Optimizing graph neural network architectures for schizophrenia spectrum disorder prediction using evolutionary algorithms.

BACKGROUND AND OBJECTIVE: The accurate diagnosis of schizophrenia spectrum disorder plays an important role in improving patient outcomes, enabling timely interventions, and optimizing treatment plans. Functional connectivity analysis, utilizing functional magnetic resonance imaging data, has been demonstrated to offer invaluable biomarkers conducive to clinical diagnosis. However, previous studies mainly focus on traditional machine learning methods or hand-crafted neural networks, which may not fully capture the spatial topological relationship between brain regions. METHODS: This paper proposes an evolutionary algorithm (EA) based graph neural architecture search (GNAS) method. EA-GNAS has the ability to search for high-performance graph neural networks for schizophrenia spectrum disorder prediction. Moreover, we adopt GNNExplainer to investigate the explainability of the acquired architectures, ensuring that the model's predictions are both accurate and comprehensible. RESULTS: The results suggest that the graph neural network model, derived using genetic algorithm search, outperforms under five-fold cross-validation, achieving a fitness of 0.1850. Relative to conventional machine learning and other deep learning approaches, the proposed method yields superior accuracy, F1 score, and AUC values of 0.8246, 0.8438, and 0.8258, respectively. CONCLUSION: Based on a multi-site dataset from schizophrenia spectrum disorder patients, the findings reveal an enhancement over prior methods, advancing our comprehension of brain function and potentially offering a biomarker for diagnosing schizophrenia spectrum disorder.

Molecular diagnostic yield of exome sequencing in a Chinese cohort of 512 fetuses with anomalies.

BACKGROUND: Currently, whole exome sequencing has been performed as a helpful complement in the prenatal setting in case of fetal anomalies. However, data on its clinical utility remain limited in practice. Herein, we reported our data of fetal exome sequencing in a cohort of 512 trios to evaluate its diagnostic yield. METHODS: In this retrospective cohort study, the couples performing prenatal exome sequencing were enrolled. Fetal phenotype was classified according to ultrasound and magnetic resonance imaging findings. Genetic variants were analyzed based on a phenotype-driven followed by genotype-driven approach in all trios. RESULTS: A total of 97 diagnostic variants in 65 genes were identified in 69 fetuses, with an average detection rate of 13.48%. Skeletal and renal system were the most frequently affected organs referred for whole exome sequencing, with the highest diagnostic rates. Among them, short femur and kidney cyst were the most common phenotype. Fetal growth restriction was the most frequently observed phenotype with a low detection rate (4.3%). Exome sequencing had limited value in isolated increased nuchal translucency and chest anomalies. CONCLUSIONS: This study provides our data on the detection rate of whole exome sequencing in fetal anomalies in a large cohort. It contributes to the expanding of phenotypic and genotypic spectrum.

miR-1290 induces endothelial-to-mesenchymal transition and promotes vascular restenosis after angioplasty by targeting FGF2.

BACKGROUND AND AIMS: Endothelial-to-mesenchymal transition (EndMT) is an important reason for restenosis but the underlying mechanisms need to be further explored. Therefore, the purpose of this study is to screen significantly different microRNAs (miRNAs) and assess their functions and downstream pathways. METHODS: This study screened several miRNAs with significant differences between human arterial segments from restenosis patients and healthy volunteers using whole transcriptome resequencing and real-time quantitative reverse transcription PCR (qRT-PCR). We explored the correlation between miR-1290 and EndMT using Western blot, qRT-PCR, Pearson correlation analysis and further functional gain and loss experiments. Subsequently, we identified the direct downstream target of miR-1290 by bioinformatics analysis, RNA pull-down, double Luciferase reporter gene and other functional experiments. Finally, rat carotid artery balloon injury model demonstrated the therapeutic potential of miR-1290 regulator. RESULTS: We screened 129 differentially expressed miRNAs. Among them, miR-1290 levels were significantly higher in restenosis arteries than in healthy arteries, and as expected, EndMT was functionally enhanced with miR-1290 overexpression and comparatively weakened when miR-1290 was knocked down. In addition, fibroblast growth factor-2 (FGF2) was established as the downstream target of miR-1290. Finally, we utilized an animal model and found that low miR-1290 levels could alleviate EndMT and the progression of restenosis. CONCLUSIONS: Our study demonstrated the strong regulatory effects of miR-1290 on EndMT, endometrial hyperplasia and restenosis, which could be useful as biomarker and therapeutic target for stent implantation in patients with arterial occlusive disease of the lower extremities.

Early prediction of dementia using fMRI data with a graph convolutional network approach.

Objective. Alzheimer's disease is a progressive neurodegenerative dementia that poses a significant global health threat. It is imperative and essential to detect patients in the mild cognitive impairment (MCI) stage or even earlier, enabling effective interventions to prevent further deterioration of dementia. This study focuses on the early prediction of dementia utilizing Magnetic Resonance Imaging (MRI) data, using the proposed Graph Convolutional Networks (GCNs).Approach. Specifically, we developed a functional connectivity (FC) based GCN framework for binary classifications using resting-state fMRI data. We explored different types and processing methods of FC and evaluated the performance on the OASIS-3 dataset. We developed the GCN model for two different purposes: (1) MCI diagnosis: classifying MCI from normal controls (NCs); and (2) dementia risk prediction: classifying NCs from subjects who have the potential for developing MCI but have not been clinically diagnosed as MCI.Main results. The results of the experiments revealed several important findings: First, the proposed GCN outperformed both the baseline GCN and Support Vector Machine (SVM). It achieved the best average accuracy of 80.3% (11.7% higher than the baseline GCN and 23.5% higher than SVM) and the highest accuracy of 91.2%. Secondly, the GCN framework with (absolute) individual FC performed slightly better than that with global FC generally. However, GCN using global graphs with appropriate connectivity can achieve equivalent or superior performance to individual graphs in some cases, which highlights the significance of suitable connectivity for achieving performance. Additionally, the results indicate that the self-network connectivity of specific brain network regions (such as default mode network, visual network, ventral attention network and somatomotor network) may play a more significant role in GCN classification.Significance. Overall, this study offers valuable insights into the application of GCNs in brain analysis and early diagnosis of dementia. This contributes significantly to the understanding of MCI and has substantial potential for clinical applications in early diagnosis and intervention for dementia and other neurodegenerative diseases. Our code for GCN implementation is available at:https://github.com/Shuning-Han/FC-based-GCN.