👤 Emily Ungermann

Early-onset psychosis presents diagnostic challenges due to overlapping clinical presentations and complex comorbidities, typically requiring specialized tertiary care with extensive neuroimaging, neuropsychometric testing, and multidisciplinary evaluation. This case-control study investigated whether machine learning could integrate multiple diagnostic modalities to create an objective diagnostic framework for early-onset psychosis. We recruited 45 patients with early-onset psychosis and 34 healthy controls from a tertiary referral centre. Participants underwent comprehensive assessment including serum protein biomarker analysis (brain-derived neurotrophic factor, proBDNF, p75 neurotrophin receptor, S100B), neuropsychometric testing (Iowa Gambling Task, Simple Response Time, Zabor Verbal Task), and demographic evaluation. Four machine learning algorithms (logistic regression, support vector machine, random forest, XGBoost) were trained on five feature combinations using nested cross-validation with hyperparameter optimization. XGBoost demonstrated superior performance, achieving optimal classification with the complete multimodal dataset (accuracy: 0.91 ± 0.08, precision: 0.92 ± 0.08, area under curve: 0.97 ± 0.04). Feature importance analysis revealed cognitive measures, particularly Zabor Verbal Task errors and response time parameters, as most discriminative, with brain-derived neurotrophic factor pathway components showing highest biomarker importance. Machine learning effectively integrated neuropsychometric and protein biomarker data for high-accuracy early-onset psychosis classification, with multimodal approaches outperforming single-domain assessments. Show less

Transport along the endolysosomal system requires multiple fusion events at early and late endosomes. Deletion of several endosomal fusion factors, including the Vac1 tether and the Class C core vacuole/endosome tethering (CORVET) complex-specific subunits Vps3 and Vps8, results in a class D vps phenotype. As these mutants have an apparently similar defect in endosomal transport, we asked whether CORVET and Vac1 could still act in distinct tethering reactions. Our data reveal that CORVET mutants can be rescued by Vac1 overexpression in the endocytic pathway but not in CPY or Cps1 sorting to the vacuole. Moreover, when we compared the ultrastructure, CORVET mutants were most similar to deletions of the Rab Vps21 and its guanine nucleotide exchange factor Vps9 and different from vac1 deletion, indicating separate functions. Likewise, CORVET still localized to endosomes even in the absence of Vac1, whereas Vac1 localization became diffuse in CORVET mutants. Importantly, CORVET localization requires the Rab5 homologs Vps21 and Ypt52, whereas Vac1 localization is strictly Vps21-dependent. In this context, we also uncover that Muk1 can compensate for loss of Vps9 in CORVET localization, indicating that two Rab5 guanine nucleotide exchange factors operate in the endocytic pathway. Overall, our study reveals a unique role of CORVET in the sorting of biosynthetic cargo to the vacuole/lysosome. Show less

The human Batten disease gene CLN3 and yeast orthologue BTN1 encode proteins of unclear function. We show that the loss of BTN1 phenocopies that of BTN2, which encodes a retromer accessory protein involved in the retrieval of specific cargo from late endosomes (LEs) to the Golgi. However, Btn1 localizes to Golgi and regulates soluble N-ethyl-maleimide sensitive fusion protein attachment protein receptor (SNARE) function to control retrograde transport. Specifically, BTN1 overexpression and deletion have opposing effects on phosphorylation of the Sed5 target membrane SNARE, on Golgi SNARE assembly, and on Golgi integrity. Although Btn1 does not interact physically with SNAREs, it regulates Sed5 phosphorylation by modulating Yck3, a palmitoylated endosomal kinase. This may involve modification of the Yck3 lipid anchor, as substitution with a transmembrane domain suppresses the deletion of BTN1 and restores trafficking. Correspondingly, deletion of YCK3 mimics that of BTN1 or BTN2 with respect to LE-Golgi retrieval. Thus, Btn1 controls retrograde sorting by regulating SNARE phosphorylation and assembly, a process that may be adversely affected in Batten Disease patients. Show less