Current psychiatric diagnoses, although heritable, have not been clearly mapped onto distinct underlying pathogenic processes. The same symptoms often occur in multiple disorders, and a substantial proportion of both genetic and environmental risk factors are shared across disorders. However, the relationship between shared symptoms and shared genetic liability is still poorly understood.

Well-characterised, cross-disorder samples are needed to investigate this matter, but few currently exist. Our aim is to develop procedures to purposely curate and aggregate genotypic and phenotypic data in psychiatric research.

As part of the Cardiff MRC Mental Health Data Pathfinder initiative, we have curated and harmonised phenotypic and genetic information from 15 studies to create a new data repository, DRAGON-Data. To date, DRAGON-Data includes over 45 000 individuals: adults and children with neurodevelopmental or psychiatric diagnoses, affected probands within collected families and individuals who carry a known neurodevelopmental risk copy number variant.

We have processed the available phenotype information to derive core variables that can be reliably analysed across groups. In addition, all data-sets with genotype information have undergone rigorous quality control, imputation, copy number variant calling and polygenic score generation.

DRAGON-Data combines genetic and non-genetic information, and is available as a resource for research across traditional psychiatric diagnostic categories. Algorithms and pipelines used for data harmonisation are currently publicly available for the scientific community, and an appropriate data-sharing protocol will be developed as part of ongoing projects (DATAMIND) in partnership with Health Data Research UK.

Copy number variants (CNVs) have been associated with the risk of schizophrenia, autism and intellectual disability. However, little is known about their spectrum of psychopathology in adulthood.

We investigated the psychiatric phenotypes of adult CNV carriers and compared probands, who were ascertained through clinical genetics services, with carriers who were not. One hundred twenty-four adult participants (age 18–76), each bearing one of 15 rare CNVs, were recruited through a variety of sources including clinical genetics services, charities for carriers of genetic variants, and online advertising. A battery of psychiatric assessments was used to determine psychopathology.

The frequencies of psychopathology were consistently higher for the CNV group compared to general population rates. We found particularly high rates of neurodevelopmental disorders (NDDs) (48%), mood disorders (42%), anxiety disorders (47%) and personality disorders (73%) as well as high rates of psychiatric multimorbidity (median number of diagnoses: 2 in non-probands, 3 in probands). NDDs [odds ratio (OR) = 4.67, 95% confidence interval (CI) 1.32–16.51; p = 0.017) and psychotic disorders (OR = 6.8, 95% CI 1.3–36.3; p = 0.025) occurred significantly more frequently in probands (N = 45; NDD: 39[87%]; psychosis: 8[18%]) than non-probands (N = 79; NDD: 20 [25%]; psychosis: 3[4%]). Participants also had somatic diagnoses pertaining to all organ systems, particularly conotruncal cardiac malformations (in individuals with 22q11.2 deletion syndrome specifically), musculoskeletal, immunological, and endocrine diseases.

Adult CNV carriers had a markedly increased rate of anxiety and personality disorders not previously reported and high rates of psychiatric multimorbidity. Our findings support in-depth psychiatric and medical assessments of carriers of CNVs and the establishment of multidisciplinary clinical services.

Cardiac intensivists frequently assess patient readiness to wean off mechanical ventilation with an extubation readiness trial despite it being no more effective than clinician judgement alone. We evaluated the utility of high-frequency physiologic data and machine learning for improving the prediction of extubation failure in children with cardiovascular disease.

This was a retrospective analysis of clinical registry data and streamed physiologic extubation readiness trial data from one paediatric cardiac ICU (12/2016-3/2018). We analysed patients’ final extubation readiness trial. Machine learning methods (classification and regression tree, Boosting, Random Forest) were performed using clinical/demographic data, physiologic data, and both datasets. Extubation failure was defined as reintubation within 48 hrs. Classifier performance was assessed on prediction accuracy and area under the receiver operating characteristic curve.

Of 178 episodes, 11.2% (N = 20) failed extubation. Using clinical/demographic data, our machine learning methods identified variables such as age, weight, height, and ventilation duration as being important in predicting extubation failure. Best classifier performance with this data was Boosting (prediction accuracy: 0.88; area under the receiver operating characteristic curve: 0.74). Using physiologic data, our machine learning methods found oxygen saturation extremes and descriptors of dynamic compliance, central venous pressure, and heart/respiratory rate to be of importance. The best classifier in this setting was Random Forest (prediction accuracy: 0.89; area under the receiver operating characteristic curve: 0.75). Combining both datasets produced classifiers highlighting the importance of physiologic variables in determining extubation failure, though predictive performance was not improved.

Physiologic variables not routinely scrutinised during extubation readiness trials were identified as potential extubation failure predictors. Larger analyses are necessary to investigate whether these markers can improve clinical decision-making.