Clock gene signature predicts insomnia and links to sleep/circadian parameters
The expression of clock genes in blood cells can distinguish people with chronic insomnia from healthy controls and identify the biologically more severe subtype of the condition, according to a study published July 1 in Translational Psychiatry.
Researchers led by Catarina Carvalhas-Almeida at the University of Coimbra, Portugal, in collaboration with the University of Pennsylvania, used machine learning to identify a three-gene signature from peripheral blood mononuclear cells (PBMCs) that reliably separates insomnia patients from controls and further differentiates between two clinically important subtypes: insomnia with short sleep duration (ISSD) and insomnia with normal sleep duration (INSD). The findings suggest that a minimally invasive blood test could provide objective biomarkers for a disorder that has long relied on subjective patient reports for diagnosis.
What they found
The study enrolled chronic insomnia patients and healthy controls and measured multiple physiological parameters: plasma cortisol levels, wrist and axillary temperature rhythms, and clock gene expression in PBMCs. Participants underwent polysomnography to objectively measure sleep duration, allowing the insomnia group to be split into ISSD (total sleep time under 6 hours on PSG) and INSD (sleep duration 6 hours or more).
Chronic insomnia patients showed several distinct biological signatures compared with controls. Body temperature rhythms were reduced in amplitude. Cortisol profiles showed elevated levels during the wake period before sleep onset, indicating dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis. And expression levels of multiple core clock genes (including BMAL1, PER1, PER2, REV-ERBalpha, and REV-ERBbeta) were significantly altered in PBMCs.
Importantly, most of these alterations were more pronounced in the ISSD group than in the INSD group, reinforcing the view that ISSD represents a biologically distinct and more severe form of insomnia. The short-sleep phenotype has been linked in prior work to higher cardiovascular risk, metabolic dysfunction, and increased mortality compared with insomnia with normal sleep duration.
Using machine learning analysis, the research team identified a panel of three clock genes that performed as sensitive biomarkers. This gene signature achieved two clinically valuable classifications: distinguishing chronic insomnia patients from healthy controls, and differentiating ISSD from INSD. The approach leverages the fact that clock genes are expressed not only in the brain’s suprachiasmatic nucleus but also in peripheral tissues, where their expression patterns reflect systemic circadian disruption.
Why it matters
Chronic insomnia affects an estimated 10 to 15 percent of adults, yet diagnosis remains almost entirely subjective. Clinical assessments rely on patient-reported sleep diaries and questionnaires, which are susceptible to recall bias and misperception of sleep time. Objective tools such as polysomnography and actigraphy are available but are expensive, time-intensive, and not routinely used in primary care settings where most insomnia is managed.
The distinction between ISSD and INSD carries clinical weight. Patients with ISSD show more consistent physiological abnormalities, including higher cortisol levels, greater sympathetic nervous system activation, and stronger associations with adverse health outcomes. Identifying this subgroup early could guide treatment decisions and risk stratification. However, distinguishing the two subtypes currently requires overnight polysomnography, which is impractical for widespread screening.
A blood-based gene expression panel could address both problems at once: providing objective confirmation of insomnia and simultaneously classifying the subtype. Because PBMCs are collected through a standard blood draw, the approach is minimally invasive and could be integrated into routine clinical workflows.
The study also strengthens the link between circadian biology and insomnia. Clock genes are the molecular machinery of the body’s internal timekeeping system, and their altered expression in PBMCs suggests that systemic circadian disruption is a core feature of chronic insomnia, particularly the short-sleep subtype. This positions insomnia not merely as a behavioral or psychological condition but as a biological disorder with measurable molecular signatures.
Limitations
The published abstract does not report specific effect sizes or detailed classification performance metrics such as sensitivity, specificity, or area under the curve for the machine learning model. Full evaluation of the biomarker panel’s diagnostic accuracy will require access to the complete manuscript. Additionally, sample characteristics (including sample size, age range, sex distribution, and comorbidity profiles) are not detailed in the abstract, and these factors influence the generalizability of the findings. Replication in larger, more diverse populations will be necessary before the gene signature can be considered for clinical use.
Bottom line
A three-gene clock expression signature in blood cells shows promise as an objective biomarker for chronic insomnia and its subtypes. If validated in larger studies, this approach could transform insomnia diagnosis from a purely subjective assessment to one supported by molecular evidence. The findings also reinforce that insomnia, especially the short-sleep subtype, is a disorder of circadian biology with measurable physiological consequences, not simply a complaint about poor sleep.
Source
Carvalhas-Almeida C, et al. “Clock gene signature predicts insomnia and links to sleep/circadian parameters.” Translational Psychiatry, July 1, 2026. DOI: 10.1038/s41398-026-04183-3. PMID: 42386720.