American Journal of Medical Genetics Part B: Neuropsychiatric Genetics

Obsessive compulsive disorder (OCD) is significantly heritable, but only a fraction of the contributory genetic variation has been identified, and the molecular etiology involved remains obscure. Identifying rare contributory variants of large effect would be an important milestone in helping to elucidate the mechanisms involved. Analysis of densely affected pedigrees is a potentially useful strategy to bypass the sample size challenges of standard case-control approaches. Here we performed whole genome sequencing (WGS) of 25 individuals across two multiplex OCD pedigrees. We prioritised rare variants using a Bayesian inference approach which incorporates variant pathogenicity and co-segregation with OCD. In the first pedigree, we identified a highly deleterious missense variant in NPY5R, carried by the majority of affected individuals. This gene is brain-expressed and has previously been implicated in panic disorder and internet addiction GWAS studies. In the second pedigree, we identified a large deletion of DLGAP1 and a missense variant in MAPK8IP3, that perfectly co-segregated in a specific branch of the family: both genes have previously been implicated in OCD and autism. Both genes contribute to a protein interaction network including ERBB4 and RAPGEF1 which we had previously identified in a large Tourette Syndrome pedigree. Our analysis suggests that both energy homeostasis and downstream signalling from the post-synaptic density may both be important avenues for future research.

BackgroundCognitive-behavioral therapy (CBT) is a widely used treatment for mental disorders, yet the biological mechanisms underlying its effects, and the factors contributing to response, remain poorly understood. DNA methylation, an epigenetic mechanism shaped by both genetic and environmental factors, may offer insights into individual differences in psychotherapy outcomes. MethodsSaliva samples were collected before treatment, after treatment, and three months post-treatment from individuals with OCD undergoing the Bergen 4-Day Treatment (n = 889). DNA methylation was measured using the Illumina EPIC v02 array, followed by epigenome-wide DNA methylation analyses of CBT response. ResultsWe identified ten differentially methylated regions (DMRs) associated with treatment response at baseline, 23 DMRs showing consistent associations with response across multiple time points, and three DMRs displaying longitudinal methylation changes associated with response. These loci were annotated to genes with roles in neuroplasticity, stress response, immune function, mitochondrial processes, and gene regulation. Baseline and stable methylation signals were largely influenced by genetic variation, whereas all longitudinal associations appeared to be confounded by psychoactive medication use and psychiatric comorbidities. In addition, changes in monocyte and CD4+T cell proportions were associated with treatment response. ConclusionsWe identified DNA methylation markers associated with CBT response in OCD at baseline. Stable methylation patterns associated with treatment response are likely driven by genetic factors. Longitudinal methylation analyses should be interpreted cautiously, as medication and comorbidities can exert substantial effects - even when they remain unchanged over time. Baseline methylation profiles may ultimately help predict treatment outcomes, thereby advancing precision psychiatry.

BackgroundAttention-deficit/hyperactivity disorder (ADHD) is a common heritable neurodevelopmental disorder, affecting [~]7 million children (11.4%) in the U.S. However, ADHDs underlying genetic architecture remains largely unknown. Transcriptome-wide association studies (TWAS), which integrate expression quantitative trait loci (eQTL) and GWAS summary data, can identify differentially expressed risk genes underlying complex phenotypes. Here we conduct a TWAS of ADHD using expression data from multiple brain tissues to improve understanding of the complex genetic architecture underlying this psychopathology. MethodsWe applied the TWAS framework OTTERS to train multiple gene expression imputation models using cis-eQTL summary statistics from MetaBrain for three brain regions: cortex (n=2,683), basal ganglia (n=208), and cerebellum (n=492), and GWAS summary statistics from the most recent meta-analysis of ADHD (n=225,534; case fraction =0.17). We further conducted fine-mapping, colocalization analysis, and functional enrichment analysis. ResultsWe identified 29 significant TWAS risk genes for ADHD (11 in cortex, 4 in basal ganglia, and 14 in cerebellum). Six genes appear novel for ADHD (MPL, C1orf210, MDFIC, NKX2-2, FAM183A, HIGD1A) while four genes were previously implicated in autism spectrum disorder (XRN2, KIZ, NKX2-4, NKX2-2). Pathway analysis indicated cortex and basal ganglia were enriched for neurodevelopmental pathways and regulation of cell development, and the protein-protein interaction network was statistically significant (p=1.12E-04). ConclusionThis multi-tissue TWAS refines the genetic architecture of ADHD by identifying genes whose genetically regulated expression is associated with risk, including six candidates not previously linked to ADHD. Together, these findings provide novel insights for potential targets in translational research and drug discovery.

ObjectiveTo characterize the familial patterns of misophonia and other commonly co-occurring neuropsychiatric conditions. MethodsWe examined cross-sectional survey responses from 101 probands with misophonia and their biological parents enrolled in a genetics study. ResultsProbands had a mean age of 24.6 {+/-} 11.6 years (8-64 years), were predominantly female (88%), and had high rates of co-occurring neuropsychiatric conditions, including anxiety (70%), depression (38%), ADHD (31%), and OCD (25%). Among probands, 39% had a first-degree relative with misophonia, and 48% had at least one any-degree relative with misophonia. In addition, many probands had at least one first-degree relative with anxiety (65%), depression (57%), ADHD (40%), OCD (20%), and autism (13%). Comparing rates of neuropsychiatric conditions reported by parents, mothers had significantly higher rates of misophonia (29% maternal vs. 9% paternal, p = 0.001) and anxiety (44% maternal vs. 26% paternal, p = 0.02) than fathers. ConclusionThese findings provide new insight into the familial patterns of misophonia and co-occurring neuropsychiatric conditions. Future research on underlying genetic and environmental factors is needed to shed light on the observed shared predispositions for misophonia and other neuropsychiatric conditions in families.

BackgroundBipolar disorder (BIP) frequently co-occurs with heightened substance use (SU) and substance use disorders (SUDs). Although the strong co-occurrence of these heritable traits points to shared genetic susceptibility, the extent to which there are differences in how SU and SUD overlap with BIP genetic architecture remains unclear. MethodsWe quantified the polygenic overlap between BIP and SUDs (alcohol, cannabis, opioid, and tobacco), and BIP and SU traits (drinks per week, lifetime cannabis use, prescription_opioid use, and smoking initiation) using GWAS summary statistics and trivariate MiXeR. We then isolated the general and unique genetic contributions of SUD and SU using GWAS-by-subtraction via GenomicSEM. Next, we tested associations between polygenic risk scores (PRSs) derived from these latent factors and diagnostic and behavioral outcomes in the Norwegian Mother, Father and Child Cohort Study. Finally, we applied GSA-MiXeR to explore pleiotropic pathway enrichment shared between the latent factors and BIP. ResultsWe found extensive polygenic overlap between traits, with SUDs being more genetically correlated with BIP than SU traits. The unique SUD factor correlated positively with psychiatric disorders, whereas unique SU correlated negatively. PRSs for BIP, shared SUD/SU, and unique SUD were significantly associated with BIP, SUD, and comorbid SUD-BIP; PRS for unique SU was only associated with self-reported lifetime SU. GSA-MiXeR revealed richer gene-set enrichment for SUD/BIP than SU/BIP implicating dopamine signaling and interneuron function. ConclusionBy dissecting the genetic liability to SUD and SU and investigating their relationship with BIP we find a genetic link driven by substance dependence but not substance use more broadly.

Background Major Depressive Disorder (MDD) is clinically and biologically heterogeneous. Here, we leveraged the genetics of individual depressive symptoms to dissect the disorder's underlying heterogeneity. Methods We utilized the BIObanks Netherlands Internet Collaboration (BIONIC). A series of genome-wide association studies (effective-N range: 14,407-47,110) compared controls (N=48,286) with partially different subsets of lifetime MDD cases (range: 3,892-15,577), each endorsing one of 12 individual DSM-based depressive symptoms. Results were combined in genetic correlations that informed factor analyses with Genomic Structural Equation Modeling, decomposing underlying MDD liability dimensions. The identified factors were assessed and further characterized using multivariate regression of neurodevelopmental/psychiatric and cardiometabolic traits. Results All symptoms demonstrated substantial SNP-based heritability (h2SNP:0.088-0.127). Despite high between-symptom genetic correlations, factor analyses yielded two highly correlated (rg=0.85) but still distinct latent factors: factor 1 (F1), capturing appetite/weight loss, insomnia, guilt/worthlessness, psychomotor slowing and suicidality, and factor 2 (F2), reflecting concentration problems, anhedonia, depressed mood, appetite/weight gain and fatigue. Overall, F1 had a stronger genetic overlap with neurodevelopmental/psychiatric phenotypes (e.g., autism: standardized estimate {beta}=0.45, p=4.49 x10-; schizophrenia: {beta}=0.40, p=1.73x10-), while F2 significantly overlapped with cardiometabolic traits (e.g., metabolic syndrome: {beta}=0.44, p=8.69x10-; coronary artery disease: {beta}=0.31, p=0.009). Conclusions We identified two genetic dimensions of MDD, each linked to partially distinct clinical manifestations and underlying biology, with one reflecting neurodevelopmental/psychiatric liabilities and the other capturing a strong cardiometabolic vulnerability. Disentangling such distinct dimensions may help guide patient stratification and targeted treatment, thereby advancing precision psychiatry.

Background: Bipolar disorder (BD) is increasingly conceptualized as a heterogeneous condition with a neurodevelopmental phenotype (NDP) identifying a subgroup with early neurodevelopmental vulnerability and poorer clinical outcomes. Sensory processing (SP) abnormalities are a core feature of neurodevelopmental disorders but remain poorly characterized in BD and may reflect underlying neurodevelopmental liability. We examined whether NDP load is associated with specific SP alterations in euthymic BD patients and whether NDP-based stratification explains SP variability better than conventional BD subtype (BD 1/2). Methods: We assessed 102 euthymic BD patients and 45 healthy controls (HC) using the Adolescent/Adult Sensory Profile (AASP). NDP load (0-3) was computed from nine clinical variables grouped into neonatal, comorbidity, and neurodevelopmental clusters; a median split defined BD without NDP (BD) and BD with NDP (BD-ND). Associations between NDP load and AASP quadrants were analyzed using Spearman correlations with FDR correction. Group differences (BD, BD-ND, HC) were assessed using Welch ANOVA and post-hoc tests. Nested and multivariable linear regressions examined whether NDP classification explained SP variance beyond BD subtype, adjusting for age, sex, anxiety, and residual mood symptoms. Results: Higher NDP load correlated with greater low registration (rho=0.35, p<0.001, q=0.004), sensory sensitivity (rho=0.30, p=0.001, q=0.004), and sensation avoiding (rho=0.23, p=0.014, q=0.040), but not sensation seeking. BD-ND showed higher low registration, sensory sensitivity, and sensation avoiding than BD and HC (all qs<0.01). NDP classification explained more SP variance than BD subtype; with robust associations after adjustment. Conclusions: Sensory processing alterations in BD are dimensionally associated with neurodevelopmental load and more accurately captured by NDP-based stratification than diagnostic subtype. SP alterations may represent a transdiagnostic marker of neurodevelopmental liability within BD, supporting biologically informed stratification approaches.

BackgroundImpairments in cognitive functioning (CF) contribute to the onset, severity, and persistence of psychiatric symptoms. While specific CF domains may relate differentially to psychopathology, evidence also supports a general factor of cognitive impairment (the C-factor). We aimed to examine how general and domain-specific CF impairments relate to psychopathology using both diagnosis-specific and transdiagnostic symptom frameworks. MethodsData were drawn from five cognitive tasks administered in the deep-phenotyped, naturalistic MIND-Set cohort. A bifactor model of CF was estimated in a discovery sample (n = 206) and internally validated in a separate subsample (n = 312). Factor scores were then explored in relation to broad diagnostic clusters (stress-related disorders, neurodevelopmental disorders, comorbid disorders, and healthy controls), presence of specific diagnoses, number of diagnoses, and transdiagnostic symptom domains. ResultsThe bifactor model comprised a general CF factor (C-factor) and five specific subfactors--Reaction Time, Incompatibility, Working Memory, Inhibition, and Flexibility--and successfully replicated, although the general factor was relatively weak. Diagnosis-specific analyses showed that only individuals with stress-related disorders differed significantly from healthy controls on the C-factor and the Incompatibility factor. Higher impairment on the Incompatibility factor was associated with mood disorder diagnoses, while both the C-factor and Incompatibility factor were correlated with greater diagnostic burden. At the symptom level, the Incompatibility factor was associated with Negative Valence and Arousal domains, the C-factor with Negative Valence, and the Flexibility factor with Arousal. ConclusionThese findings indicate that broader cognitive impairment and deficits on tasks requiring inhibition under cognitive load are primarily related to mood disorders, ADHD, and transdiagnostic symptoms of negative valence and arousal. More generally, cognitive impairment appears to reflect symptom burden and transdiagnostic expression rather than diagnostic category alone, suggesting that dimensional symptom measures may provide a more informative framework for understanding cognitive impairment in clinical populations.

Polygenic risk scores (PRS) have shown increasing utility for risk stratification across complex diseases, but for psychiatric disorders such as bipolar disorder (BD), current PRS explain only a fraction of disorder liability (~1-9%), with predictive performance further diminished in non-European populations and real-world clinical cohorts. To explore the potential of integrating social and environmental risk factors alongside genetic liability to improve risk prediction, we evaluated the relationship between a PRS for BD (PRSBD) and six social risk measures - perceived stress, discrimination in medical settings, neighborhood social cohesion, perceived neighborhood disorder, cost-related medication nonadherence, and adverse childhood experiences - to BD case status in 115,275 participants (7,000 cases; 108,275 controls) from the All of Us Research Program. PRSBD was associated with BD case status across ancestry groups, though liability-scale variance explained was attenuated relative to what has been reported for curated research cohorts (R2 = 1.86% in European, 0.60% in African, 1.65% in Latino/Admixed American ancestries). Each social risk factor tested exhibited a larger effect size than PRSBD, with perceived stress (OR = 2.05 per SD) and adverse childhood experiences (OR = 2.68 for [&ge;]4 ACEs) demonstrating the strongest associations. Individuals in the lowest genetic risk decile with high social burden exhibited BD prevalence comparable to or exceeding those in the highest genetic risk decile with low social burden. These findings demonstrate the substantial explanatory power of social risk factors and support the development of integrated genetic-social risk frameworks for more accurate and equitable psychiatric risk prediction.

ObjectiveCognitive behavioral therapy (CBT) is an effective first-line treatment for obsessive-compulsive disorder (OCD), yet it remains difficult to predict who will respond to this intervention. This study investigates associations between neural activity during inhibitory control tasks and CBT outcomes, and whether task-based fMRI data could serve as a predictive marker of individual CBT response. MethodsUsing fMRI data from individuals performing an inhibitory control task across five samples (n=130, age range=8-57, 54% female) of the ENIGMA-OCD consortium, univariate associations were analyzed between activity during response inhibition and error processing and three CBT outcomes: response, remission, and pre-post treatment change in symptom severity. Random forest and support vector machine models using leave-one-sample-out cross-validation were used for prediction of CBT response and remission from fMRI activity and clinical data. ResultsRemission after CBT was associated with weaker activity in default mode regions during response inhibition and in the right supramarginal gyrus during error processing. Greater symptom reduction was linked to weaker pre-treatment activity across frontoparietal, dorsal attention, visual, and subcortical regions during response inhibition, but to stronger default mode activity during error processing. Despite these robust group-level effects, machine learning models failed to predict individual outcomes above chance level with either neuroimaging or clinical data. ConclusionWeaker activity during response inhibition in a widespread network, as well as stronger activity in default mode regions during error processing before treatment, appear beneficial to CBT response. However, these findings cannot yet be translated into individually predictive markers of CBT outcome.

Alternative-splicing events (ASE) increase transcriptomic variability and play key roles in biological functions. The contribution of ASE to bipolar disorder (BD) remains largely unexplored. We performed a Transcriptome-Wide Alternative-Splicing Analysis (TWASA) to identify ASEs and genes potentially involved in BD. The study comprised 635 individuals: a discovery sample (DS) of 31 individuals from eight multiplex BD families (16 BD cases; 15 unaffected relatives), and a replication sample (RS) of 604 subjects (372 BD cases; 232 controls). Sequencing was conducted on RNA from lymphoblastoid cell lines (DS) and whole blood (RS). TWASA was performed using VAST-TOOLS (VT), rMATS (RM), and MAJIQ/MOCCASIN (MCC). Gene-set association analyses of genes containing ASEs were performed across six psychiatric disorders. Novel ASE (nASE) were investigated in the DS using FRASER. Limited gene overlap was observed across TWASA tools. MCC identified 2,031 complex ASEs involving 1,508 genes, showing the strongest genetic association with BD across psychiatric phenotypes. Prioritization of MCC-identified ASE genes yielded 441 candidates, including DOCK2 as top candidate from the DS. Replication was obtained for 98 genes, five with an identical ASE, and four (RBM26, QKI, ANKRD36, and TATDN2) showing a concordant percentage-spliced-in direction with the DS. Finally, 578 nASE were identified in the DS, with no evidence of familial segregation or differences in ASE types. This first TWASA in BD reveals tool-specific variability, complex ASE for genes specifically associated with BD, and novel candidate genes for BD. Alternative transcript isoform abundance may represent a mechanism contributing to BD pathophysiology.

Adult ADHD is increasingly recognized across the lifespan, yet the psychometric equivalence of the Adult ADHD Self-Report Scale (ASRS) remains unverified for older populations. This study examined age-related Differential Item Functioning (DIF) in 600 adults (n = 100 per decade, ages 20-80) who completed the 18-item ASRS. Using a bi-factor Graded Response Model, we extracted latent ADHD trait scores ({omega}H = .895) and assessed DIF via ordinal logistic regression with adaptive age modeling. Five of 18 items exhibited significant uniform DIF. At equivalent latent severity, older adults were less likely to endorse hyperactivity symptoms in Part A (fidgeting, feeling "driven by a motor") but more likely to endorse specific symptoms in Part B (careless mistakes, misplacing items, interrupting). From ages 20 to 80, expected Part A scores decreased by 1.36 points (~0.27 per decade), while Part B scores increased by 1.15 points (~0.23 per decade). These findings indicate a phenotypic redistribution of ADHD symptoms as individuals age. Because the 6-item Part A screener serves as the primary clinical gatekeeper, its concentration of negative DIF suggests standard screening practice may systematically underestimate ADHD severity in older adults. We recommend using the full 18-item ASRS when screening older populations and suggest that developing age-adjusted norms would improve diagnostic accuracy.

Pediatric bipolar disorder is challenging to diagnose accurately due to symptom heterogeneity. More standardized and data-driven approaches are needed to enhance diagnostic reliability. We evaluated a clinical decision tool (nomogram), statistical methods (logistic regression, LASSO), machine learning (support vector machine, random forest, k-nearest neighbors, extreme gradient boosting), and deep learning model (multilayer perceptron) for pediatric bipolar disorder prediction across two datasets collected in academic (N=550) and community (N=511) clinical settings. We compared three modeling strategies: cross-dataset validation, cross-dataset with interaction terms, and mixed-dataset. We assessed model performance using discrimination ability, calibration, and predictor importance ranking. In the baseline cross-dataset approach, all models showed good internal discrimination in the academic dataset; but external discrimination in the community dataset substantially declined. Interaction-enhanced models slightly improved internal discrimination but not external performance or calibration. Recalibration prominently improved cross-dataset calibration without compromising discrimination, indicating that transportability problems were largely driven by probability scaling. Models trained on mixed datasets exhibited much stronger external discrimination and calibration. Across models and training strategies, family risk and PGBI-10M were consistently ranked as the most important predictors. Predictive models for pediatric bipolar disorder showed strong internal performance but limited cross-setting generalizability due to dataset shift and miscalibration. Increasing model complexity did not improve external performance, whereas training on pooled data substantially improved both discrimination and calibration. Findings suggest that sampling diversity, rather than model complexity, is more valuable for developing clinically useful and generalizable psychiatric prediction models, underscoring the importance of open and collaborative datasets.

ImportanceEating disorders (EDs) are heritable, yet the developmental pathways through which genetic liability manifests in early life remain unclear. ObjectiveTo investigate the associations between genetic liability for anorexia nervosa (AN) and binge eating (BE) and disordered eating behaviors (DEB) across childhood, and to identify the mediating roles of metabolic and psychosocial traits. Design, Setting, and ParticipantsThis longitudinal observational study used genomic and behavioral data from the Adolescent Brain Cognitive Development SM (ABCD(R)) Study, a multisite, population-based cohort of children recruited between 2016 and 2018 at ages 9 to 10 years from 21 research centers across the United States. A three-wave temporal design was employed, utilizing data from baseline (T0), Year 1 (T1), and Year 2 (T2) follow-ups. Primary analyses focused on 5,618 participants of genetically inferred European (EUR) ancestry, with exploratory analyses conducted in a diverse sample of 9,132 participants. ExposuresPolygenic scores (PGS) for AN and BE were calculated using summary statistics from the most recent genome-wide association studies. Mediators included BMI, ADHD, anxiety/depression, and social problems from the Child Behavioral Checklist assessed at Year 1 follow-up (T1). Main Outcomes and MeasuresParent reported DEB symptoms via the Kiddie Schedule for Affective Disorders and Schizophrenia (KSADS). For longitudinal association analyses, DEB were pooled across T0, T1 and T2 to assess the relationship between genetic liability and childhood symptom severity. For mediation analyses, DEB at T2 follow-up were used to ensure a clear temporal sequence between mediators at T1 and the outcomes. ResultsAmong 5,618 EUR participants (mean [SD] age, 9.91 [0.62] years; 47% female), longitudinal association models revealed that higher AN-PGS was associated with increased AN symptoms, while BE-PGS was associated with increased BE and AN symptoms. These patterns were largely consistent in exploratory cross-ancestry analyses. Mediation analyses showed that BMI mediated genetic risks across sexes, while ADHD and anxiety/depression symptoms emerged as additional mediators in females. Conclusions and RelevanceGenetic liabilities to AN and BE contribute to childhood DEB through sex-dependent pathways, highlighting the developmental continuity of ED risk from childhood. Integrating genetic profiles with behavioral markers may facilitate early identification and support multifaceted interventions. Key points QuestionDo genetic risks for anorexia nervosa (AN) and binge eating (BE) contribute to childhood disordered eating behaviors, and what mechanisms mediate these effects? FindingsIn this longitudinal study of 5,618 children of European ancestry, AN polygenic scores (AN-PGS) were associated with early AN symptoms, while BE-PGS showed transdiagnostic associations with both AN and BE symptoms. These links were mediated by BMI and psychosocial traits, including sex-specific pathways through ADHD and anxiety/depression symptoms in females. MeaningOur findings suggest that genetic liability to eating disorders manifests early in life through distinct metabolic and psychosocial pathways, highlighting a window for sex-specific targeted prevention.

BackgroundTo clarify the working mechanisms of psychotherapy for obsessive-compulsive disorder (OCD), we studied the neural effects of two psychotherapies: cognitive behavioral therapy with exposure and response prevention (CBT-ERP) and inference-based cognitive behavioral therapy (I-CBT). MethodsFifty-five individuals with OCD completed an emotional processing task during fMRI before and after 20 weekly psychotherapy sessions, using general fear and OCD-related visual stimuli. Forty-two healthy controls performed the task once. We used Bayesian region-of-interest analyses to assess changes in brain activation in prefrontal, limbic, sensory, subcortical, and visual areas, and their association with symptom improvement. ResultsAfter treatment, the CBT-ERP group (N=28) showed strong credible evidence for decreased activation across all brain regions during fear (but not OCD) versus neutral stimuli, especially in treatment responders. Conversely, the I-CBT group (N=27) showed increased activation during fear versus neutral stimuli in the precentral gyrus and lateral occipital cortex (LOC), which correlated with symptom improvement. A similar but weaker pattern was observed for OCD-related stimuli. Across all ROIs, baseline fear-related activity was associated with symptom improvement in CBT-ERP, while lower baseline activity was associated with improvement in I-CBT in, amongst others, the precentral gyrus and dorsolateral prefrontal cortex. Lower baseline LOC activation during OCD-related stimuli was linked to symptom improvement after both psychotherapies. ConclusionsThe results support CBT-ERPs mechanism of fear reduction and I-CBTs mechanism of sensory engagement. Visual brain activity during emotional processing may predict treatment response across psychotherapies.

Depression is a heterogeneous disorder, often diagnosed based on symptom co-occurrence. However, individuals may present with markedly different symptom profiles, potentially reflecting distinct underlying mechanisms. Identifying common patterns of symptoms using data-driven approaches could help clarify the heterogeneity of depression. Furthermore, examining the sociodemographic and lifestyle characteristics, health status, and polygenic scores of individuals with specific symptom profiles may offer insights into underlying risk factors. Unsupervised machine learning models were applied to large-scale data from the UK Biobank. Independent groups of individuals were assessed at two time points (the Mental Health Questionnaire: Q1; and the Mental Well-being Questionnaire: Q2) and reporting on historical or current episodes of depression. Two machine learning models, multivariate Bernoulli-mixtures and agglomerative hierarchical clustering, were used to identify common sets of symptoms and cluster individuals by symptom similarity. Consistency of results was examined between Q1 and Q2 and between clustering models. Associations between cluster membership probabilities and sociodemographic and lifestyle factors (sex, age, body mass index, smoking status, ethnicity, and deprivation), eight health conditions, and polygenic scores for bipolar disorder, schizophrenia, and attention-deficit/hyperactivity disorder (ADHD) were examined using regression models. Symptom clusters were highly consistent across Q1 and Q2 (mean correlation > 0.81) and between machine learning models (Rand Index > 0.83). Clusters aligned with the existing clinical subtypes, atypical and melancholic depression, alongside other potentially novel clusters reflecting a range of different symptom profiles. Atypical clusters (hypersomnia with weight gain) appeared in both Q1 and Q2 and were associated with younger age and higher body mass index. Distinct clusters combining insomnia, weight gain, and having thoughts of death were associated with asthma, suggesting potential inflammatory dysregulation. Further clusters were characterised by psychomotor changes and showed strong associations with Parkinsons disease, both before and after the mental health questionnaire was conducted. These findings highlight robust and clinically meaningful symptom subtypes within depression and support the use of data-driven approaches to improve diagnostic refinement and inform personalised treatment strategies.

Because both monothetic and polythetic diagnostic classification approaches focus on the presence of individual symptom(s) to identify individuals in a clinical population, they may be diagnostically sensitive clinical markers of multidimensional disorders such as developmental language disorder (DLD). DLD researchers have also used likelihood ratios (LHs) to identify possible diagnostic clinical markers of DLD, however the diagnostic sensitivity of LHs varies markedly across studies. A recent multidimensional computational elastic-net regression examined a total of 71 measures of spoken language and cognitive processing from a cohort of 223 children ages 7;0 to 11;0 with and without DLD (DLD = 110; typically developing (TD) controls = 113). All 200 iterations of the model had high discriminative power (87% - 88%) in positively identifying and distinguishing the DLD participants across all thresholds. Notably, the models identified a sparse DLD-specific deficit profile which only included nine of the 71 measures. In this study, we ask if the individual LHs for each of these nine measures are equally sensitive in identifying and discriminating the children with DLD from TD controls or if diagnostic markers of multidimensional disorders such as DLD can only be identified based on computational modeling approaches. The LHs for each of the nine measures were in the moderately high ranged (3.25 - 10). However, at the the highest LH cut points for each measure, there was little to no overlap in the children each measure identified as having DLD. Follow up analysis revealed that the elastic net model-derived predictive scores for each participant were significantly correlated with the participants language ability. The model also identified a subgroup of TD participants as having the same DLD-deficit profile as the DLD participants. This subgroup were younger, predominantly male participants whose standardized language assessment scores were lower as compared to the larger TD cohort. Taken together, the results from this study show that, because multidimensional modeling approaches such as elastic net regression leverage the variability in the deficit profiles across individual members of a diagnostic group and the unique contributions of each of the behavioral features of the phenotype, they may be an effective tool in deriving diagnostically specific deficit profiles for phenotypically complex, multicausal, multidimensional, neurodevelopmental disorders such as DLD. The results also demonstrate the robustness of the derived DLD-specific deficit profile in identifying individuals with "mild" or subclinical DLD, demonstrating the potential utility of this approach in both clinical and research arenas. What this paper adds.O_ST_ABSWhat is already known on this subject.C_ST_ABSThe identification of diagnostic markers for DLD has been a challenge for both clinicians and researchers across multiple decades. Monothetic classification markers such as non-word repetition, optional infinitive, or syntax dependencies have been explored, as well as polythetic classification approaches where a list of diagnostic symptoms is used together. However, each assumes different criteria and symptoms that should be included as diagnostic markers of DLD. What this study adds.Our study assessed the feasibility and effectiveness of monothetic vs. polythetic classification approaches for identifying DLD. Since our prior work, which used elastic net logistic regression computational modeling with strong discriminatory power, consistently selected nine key features as the DLD-deficit profile, in this effort, we calculated each of the nine features likelihood ratios to examine each measures ability to identify children with DLD. The monothetic approach failed to identify a consistent set of children with DLD, and the polythetic classification approach also did not identify participants who were shown to have mild DLD by the elastic net modeling approach. Instead, our analysis showed that a computational modeling approach, such as elastic net regression, that included small but important input from multiple cognitive and linguistic aspects of children, could better capture multifaceted information about the disorder, better account for individual variability, and consistently identify most participants with DLD. Clinical implications of this study.Elastic net logistic regression identifies a small subset of important features for distinguishing DLD and can assign a probability of DLD presence for each participant. Instead of the polythetic and monothetic approaches commonly used in the field, our study shows that integrating advanced computational modeling, such as elastic net regression, with clinician judgment can better refine assessment processes and address prior and ongoing inconsistencies in the DLD literature and diagnostic practices.

Background: Early adolescence is a common period of onset for depressive symptoms. In part, this may reflect a developmental manifestation of individual's genetic propensities as they undergo physiological and hormonal changes and interact with new environments. Many commonly proposed mechanisms assume direct effects of an individual's own genes on emerging variation in their depressive symptomatology. However, estimates of genetic influence based on analyses in unrelated individuals capture not only direct genetic effects but also genetic effects from parents and other biologically related family members. Aim: In data from the Norwegian Mother, Father and Child Cohort (MoBa), we used linear mixed models to distinguish developmentally-stable and adolescence-specific direct and parental indirect genetic effects. We examined effects of polygenic scores for major depressive disorder (MDD), ADHD, anxiety disorders, and educational attainment (EA) on depressive symptoms, which were assessed by maternal reports at ages 8 and 14. Results: Children's own MDD polygenic scores showed adolescence-specific effects on depressive symptoms ( b_PGS*wave=0.041, [95% CI: 0.017, 0.065]). Developmentally-stable direct effects from children's polygenic scores for MDD (b=0.016, [0.006, 0.039]), ADHD (b=0.024, [0.008, 0.041]) and EA (b=-0.02, [ -0.038, -0.002]) were also evident. The only evidence of indirect genetic effects was a stable effect of maternal EA polygenic scores (b=0.04, [0.024, 0.054]). Conclusion: Direct genetic effects linked to genetic liability to MDD accounted for emerging variation in depressive symptoms in adolescence. These results imply that specific etiological mechanisms related to MDD may become particularly relevant for depressive symptoms during early adolescence compared to at earlier ages.

ObjectiveTo identify immunomodulatory drug targets with genetic evidence in major depressive disorder (MDD), probe symptom-level heterogeneity in their effects, and identify drug repurposing opportunities. MethodsWe used cis-Mendelian randomisation to evaluate the targets of 204 immunomodulatory compounds, including immunosuppressants, cytokine inhibitors, and anti-infectives. As exposures, we selected genetic instruments from nine genome-wide association studies (GWASs) of protein or gene transcript levels in blood and six brain regions, capturing peripheral and central inflammatory processes. As outcomes, we used summary statistics from GWASs of MDD and eight individual depressive symptoms. We prioritised targets based on consistency across tissues and robustness of Mendelian randomisation estimates, and prioritised compounds based on predicted therapeutic effects. ResultsWe prioritised 13 drug targets (C1S, CRBN, CUL4A, DEPE1, FCGRT, FKBP1A, HRH1, IL1RL2, IMPDH1, MMP7, POLE2, PRIM1, and S1PR4) associated with MDD risk. These are targeted by 46 compounds - mostly inhibitory - including doxycycline, sutimlimab, and pomalidomide, which may have therapeutic benefits for MDD. In symptom-level analyses, most targets showed heterogeneity across symptoms. Among prioritised targets, three showed consistent effects across more than half the symptoms, while the remainder showed symptom-specific or opposing effects between symptoms. ConclusionWe provide genetic evidence supporting the repurposing of immunomodulatory drugs for MDD, including compounds acting on novel therapeutic targets. Effects differ across depressive symptoms, suggesting that symptoms respond to different drug mechanisms. These findings highlight the importance of considering individual symptoms, rather than MDD as a unitary condition, when developing treatments with a broad spectrum of action or targeting specific symptom profiles.

Objective: Autistic individuals may face elevated risk for PTSD, yet the degree to which this risk differs by sex remains unknown. We examined the association between autism and incident PTSD, characterized sex differences in risk, identified high-risk subgroups, and described post-diagnosis clinical trajectories. Method: We conducted a population-based matched cohort study using Swedish national registers. Individuals born 1990 through 2010 were followed from age 6 years through December 31, 2017. Autistic individuals (N=42,862) were matched 1:10 to controls (N=412,251) on sex and birth year. Cox proportional hazards regression estimated hazard ratios (HRs) for incident PTSD. Among those who developed PTSD, we compared care utilization, hospitalization rates, and persistence of care contacts. Results: During mean follow-up of 5.1 years, 401 autistic individuals (0.9%) and 903 controls (0.2%) developed PTSD (incidence rates: 18.3 vs 4.2 per 10,000 person-years). Autism was associated with 4.4-fold increased PTSD risk (HR=4.37; 95% CI, 3.93-4.86). Risk was higher among females (HR=4.79) than males (HR=3.39; P interaction=.006). Among autistic individuals, comorbid ADHD conferred additional risk (HR=1.38; 95% CI, 1.14-1.68). Ten-year cumulative incidence reached 6.0% among autistic females with ADHD. Autistic individuals with PTSD had higher care utilization (mean visits: 5.0 vs 3.9; P<.001), more psychiatric hospitalizations (27.9% vs 19.8%; P=.002), and more persistent courses (24.8% vs 12.3% with contacts in all 3 post-diagnosis years; P=.001). Conclusion: Autism is associated with substantially elevated PTSD risk, particularly among females with comorbid ADHD. When PTSD occurs, autistic individuals experience more severe and persistent clinical courses, supporting targeted screening and sustained follow-up.