Neurogenetics is the study of how genetic variation — inherited or arising de novo — produces neurological disease. It sits at the intersection of neurology and medical genetics, spanning diagnosis, counseling, and a fast-growing set of mechanism-targeted therapies.

The nervous system is disproportionately vulnerable to genetic disruption for a structural reason: a large majority of genes are expressed in the brain, neurons are largely post-mitotic and long-lived (so they cannot dilute or repair damage by dividing), and neurodevelopment is an exquisitely orchestrated program in which many single genes are individually essential. This is why the nervous system is the organ system most often involved in Mendelian disease.

Next-generation sequencing transformed the field — not merely by reading DNA faster, but by making trio (parent–child–parent) analysis affordable, which is what unmasked the large contribution of de novo variation. The practical consequence is scale: identifiable genetic causes now underlie roughly 50% of pediatric-onset epilepsies and 30–40% of childhood intellectual disability, alongside many early-onset movement and neurodegenerative disorders. A genetic diagnosis is rarely just a label — it can redirect treatment, define surveillance, sharpen recurrence counseling, and end a diagnostic odyssey.

Neurological disease spans a continuum of genetic architecture, and locating a condition on that continuum is what tells you which test to order and how to counsel.

At one end sit monogenic (Mendelian) disorders — a single high-penetrance gene, often presenting early and severely. These dominate severe early-onset neurology for an evolutionary reason: strongly deleterious variants are removed from the population each generation, so they are continually replenished by new mutation rather than inheritance. This is the de novo insight that trio exome sequencing revealed — in the Deciphering Developmental Disorders study, ~42% of children with a developmental disorder carried a pathogenic de novo coding variant (McRae et al., Nature 2017). It also explains the clinical paradox that an unremarkable family history does not argue against a genetic cause in severe early-onset disease.

At the other end, common late-onset conditions (Alzheimer disease, most Parkinson disease) are predominantly polygenic — many common variants of small effect, shaped by environment — with only a minority of patients carrying rare high-penetrance variants. Chromosomal disorders (aneuploidy, large CNVs) and mosaicism (post-zygotic variants confined to a subset of cells, e.g., focal cortical dysplasia) round out the architecture; mosaicism is a reminder that a normal blood test does not exclude a variant present only in affected tissue.

Neurogenetic diseases are grouped by clinical phenotype, but two principles complicate that grouping and shape how we test.

Genetic heterogeneity — the same phenotype arising from variants in many different genes — is the rule, not the exception: dozens of genes can produce an epileptic encephalopathy or a hereditary ataxia. This is the single strongest argument for broad testing (panels, exome) over serial single-gene testing once the differential is wide.

Pleiotropy — one gene causing several distinct diseases — is its mirror image: CACNA1A, for example, causes episodic ataxia type 2, familial hemiplegic migraine, and SCA6, depending on the variant and mechanism. Pleiotropy warns against anchoring on a single diagnosis from a gene name alone.

With those caveats, learning the prototypical gene for each category (below) gives a scaffold for building a focused differential — the high-yield first guesses a phenotype should bring to mind.

The genetic evaluation still begins at the bedside, and a careful history and exam often do more to narrow the differential than any single test.

The three-generation pedigree is one of the highest-yield 'tests' you can perform without a laboratory. It can reveal the mode of inheritance before a sample is drawn — vertical transmission suggesting autosomal dominant, an affected sibship with unaffected parents suggesting recessive, an exclusively maternal-line male pattern suggesting X-linked — and it simultaneously identifies at-risk relatives and flags consanguinity, which sharply raises the prior probability of a recessive disorder. A 'negative' family history is itself informative: in severe early-onset disease it points toward a de novo cause rather than away from a genetic one.

The exam narrows the gene list. Skin is a particularly rich window because skin and brain share an ectodermal origin — café-au-lait macules (NF1), ash-leaf macules and angiofibromas (tuberous sclerosis), and other neurocutaneous stigmata can effectively make the diagnosis. Subtle features (ear pits, hypertelorism, clinodactyly), multi-system involvement, and a treatment response (e.g., to pyridoxine or a ketogenic diet) all shift both the differential and the choice of test. Neuroimaging adds a parallel axis of pattern recognition — a simplified gyral pattern (LIS1/DCX lissencephaly), a leukodystrophy signature, striatal necrosis (mitochondrial disease), or subependymal nodules (TSC).

Choosing a genetic test is largely a matter of matching the test's resolution to the suspected variant type — and the field has shifted decisively toward comprehensive first-line testing.

Each test sees a different slice of variation. Karyotype sees whole-chromosome changes and large balanced rearrangements but little else; chromosomal microarray sees copy-number gains and losses (and, on SNP platforms, uniparental disomy and absence of heterozygosity) but no single-nucleotide variants; gene panels and exome see coding SNVs and small indels but call CNVs and repeat expansions poorly; genome sequencing captures coding and non-coding SNVs, indels, CNVs, and structural variants in one assay. No single test sees everything — which is why CMA, repeat-expansion testing, and methylation studies still complement sequencing.

The strategic shift away from serial single-gene testing reflects both higher yield (~30–40% in unexplained neurodevelopmental disease) and a formal evidence base: the ACMG now recommends exome or genome sequencing as a first- or second-tier test for children with congenital anomalies, developmental delay, or intellectual disability (Manickam et al., ACMG 2021). The remaining art is recognizing when a narrow differential still justifies a targeted test — trinucleotide-repeat PCR for a classic Huntington or Fragile X presentation, or a methylation study for suspected Prader-Willi/Angelman — because those mechanisms are invisible to standard sequencing.