neuropsychiatric conditions are caused by a single identifiable locus you can test for directly, but most common psychiatric disorders are highly polygenic, with risk spread across thousands of variants that only large, replicated GWAS, not a single candidate gene or an individual risk score, can reliably identify.