Biology subjects

Donohoe, G.

Publications and source records attributed to Donohoe, G..

13 recordsLinked to original sources

bioRxiv2025

The Dynamic Nature of Genetic Risk for Schizophrenia Within Genes Regulated by FOXP1 During Neurodevelopment

FOXP1 (Forkhead-box protein P1) is a crucial transcription factor in neural development and is associated with schizophrenia (SCZ). FOXP1-regulated genes may contribute to genetic risk of SCZ and this may vary across different stages of neurodevelopment. We analyzed transcriptomic data from mouse and human models of FOXP1 loss-of-function across prenatal and postnatal developmental stages, including neural stem cells from embryonic mice (E14.5) and human brain organoids (equivalent to second trimester), and cortical tissues from different mouse postnatal stages P0, P7, and P47. P0 in mice corresponds to the third trimester in humans, while P7 and P47 represent early childhood and adolescence, respectively. The effect of FOXP1 disruption on gene expression in cortical tissues/cells was assayed using RNA-seq, including time-course and pairwise gene expression analysis. Linkage disequilibrium score regression assessed if FOXP1-regulated genes were enriched for SCZ heritability. Gene-set enrichment analysis investigated if FOXP1-regulated genes were enriched for SCZ-associated genes reported as differentially expressed in single cortical cell studies. SynGO analysis mapped FOXP1-regulated genes to synaptic locations and functions. FOXP1-regulated genes are enriched for SCZ heritability, with significant results for E14.5, P7 and P47 but not P0. The P7 gene-set showed the strongest enrichment for SCZ-associated genes from single cortical cell studies. FOXP1-regulated genes at both P7 and P47 were involved in multiple synaptic functions and were mainly enriched within glutamatergic excitatory neurons, with P47 also showing enrichment within GABAergic inhibitory neurons across regions of the postnatal cortex. Prenatal FOXP1-regulated genes were enriched in progenitor cells and also mapped to the synapse. Genetic risk for SCZ within FOXP1-regulated genes follows a dynamic trajectory across developmental stages, showing stronger effects at timepoints that map to early childhood, followed by adolescence and second trimester. Author SummarySchizophrenia is a complex disorder caused by many genes. While existing treatments can help manage some positive symptoms, such as hallucinations and delusions, they are not effective in treating other life-limiting symptom areas. As a result, individuals with schizophrenia continue to face significant challenges, including disability, unemployment, homelessness, and social isolation. Genome-wide association studies of schizophrenia have been effective at identifying individual SNPs and genes that contribute to these phenotypes but have struggled to immediately uncover the bigger picture of the underlying biology of the disorder. Here we take an individual gene associated with schizophrenia risk, FOXP1, which is an important regulator of brain development. Using functional genomics data from models where FOXP1 has been disrupted, we identified sets of genes regulated by FOXP1 across different developmental stages, focusing on the cortical region of the brain. Our findings reveal that FOXP1-regulated genes are most strongly associated with schizophrenia during early childhood and adolescence. These genes are associated with synaptic components, including presynaptic and postsynaptic structures. By integrating developmental timing, cell type specificity, and functional pathways, this study provides valuable insights into the molecular mechanisms underlying schizophrenia.

genomics↗

bioRxiv2024

Sex differences in deep brain shape and asymmetry persist across schizophrenia and healthy individuals: A meta-analysis from the ENIGMA-Schizophrenia Working Group

BackgroundSchizophrenia (SCZ) is characterized by a disconnect from reality that manifests as various clinical and cognitive symptoms, and persistent neurobiological abnormalities. Sex-related differences in clinical presentation imply separate brain substrates. The present study characterized deep brain morphology using shape features to understand the independent effects of diagnosis and sex on the brain, and to determine whether the neurobiology of schizophrenia varies as a function of sex. MethodsThis study analyzed multi-site archival data from 1,871 male (M) and 955 female (F) participants with SCZ, and 2,158 male and 1,877 female healthy controls (CON) from twenty-three cross-sectional samples from the ENIGMA Schizophrenia Workgroup. Harmonized shape analysis protocols were applied to each sites data for seven deep brain regions obtained from T1-weighted structural MRI scans. Effect sizes were calculated for the following main contrasts: 1) Sex effects;2) Diagnosis-by-Sex interaction; 3) within sex tests of diagnosis; 4) within diagnosis tests of sex differences. Meta-regression models between brain structure and clinical variables were also computed separately in men and women with schizophrenia. ResultsMass univariate meta-analyses revealed more concave-than-convex shape differences in all regions for women relative to men, across diagnostic groups (d = -0.35 to 0.20, SE = 0.02 to 0.07); there were no significant diagnosis-by-sex interaction effects. Within men and women separately, we identified more-concave-than-convex shape differences for the hippocampus, amygdala, accumbens, and thalamus, with more-convex-than-concave differences in the putamen and pallidum in SCZ (d = -0.30 to 0.30, SE = 0.03 to 0.10). Within CON and SZ separately, we found more-concave-than-convex shape differences in the thalamus, pallidum, putamen, and amygdala among females compared to males, with mixed findings in the hippocampus and caudate (d = -0.30 to 0.20, SE = 0.03 to 0.09). Meta-regression models revealed similarly small, but significant relationships, with medication and positive symptoms in both SCZ-M and SCZ-F. ConclusionsSex-specific variation is an overriding feature of deep brain shape regardless of disease status, underscoring persistent patterns of sex differences observed both within and across diagnostic categories, and highlighting the importance of including it as a critical variable in studies of neurobiology. Future work should continue to explore these dimensions independently to determine whether these patterns of brain morphology extend to other aspects of neurobiology in schizophrenia, potentially uncovering broader implications for diagnosis and treatment. Key PointsO_LIStatistical analyses revealed significant main effects for diagnosis and sex in deep brain shape morphology. Among patients with schizophrenia, there was a pattern of thinning and surface contraction in the bilateral hippocampus, amygdala, accumbens, and thalamus, and a pattern of significant thickening and surface expansion in the bilateral putamen and pallidum compared to healthy control participants. Between males and females, there was a pattern of significant thinning and surface contraction in the bilateral thalamus, pallidum, putamen, and amygdala in females compared to males. C_LIO_LIThere was no significant interaction between diagnosis and biological sex, suggesting that sex differences in deep brain shape and asymmetry among patients with schizophrenia reflect those observed in healthy individuals. C_LIO_LISmall but statistically significant relationships exist between brain structure and clinical correlates of schizophrenia were similar for both men and women with the disease, such that higher CPZ was associated with shape-derived thinning and surface contraction in the caudate, accumbens, hippocampus, amygdala, and thalamus, and elevated positive symptoms were associated with shape-derived thinning and surface contraction in the bilateral caudate, right hippocampus, and right amygdala. C_LI

neuroscience↗

bioRxiv2023

Direct targets of MEF2C are enriched for genes associated with schizophrenia and cognitive function and are involved in neurogenesis and mitochondrial function

Myocyte Enhancer Factor 2C (MEF2C) is a transcription factor that plays a crucial role in neurogenesis and synapse development. Genetic studies have identified MEF2C as a gene that influences cognition and risk for neuropsychiatric disorders, including autism spectrum disorder (ASD) and schizophrenia (SCZ). Here, we investigated the involvement of MEF2C in these phenotypes using human-derived neural stem cells (NSCs) and induced neurons (iNs), which represented early and late neurodevelopmental stages. For these cellular models, MEF2C function had previously been disrupted, either by direct or indirect mutation, and gene expression assayed using RNA-seq. We integrated these RNA-seq data with MEF2C ChIP-seq data to identify dysregulated direct target genes of MEF2C in the NSCs and iNs models. Several MEF2C direct target gene-sets were enriched for SNP-based heritability for intelligence, educational attainment and SCZ, as well as being enriched for genes containing rare de novo mutations reported in ASD and/or developmental disorders. These gene-sets are enriched in both excitatory and inhibitory neurons in the hippocampus and cortex and are involved in a wide range of biological processes including neurogenesis, metabolic processes, protein modifications, as well as mitochondrial function and energy production. We observed a trans expression quantitative trait locus (eQTL) effect of a single SNP at MEF2C (rs6893807, which is associated with IQ) on the expression of a target gene, BNIP3L. BNIP3L is a prioritized risk gene from the largest genome-wide association study of SCZ and has a function in mitophagy in mitochondria. Overall, our analysis reveals that either direct or indirect disruption of MEF2C dysregulates sets of genes that contain multiple alleles associated with SCZ risk and cognitive function and implicates neurogenesis and mitochondrial function in the etiology of these phenotypes. Author SummarySchizophrenia is a complex disorder caused by many genes. Current drugs for schizophrenia are only partially effective and do not treat cognitive deficits, which are key factors for explaining disability, leading to unemployment, homelessness and social isolation. Genome-wide association studies of schizophrenia and cognitive function have been effective at identifying individual SNPs and genes that contribute to these phenotypes but have struggled to immediately uncover the bigger picture of the underlying biology of the disorder. Here we take an individual gene associated with schizophrenia and cognitive function called MEF2C, which on its own is a very important regulator of brain development. We use functional genomics data from studies where MEF2C has been mutated to identify sets of other genes that are influenced by MEF2C in developing and mature neurons. We show that several of these gene-sets are enriched for common variants associated with schizophrenia and cognitive function, and for rare variants that increase risk of various neurodevelopmental disorders. These gene-sets are involved in neurogenesis and mitochondrial function, providing evidence that these biological processes may be important in the context of the molecular mechanisms that underpin schizophrenia and cognitive function.

genetics↗

bioRxiv2023

Social isolation-induced transcriptomic changes in mouse hippocampus impact the synapse and show convergence with human genetic risk for neurodevelopmental phenotypes

Early life stress (ELS) can impact brain development and is a risk factor for neurodevelopmental disorders such as schizophrenia. Post-weaning social isolation (SI) is used to model ELS in animals, using isolation stress to disrupt a normal developmental trajectory. We aimed to investigate how SI affects the expression of genes in mouse hippocampus and to investigate how these changes related to the genetic basis of neurodevelopmental phenotypes. BL/6J mice were exposed to post-weaning SI (PD21-25) or treated as group-housed controls (n = 7-8 per group). RNA sequencing was performed on tissue samples from the hippocampus of adult male and female mice. Four hundred and 1,215 differentially-expressed genes (DEGs) at a false discovery rate of < 0.05 were detected between SI and control samples for males and females respectively. DEGS for both males and females were significantly overrepresented in gene ontologies related to synaptic structure and function, especially the post-synapse. DEGs were enriched for common variant (SNP) heritability in humans that contributes to risk of neuropsychiatric disorders (schizophrenia, bipolar disorder) and to cognitive function. DEGs were also enriched for genes harbouring rare de novo variants that contribute to autism spectrum disorder and other developmental disorders. Finally, cell type analysis revealed populations of hippocampal astrocytes that were enriched for DEGs, indicating effects in these cell types as well as neurons. Overall, these data suggest a convergence between genes dysregulated by the SI stressor in the mouse and genes associated with neurodevelopmental disorders and cognitive phenotypes in humans. Author SummaryEarly life stress increases risk of developing neuropsychiatric and neurodevelopmental disorders. Early life stress can be modelled in animals using social isolation (SI) where animals are separated from others after they have stopped weaning and are housed individually rather than in groups. Here, we investigated the effect of SI on gene expression in the hippocampus, a brain region that regulates stress response and emotion, and how this relates to the known genetic aetiology of neuropsychiatric disorders and traits such as cognitive function. We found that genes altered by SI play a role in how synapses form and function - these are the connection points between nerve cells in the brain. We also found these altered genes are also the genes where common changes in the DNA code can increase risk for schizophrenia, bipolar disorder and influence cognitive ability, and where rare changes in the DNA code increase risk for autism spectrum disorder and developmental disorder. Finally, these genes altered by SI are also highly expressed in astrocytes, cells that help nerve cells to function in the hippocampus. Overall, these data suggest a convergence between genes affected by SI, the environmental stressor, and the genes associated with neurodevelopmental disorders and cognition.

genomics↗

bioRxiv2022

Associations between methylation age and brain age in late adolescence

Recent research suggests that biological age, based on DNA methylation or neuroimaging measures, may predict health traits in adulthood more accurately than chronological age. However, whether these findings apply to earlier stages in life is unknown. We therefore aimed to characterise the performance of and interdependence between measures of biological age during adolescence, leveraging longitudinal data from a subsample of young adolescents from the population-based ALSPAC cohort (n=386). We derived four methylation age measures in late adolescence (17-19 years) and a measure of brain age derived from structural neuroimaging data (18-24 years). We then examined associations between these measures of biological age, and their relationship with five measures of physical, cognitive and mental health (8-18 years). Brain age was largely independent of different measures of methylation age, after accounting for age, cell type composition, array and study (beta range: -0.60 to 0.60, all p>0.05). Smoking and BMI were associated with three measures of advanced methylation age (beta range: -0.35 to 0.52, all p<0.05), but not brain age. Depressive symptoms and cognitive ability were unrelated to all measures of biological age. Our findings emphasize the variability of and independence between these methylation- and brain-based measures of age in adolescents. They also highlight the importance of tracking the mosaic of ageing in younger populations.

neuroscience↗

bioRxiv2021

Identification of 13 independent genetic loci associated with cognitive resilience in healthy aging in 330,097 individuals in the UK Biobank.

Cognitive resilience is the ability to withstand the negative effects of stress on cognitive functioning and is important for maintaining quality of life while aging. UK Biobank (UKB) does not have direct measurements of the same cognitive phenotype at distal timepoints. Therefore, we used number of education years (EY) as a proxy phenotype for past cognitive performance. Current cognitive performance was determined based on processing speed. This approach captured an average time span of 40 years between past and current cognitive performance in 330,097 individuals. A confounding factor was that EY is highly polygenic and masked the genetics of resilience. To overcome this, we employed Genomics Structural Equation Modelling (GenomicSEM) to perform a GWAS-by-subtraction using two GWAS, one GWAS of EY and resilience and a second GWAS of EY but not resilience. Subtracting one from the other generated a GWAS of Resilience. Replication of this approach was shown using independent discovery and replication samples within UKB and the full UKB GWAS results show significant genetic correlation with a GWAS of cognitive change in the independent Health and Retirement Study (N=9,526; P=1.5x10-3). We found 13 independent genetic loci for Resilience. Functional analyses showed enrichment in several brain regions and involvement of specific cell types, including GABAergic neurons (P=6.59x10-8) and glutamatergic neurons (P=6.98x10-6) in the cortex. Gene-set analyses implicated the biological process "neuron differentiation" (P=9.7x10-7) and the cellular component "synaptic part" (P=2.14x10-6). Mendelian randomization analysis showed a causative effect of white matter volume on cognitive resilience. These results enhance neurobiological understanding of resilience.

genomics↗

bioRxiv2020

Large-scale analysis of DNA methylation identifies cellular alterations in blood from psychosis patients and molecular biomarkers of treatment-resistant schizophrenia.

ObjectivePsychosis - a complex and heterogeneous neuropsychiatric condition characterized by hallucinations and delusions - is a common feature of schizophrenia. There is evidence for altered DNA methylation (DNAm) associated with schizophrenia in both brain and peripheral tissues. We aimed to undertake a systematic analysis of variable DNAm associated with psychosis, schizophrenia, and treatment-resistant schizophrenia, also exploring measures of biological ageing, smoking, and blood cell composition derived from DNAm data to identify molecular biomarkers of disease. MethodsWe quantified DNAm across the genome in blood samples from 4,483 participants from seven case-control cohorts including patients with schizophrenia or first-episode psychosis. Measures of biological age, cellular composition and smoking status were derived from DNAm data using established algorithms. DNAm and derived measures were analyzed within each cohort and the results combined by meta-analysis. ResultsPsychosis cases were characterized by significant differences in measures of blood cell proportions and elevated smoking exposure derived from the DNAm data, with the largest differences seen in treatment-resistant schizophrenia patients. DNAm at 95 CpG sites was significantly different between psychosis cases and controls, with 1,048 differentially methylated positions (DMPs) identified between schizophrenia cases and controls. Schizophrenia-associated DMPs colocalize to regions identified in genetic association studies, with genes annotated to these sites enriched for pathways relevant to disease. Finally, a number of the schizophrenia associated differences were only present in the treatment-resistant schizophrenia subgroup. ConclusionsWe show that DNAm data can be leveraged to derive measures of blood cell counts and smoking that are strongly associated with psychosis. Our DNAm meta-analysis identified multiple DMPs associated with both psychosis and a more refined diagnosis of schizophrenia, with evidence for differential methylation associated with treatment-resistant schizophrenia that potentially reflects exposure to clozapine.

genomics↗

bioRxiv2019

Genes influenced by MEF2C contribute to neurodevelopmental disease via gene expression changes that affect multiple types of cortical excitatory neurons

Myocyte enhancer factor 2 C (MEF2C) is an important transcription factor during neurodevelopment. Mutation or deletion of MEF2C causes intellectual disability (ID) and common variants within MEF2C are associated with cognitive function and schizophrenia risk. We investigated if genes influenced by MEF2C during neurodevelopment are enriched for genes associated with neurodevelopmental phenotypes, and if this can be leveraged to identify biological mechanisms and individual brain cell types affected. We used a set of 1,052 genes that were differentially expressed in the adult mouse brain following early embryonic deletion of Mef2c in excitatory cortical neurons. Using GWAS data, we found these differentially expressed genes (DEGs) to be enriched for genes associated with schizophrenia, intelligence and educational attainment but not autism spectrum disorder (ASD). Using sequencing data from trios studies, we found these DEGs to be enriched for genes containing de novo mutations reported in ASD and ID, but not schizophrenia. Using single cell RNA-seq data, we identified that a number of different excitatory glutamatergic neurons in the cortex were enriched for these DEGs including deep layer pyramidal cells and cells in the retrosplenial cortex, entorhinal cortex and subiculum. These data indicate that genes influenced by MEF2C during neurodevelopment contribute to cognitive function and risk of neurodevelopmental disorders. Within excitatory neurons, common SNPs in these genes contribute to cognition and SZ risk via an effect on synaptic function based on gene ontology analysis. In contrast, rare mutations contribute to earlier onset ASD and ID via an effect on cell morphogenesis. Author SummarySchizophrenia is a complex disorder caused by many genes. Current drugs for schizophrenia are only partially effective and do not treat cognitive deficits, which are key factors for explaining disability. Here we take an individual gene identified in genetic studies of schizophrenia and cognition called MEF2C, which on its own is a very important regulator of brain development. We use data from a mouse study where MEF2C has been stopped from functioning or knocked out during brain development. The effect of that knock out has been measured when the mice reach adulthood, in the form of a set of differentially expressed genes (DEGs) from the somatosensory cortex. We found that this set of DEGs contains more genes than expected by chance that are associated with schizophrenia and cognition or contain rare new (de novo) mutations reported in autism and intellectual disability. Using gene expression data from single brain cells, we identified that a number of specific excitatory glutamatergic neurons in the cortex were enriched for these DEGs. This study provides evidence that the molecular mechanisms that underpin schizophrenia and cognitive function include disruption of cell types influenced by MEF2C.

genomics↗

bioRxiv2019

10Kin1day: A bottom-up neuroimaging initiative

We organized 10Kin1day, a pop-up scientific event with the goal to bring together neuroimaging groups from around the world to jointly analyze 10,000+ existing MRI connectivity datasets during a 3-day workshop. In this report, we describe the motivation and principles of 10Kin1day, together with a public release of 8,000+ MRI connectome maps of the human brain.

neuroscience↗

bioRxiv2018

Cortical patterning of abnormal morphometric similarity in psychosis is associated with brain expression of schizophrenia related genes

Schizophrenia has been conceived as a disorder of brain connectivity but it is unclear how this network phenotype is related to the emerging genetics. We used morphometric similarity analysis of magnetic resonance imaging (MRI) data as a marker of inter-areal cortical connectivity in three prior case-control studies of psychosis: in total, N=185 cases and N=227 controls. Psychosis was associated with globally reduced morphometric similarity (MS) in all 3 studies. There was also a replicable pattern of case-control differences in regional MS which was significantly reduced in patients in frontal and temporal cortical areas, but increased in parietal cortex. Using prior brain-wide gene expression data, we found that the cortical map of case-control differences in MS was spatially correlated with cortical expression of a weighted combination of genes enriched for neu-robiologically relevant ontology terms and pathways. In addition, genes that were normally over-expressed in cortical areas with reduced MS were significantly up-regulated in a prior post mortem study of schizophrenia. We propose that this combination of neuroimaging and transcriptional data provides new insight into how previously implicated genes and proteins, as well as a number of unreported proteins in their vicinity on the protein interaction network, may interact to drive structural brain network changes in schizophrenia.

neuroscience↗

bioRxiv2018

Deep Learning for Quality Control of Subcortical Brain 3D Shape Models

We present several deep learning models for assessing the morphometric fidelity of deep grey matter region models extracted from brain MRI. We test three different convolutional neural net architectures (VGGNet, ResNet and Inception) over 2D maps of geometric features. Further, we present a novel geometry feature augmentation technique based on parametric spherical mapping. Finally, we present an approach for model decision visualization, allowing human raters to see the areas of subcortical shapes most likely to be deemed of failing quality by the machine. Our training data is comprised of 5200 subjects from the ENIGMA Schizophrenia MRI cohorts, and our test dataset contains 1500 subjects from the ENIGMA Major Depressive Disorder cohorts. Our final models reduce human rater time by 46-70%. ResNet outperforms VGGNet and Inception for all of our predictive tasks.

neuroscience↗

bioRxiv2018

Association between schizophrenia and both loss of function and missense mutations in paralog conserved sites of voltage-gated sodium channels

Sequencing studies have highlighted candidate sets of genes involved in schizophrenia, including activity-regulated cytoskeleton-associated protein (ARC) and N-methyl-d-aspartate receptor (NMDAR) complexes. Two genes, SETD1A and RBM12, have also been associated with robust statistical evidence. Larger samples and novel methods for identifying disease-associated missense variants are needed to reveal novel genes and biological mechanisms associated with schizophrenia. We sequenced 187 genes, selected for prior evidence of association with schizophrenia, in a new dataset of 5,207 cases and 4,991 controls. Included were members of ARC and NMDAR post-synaptic protein complexes, as well as voltage-gated sodium and calcium channels. We observed a significant case excess of rare (<0.1% in frequency) loss-of-function (LoF) mutations across all 187 genes (OR = 1.36; Pcorrected = 0.0072) but no individual gene was associated with schizophrenia after correcting for multiple testing. We found novel evidence that LoF and missense variants at paralog conserved sites were enriched in sodium channels (OR = 1.26; P = 0.0035). Meta-analysis of our new data with published sequencing data (11,319 cases, 15,854 controls and 1,136 trios) supported and refined this association to sodium channel alpha subunits (P = 0.0029). Meta-analysis also confirmed association between schizophrenia and rare variants in ARC (P = 4.0 x 10-4) and NMDAR (P = 1.7 x 10-5) synaptic genes. No association was found between rare variants in calcium channels and schizophrenia.\n\nIn one of the largest sequencing studies of schizophrenia to date, we provide novel evidence that multiple voltage-gated sodium channels are involved in schizophrenia pathogenesis, and increase the evidence for association between rare variants in ARC and NMDAR post-synaptic complexes and schizophrenia. Larger samples are required to identify specific genes and variants driving these associations.\n\nAuthor SummaryCommon and rare genetic variations are known to play a substantial role in the development of schizophrenia. Recently, sequencing studies have started to highlight specific sets of genes that are enriched for rare variation in schizophrenia, such as the synaptic gene sets ARC and NMDAR, as well as voltage-gated sodium and calcium channels. To confirm the role of these gene sets in schizophrenia, and identify specific risk genes, we sequenced 187 genes in a new sample of 5,207 schizophrenia cases and 4,991 controls. We find an excess of protein truncating mutations with a frequency <0.1% in all 187 targeted genes, and provide novel evidence that mutations altering amino acids conserved across sodium channel proteins are risk factors for schizophrenia. Through meta-analysing our new data with previously published sequencing data sets, for a total of 11,319 cases, 15,854 controls and 1,136 trios, we increase the evidence for association between rare coding variants and schizophrenia in voltage-gated sodium channels, as well as in synaptic gene sets ARC and NMDAR. Although no individual gene was associated with schizophrenia, these findings suggest larger studies will identify the specific genes driving these associations.

genetics↗

bioRxiv2017

Machine Learning for Large-Scale Quality Control of 3D Shape Models in Neuroimaging

As very large studies of complex neuroimaging phenotypes become more common, human quality assessment of MRI-derived data remains one of the last major bottlenecks. Few attempts have so far been made to address this issue with machine learning. In this work, we optimize predictive models of quality for meshes representing deep brain structure shapes. We use standard vertex-wise and global shape features computed homologously across 19 cohorts and over 7500 human-rated subjects, training kernelized Support Vector Machine and Gradient Boosted Decision Trees classifiers to detect meshes of failing quality. Our models generalize across datasets and diseases, reducing human workload by 30-70%, or equivalently hundreds of human rater hours for datasets of comparable size, with recall rates approaching inter-rater reliability.

neuroscience↗