In neurogenetics, neuropsychiatric comorbidity refers to the co-occurrence of a genetic or chromosomal condition with a behavioral, psychiatric, or neurodevelopmental disorder — intellectual disability, autism spectrum disorder (ASD), ADHD, anxiety, depression, OCD, or psychosis. This is not the exception — it is the rule.

The reason this co-occurrence is so common is mechanistic, not coincidental. The same variant that disrupts brain development to produce intellectual disability also disrupts the synaptic, neurotransmitter, and circuit-level machinery that underlies mood, attention, and reality-testing. A deletion does not respect the boundaries we draw between "neurological" and "psychiatric" symptoms — it removes a dose of genes whose products are needed across many circuits at once. So when a child with 22q11.2 deletion has both a conotruncal heart defect and a 25-30% risk of schizophrenia, these are not two unrelated problems; they are two downstream consequences of the same haploinsufficiency. The psychiatric phenotype is built into the genotype.

This reframing has a direct clinical payoff. If you know the molecular lesion, you can anticipate the psychiatric risk rather than wait for it to declare itself — surveillance becomes proactive instead of reactive. It also explains why these comorbidities are not 'softer' than the medical ones: across most neurogenetic syndromes, behavioral and psychiatric symptoms are what most erode quality of life and family functioning, frequently more than the physical features that first brought the patient to attention.

Two recurring pitfalls distort care. Diagnostic overshadowing — attributing a new, treatable psychiatric symptom to the pre-existing intellectual disability ('that's just how he is') and never evaluating it. And the inverse error — treating behavior as a fixed trait rather than as communication. In a person with limited verbal ability, a change in behavior is often the only available signal of pain, seizure, infection, or a mood disorder; it is a symptom to be decoded, not a problem to be suppressed.

Each chromosomal syndrome carries a characteristic behavioral phenotype — a probabilistic constellation of behavioral and cognitive features that recurs across unrelated individuals because they share the same genomic lesion. The concept is powerful precisely because it is probabilistic, not deterministic: the deletion does not script a fixed behavior, it shifts the probability distribution. Williams-Beuren syndrome makes severe anxiety and hypersociality likely; it does not guarantee them in any one child. Understanding this keeps clinicians from either over-predicting ('she has Williams, so she will be anxious') or dismissing what they see ('that's just the syndrome').

Why do these profiles emerge at all? A contiguous deletion or whole-chromosome aneuploidy alters the dosage of many genes simultaneously, and the specific genes in that interval bias which circuits are perturbed. Williams-Beuren syndrome reflects this most vividly: the 7q11.23 deletion produces a near-mirror image of autism — hypersocial rather than socially avoidant — illustrating that 'social drive' is itself under dosage-sensitive genetic control, and that more social interest is not the same as effective social function (the Williams child seeks connection but reads social danger poorly, hence the coexisting anxiety).

The deepest teaching point is that imprinting can make the same chromosomal region produce opposite syndromes depending on parental origin. Loss of the paternal contribution at 15q11-13 yields Prader-Willi syndrome (hyperphagia, OCD-like rigidity); loss of the maternal contribution at the overlapping region yields Angelman syndrome (happy affect, seizures, near-absent speech) through loss of maternally-expressed UBE3A. Same locus, opposite phenotypes — a reminder that in neurogenetics the parent of origin, not just the gene, shapes behavior. Recognizing a known syndrome's phenotype therefore converts genetics into a forecast: it tells you what to screen for, what to teach families to expect, and when to intervene early.

Many single-gene disorders announce themselves first as psychiatric illness, with the motor or systemic features arriving only later. This ordering is the trap: a young adult presents with psychosis, depression, or treatment-resistant OCD, is diagnosed with a primary psychiatric disorder, and the underlying genetic disease goes unrecognized for years — sometimes until the chelatable copper of Wilson disease has already caused irreversible injury, or the neurodegeneration of Niemann-Pick C is advanced.

The 22q11.2 deletion is the archetype of psychiatric risk that is genetically encoded. It confers a roughly 25-30% lifetime risk of schizophrenia — the largest known single-locus risk factor for the disorder — and across the lifespan it drives high rates of ADHD, anxiety, and OCD as well, in a striking longitudinal study of over 1,400 individuals from the international consortium (Schneider et al. 2014). Mechanistically this is plausible because the deleted interval includes genes such as DGCR8 (a core component of microRNA processing) and COMT (which sets the pace of prefrontal dopamine catabolism); losing one copy nudges dopaminergic signaling toward the imbalance implicated in psychosis. The clinical consequence: childhood-onset schizophrenia warrants testing for 22q11.2 deletion (diagnostic yield ~5%), and a known 22q11.2 adolescent who withdraws and develops odd thinking is in a prodrome until proven otherwise.

The metabolic and storage disorders teach the complementary lesson — that a treatable or progressive systemic disease can masquerade as primary psychiatry. Wilson disease presents psychiatrically in 20-30% (depression, personality change, psychosis), and the liver disease that would give it away may be subclinical, so ceruloplasmin and a slit-lamp exam belong in the workup of a young adult with new neuropsychiatric symptoms plus any movement or hepatic clue. Niemann-Pick type C can present as a schizophrenia-like psychosis that is clinically indistinguishable from idiopathic schizophrenia and that precedes the diagnostic vertical supranuclear gaze palsy and ataxia by years (Ong 2021). The unifying rule: psychosis with atypical features — early cognitive decline, a movement disorder, a history of neonatal jaundice, or poor treatment response — is psychosis that deserves a genetic and metabolic workup.

Evaluating neuropsychiatric comorbidity means holding two questions at once: what is the genetic diagnosis, and what, separately, are the psychiatric and medical problems layered on top of it. The error to avoid is collapsing the second question into the first — each comorbidity is its own diagnosis with its own treatment, and naming the syndrome does not absolve you of evaluating the depression, the seizures, or the pain.

The central diagnostic move, especially in someone with limited verbal ability, is to read behavior as a symptom and work backward to its cause before reaching for a behavioral label. A new pattern of aggression, self-injury, or withdrawal should trigger a medical search first — dental or abdominal pain, constipation, UTI, otitis, menstrual pain, reflux, medication side effects — because the patient may have no other way to report any of them. This is diagnostic overshadowing run in reverse: the danger here is calling something 'psychiatric' that is in fact 'medical.'

Two investigations deserve special emphasis because they masquerade as psychiatric change. Epilepsy is common across neurogenetic syndromes and can present purely as behavioral disturbance, confusion, or aggression — especially when seizures are nocturnal or subclinical — so EEG (overnight if needed) belongs in the workup of behavioral regression. Sleep is nearly universally disrupted in these syndromes, and disrupted sleep independently amplifies irritability, inattention, and aggression; Smith-Magenis syndrome is the extreme case, with an inverted circadian melatonin rhythm (peak by day) that makes the child sleepy and dysregulated when they should be alert. Standardized tools (ABC, DBC, PAS-ADD) then let you quantify what you find and track response over time. Because no single clinician spans genetics, psychiatry, epilepsy, sleep, communication, and behavior, the workup is inherently multidisciplinary — the team structure is not a luxury but a requirement of the problem's breadth.

Management is biopsychosocial by necessity, not by ideology. Because the behavior is multiply determined — by the genotype, by an unmet medical need, by a communication gap, by the environment — pulling a single lever (usually a medication) tends to disappoint. The most reliable gains often come from the least pharmacological moves: securing the environment (locked food storage in Prader-Willi turns 'aggression' back into thwarted hyperphagia), building a way to communicate (AAC reduces frustration-driven aggression by giving the patient words), and treating the sleep or seizure problem that was amplifying everything else.

When medication is warranted, the genetic context must steer the choice, because the same drug carries different risks in different syndromes. The clearest example is ADHD in 22q11.2 deletion: stimulants genuinely help the inattention, but they act on the very dopaminergic system already biased toward psychosis by COMT and DGCR8 haploinsufficiency, so methylphenidate is used with surveillance for emerging psychotic symptoms through the adolescent prodrome — not withheld, but watched. Antipsychotics in 22q11.2 are typically dosed lower than in idiopathic schizophrenia, with metabolic monitoring made non-negotiable by these patients' baseline metabolic vulnerability.

The broader pharmacologic principles follow from neurobiology. Anticholinergic and sedating-antihistamine effects are poorly tolerated when cognitive reserve is already limited, so agents that blunt cognition or cause paradoxical disinhibition (high-dose daily benzodiazepines) are generally avoided; anxiety in Williams and related syndromes is better served by SSRIs plus cognitively-adapted exposure therapy. And some treatments are mechanistically bespoke: Smith-Magenis syndrome's inverted melatonin rhythm is corrected not by evening melatonin alone but by suppressing the abnormal daytime peak with a morning beta-1 blocker (acebutolol) and restoring the nocturnal peak with evening melatonin — a regimen that only makes sense once you understand the underlying circadian inversion. Throughout, the goal is to match the intervention to the mechanism rather than to the surface behavior.