Stable biomarker identification for predicting schizophrenia in the human connectome

Gutiérrez-Gómez, Leonardo; Vohryzek, Jakub; Chiêm, Benjamin; Baumann, Philipp S.; Conus, Philippe; Cuenod, K. Do; Hagmann, Patric; Delvenne, Jean‐Charles

doi:10.1016/j.nicl.2020.102316

Cited by 24 publications

(14 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our model had a mean accuracy of 85%, which is slightly better than those reported in recent machine-learning studies based on brain-wide FC: 82.4% [ 15 ], 81.74% [ 16 ], and 82.61% [ 41 ]. Notably, the performance of these SVMs was highly consistent—between 80 and 85%—suggesting that brain-wide FC is a reliable feature for automatic classification of patients with schizophrenic disorder.…”

Section: Discussioncontrasting

confidence: 60%

Evaluating the performance of machine learning models for automatic diagnosis of patients with schizophrenia based on a single site dataset of 440 participants

et al. 2021

View full text Add to dashboard Cite

Background Support vector machines (SVMs) based on brain-wise functional connectivity (FC) have been widely adopted for single-subject prediction of patients with schizophrenia, but most of them had small sample size. This study aimed to evaluate the performance of SVMs based on a large single-site dataset and investigate the effects of demographic homogeneity and training sample size on classification accuracy. Methods The resting functional Magnetic Resonance Imaging (fMRI) dataset comprised 220 patients with schizophrenia and 220 healthy controls. Brain-wise FCs was calculated for each participant and linear SVMs were developed for automatic classification of patients and controls. First, we evaluated the SVMs based on all participants and homogeneous subsamples of men, women, younger (18–30 years), and older (31–50 years) participants by 10-fold nested cross-validation. Then, we hold out a fixed test set of 40 participants (20 patients and 20 controls) and evaluated the SVMs based on incremental training sample sizes (N = 40, 80, …, 400). Results We found that the SVMs based on all participants had accuracy of 85.05%. The SVMs based on male, female, young, and older participants yielded accuracy of 84.66, 81.56, 80.50, and 86.13%, respectively. Although the SVMs based on older subsamples had better performance than those based on all participants, they generalized poorly to younger participants (77.24%). For incremental training sizes, the classification accuracy increased stepwise from 72.6 to 83.3%, with >80% accuracy achieved with sample size >240. Conclusions The findings indicate that SVMs based on a large dataset yield high classification accuracy and establish models using a large sample size with heterogeneous properties are recommended for single subject prediction of schizophrenia.

show abstract

Section: Discussioncontrasting

confidence: 60%

Evaluating the performance of machine learning models for automatic diagnosis of patients with schizophrenia based on a single site dataset of 440 participants

et al. 2021

View full text Add to dashboard Cite

show abstract

“…To screen for important variables and establish the optimal classification model, LASSO analysis uses an L1-penalty (lambda) to set the coefficients of less significant variables to zero. It uses supervised machine learning to categorize data points by maximizing the distance between classes in a high-dimensional space ( 13 ). RF is a non-parametric classification method ( 14 ), which includes decision trees based on divided data sets.…”

Section: Introductionmentioning

confidence: 99%

Single-Cell Sequencing Analysis and Multiple Machine Learning Methods Identified G0S2 and HPSE as Novel Biomarkers for Abdominal Aortic Aneurysm

Xiong

et al. 2022

Front. Immunol.

View full text Add to dashboard Cite

Identifying biomarkers for abdominal aortic aneurysms (AAA) is key to understanding their pathogenesis, developing novel targeted therapeutics, and possibly improving patients outcomes and risk of rupture. Here, we identified AAA biomarkers from public databases using single-cell RNA-sequencing, weighted co-expression network (WGCNA), and differential expression analyses. Additionally, we used the multiple machine learning methods to identify biomarkers that differentiated large AAA from small AAA. Biomarkers were validated using GEO datasets. CIBERSORT was used to assess immune cell infiltration into AAA tissues and investigate the relationship between biomarkers and infiltrating immune cells. Therefore, 288 differentially expressed genes (DEGs) were screened for AAA and normal samples. The identified DEGs were mostly related to inflammatory responses, lipids, and atherosclerosis. For the large and small AAA samples, 17 DEGs, mostly related to necroptosis, were screened. As biomarkers for AAA, G0/G1 switch 2 (G0S2) (Area under the curve [AUC] = 0.861, 0.875, and 0.911, in GSE57691, GSE47472, and GSE7284, respectively) and for large AAA, heparinase (HPSE) (AUC = 0.669 and 0.754, in GSE57691 and GSE98278, respectively) were identified and further verified by qRT-PCR. Immune cell infiltration analysis revealed that the AAA process may be mediated by T follicular helper (Tfh) cells and the large AAA process may also be mediated by Tfh cells, M1, and M2 macrophages. Additionally, G0S2 expression was associated with neutrophils, activated and resting mast cells, M0 and M1 macrophages, regulatory T cells (Tregs), resting dendritic cells, and resting CD4 memory T cells. Moreover, HPSE expression was associated with M0 and M1 macrophages, activated and resting mast cells, Tregs, and resting CD4 memory T cells. Additional, G0S2 may be an effective diagnostic biomarker for AAA, whereas HPSE may be used to confer risk of rupture in large AAAs. Immune cells play a role in the onset and progression of AAA, which may improve its diagnosis and treatment.

show abstract

“…A common method for characterizing global brain function, involves assessing how activity is temporally correlated across spatially separate brain areas over an entire recording period, defining static and state-specific 'functional connectomes' (Bullmore and Sporns, 2009;Amico et al, 2017;Gutiérrez-Gómez et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

Brain dynamics predictive of response to psilocybin for treatment-resistant depression

Vohryzek

Cabral

Lord

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Psilocybin therapy for depression has started to show promise, yet the underlying causal mechanisms are not currently known. Here we leveraged the differential outcome in responders and non-responders to psilocybin (10mg and 25mg, 7 days apart) therapy for depression - to gain new insights into regions and networks implicated in the restoration of healthy brain dynamics. We used whole-brain modelling to fit the spatiotemporal brain dynamics at rest in both responders and non-responders before treatment. Dynamic sensitivity analysis of systematic perturbation of these models enabled us to identify specific brain regions implicated in a transition from a depressive brain state to a heathy one. Binarizing the sample into treatment responders (>50% reduction in depressive symptoms) versus non-responders enabled us to identify a subset of regions implicated in this change. Interestingly, these regions correlate with in vivo density maps of serotonin receptors 5-HT2A and 5-HT1A, which psilocin, the active metabolite of psilocybin, has an appreciable affinity for, and where it acts as a full-to-partial agonist. Serotonergic transmission has long been associated with depression and our findings provide causal mechanistic evidence for the role of brain regions in the recovery from depression via psilocybin.

show abstract

Stable biomarker identification for predicting schizophrenia in the human connectome

Cited by 24 publications

References 41 publications

Evaluating the performance of machine learning models for automatic diagnosis of patients with schizophrenia based on a single site dataset of 440 participants

Evaluating the performance of machine learning models for automatic diagnosis of patients with schizophrenia based on a single site dataset of 440 participants

Single-Cell Sequencing Analysis and Multiple Machine Learning Methods Identified G0S2 and HPSE as Novel Biomarkers for Abdominal Aortic Aneurysm

Brain dynamics predictive of response to psilocybin for treatment-resistant depression

Contact Info

Product

Resources

About