Early detection of myocardial involvement by non-contrast T1ρ mapping of cardiac magnetic resonance in type 2 diabetes mellitus.

Objective: This study aims to determine the effectiveness of T1ρ in detecting myocardial fibrosis in type 2 diabetes mellitus (T2DM) patients by comparing with native T1 and extracellular volume (ECV) fraction. Methods: T2DM patients (n = 35) and healthy controls (n = 30) underwent cardiac magnetic resonance. ECV, T1ρ, native T1, and global longitudinal strain (GLS) values were assessed. Diagnostic performance was analyzed using receiver operating curves. Results: The global ECV and T1ρ of T2DM group (ECV = 32.1 ± 3.2%, T1ρ = 51.6 ± 3.8 msec) were significantly higher than those of controls (ECV = 26.2 ± 1.6%, T1ρ = 46.8 ± 2.0 msec) (all P < 0.001), whether there was no significant difference in native T1 between T2DM and controls (P = 0.264). The GLS decreased significantly in T2DM patients compared with controls (-16.5 ± 2.4% vs. -18.3 ± 2.6%, P = 0.015). The T1ρ and native T1 were associated with ECV (Pearson's r = 0.50 and 0.25, respectively, both P < 0.001); the native T1, T1ρ, and ECV were associated with hemoglobin A1c (Pearson's r = 0.41, 0.52, and 0.61, respectively, all P < 0.05); and the ECV was associated with diabetes duration (Pearson's r = 0.41, P = 0.016). The AUC of ECV, T1ρ, GLS, and native T1 were 0.869, 0.810, 0.659, and 0.524, respectively. Conclusion: In T2DM patients, T1ρ may be a new non-contrast cardiac magnetic resonance technique for identifying myocardial diffuse fibrosis, and T1ρ may be more sensitive than native T1 in the detection of myocardial diffuse fibrosis.

Myocardial involvement characteristics by cardiac MR imaging in neurological and non-neurological Wilson disease patients.

OBJECTIVES: To explore the characteristics of myocardial involvement in Wilson Disease (WD) patients by cardiac magnetic resonance (CMR). METHODS: We prospectively included WD patients and age- and sex-matched healthy population. We applied CMR to analyze cardiac function, strain, T1 maps, T2 maps, extracellular volume fraction (ECV) maps, and LGE images. Subgroup analyzes were performed for patients with WD with predominantly neurologic manifestations (WD-neuro +) or only hepatic manifestations (WD-neuro -). RESULTS: Forty-one WD patients (age 27.9 ± 8.0 years) and 40 healthy controls (age 25.4 ± 2.9 years) were included in this study. Compared to controls, the T1, T2, and ECV values were significantly increased in the WD group (T1 1085.1 ± 39.1 vs. 1046.5 ± 33.1 ms, T2 54.2 ± 3.3 ms vs. 51.5 ± 2.6 ms, ECV 31.8 ± 3.6% vs. 24.3 ± 3.7%) (all p < 0.001). LGE analysis revealed that LGE in WD patients was predominantly localized to the right ventricular insertion point and interventricular septum. Furthermore, the WD-neuro + group showed more severe myocardial damage compared to WD-neuro - group. The Unified Wilson Disease Rating Scale score was significantly correlated with ECV (Pearson's r = 0.64, p < 0.001). CONCLUSIONS: CMR could detect early myocardial involvement in WD patients without overt cardiac function dysfunction. Furthermore, characteristics of myocardial involvement were different between WD-neuro + and WD-neuro - , and myocardial involvement might be more severe in WD-neuro + patients. CRITICAL RELEVANCE STATEMENT: Cardiac magnetic resonance enables early detection of myocardial involvement in Wilson disease patients, contributing to the understanding of distinct myocardial characteristics in different subgroups and potentially aiding in the assessment of disease severity. KEY POINTS: • CMR detects WD myocardial involvement with increased T1, T2, ECV. • WD-neuro + patients show more severe myocardial damage and correlation with ECV. • Differences of myocardial characteristics exist between WD-neuro + and WD-neuro - patients.

Structural basis for FGF hormone signalling.

α/ßKlotho coreceptors simultaneously engage fibroblast growth factor (FGF) hormones (FGF19, FGF21 and FGF23)1,2 and their cognate cell-surface FGF receptors (FGFR1-4) thereby stabilizing the endocrine FGF-FGFR complex3-6. However, these hormones still require heparan sulfate (HS) proteoglycan as an additional coreceptor to induce FGFR dimerization/activation and hence elicit their essential metabolic activities6. To reveal the molecular mechanism underpinning the coreceptor role of HS, we solved cryo-electron microscopy structures of three distinct 1:2:1:1 FGF23-FGFR-αKlotho-HS quaternary complexes featuring the 'c' splice isoforms of FGFR1 (FGFR1c), FGFR3 (FGFR3c) or FGFR4 as the receptor component. These structures, supported by cell-based receptor complementation and heterodimerization experiments, reveal that a single HS chain enables FGF23 and its primary FGFR within a 1:1:1 FGF23-FGFR-αKlotho ternary complex to jointly recruit a lone secondary FGFR molecule leading to asymmetric receptor dimerization and activation. However, αKlotho does not directly participate in recruiting the secondary receptor/dimerization. We also show that the asymmetric mode of receptor dimerization is applicable to paracrine FGFs that signal solely in an HS-dependent fashion. Our structural and biochemical data overturn the current symmetric FGFR dimerization paradigm and provide blueprints for rational discovery of modulators of FGF signalling2 as therapeutics for human metabolic diseases and cancer.

Identifying spatial domain by adapting transcriptomics with histology through contrastive learning.

Recent advances in spatial transcriptomics have enabled measurements of gene expression at cell/spot resolution meanwhile retaining both the spatial information and the histology images of the tissues. Accurately identifying the spatial domains of spots is a vital step for various downstream tasks in spatial transcriptomics analysis. To remove noises in gene expression, several methods have been developed to combine histopathological images for data analysis of spatial transcriptomics. However, these methods either use the image only for the spatial relations for spots, or individually learn the embeddings of the gene expression and image without fully coupling the information. Here, we propose a novel method ConGI to accurately exploit spatial domains by adapting gene expression with histopathological images through contrastive learning. Specifically, we designed three contrastive loss functions within and between two modalities (the gene expression and image data) to learn the common representations. The learned representations are then used to cluster the spatial domains on both tumor and normal spatial transcriptomics datasets. ConGI was shown to outperform existing methods for the spatial domain identification. In addition, the learned representations have also been shown powerful for various downstream tasks, including trajectory inference, clustering, and visualization.

A parameter-free deep embedded clustering method for single-cell RNA-seq data.

Clustering analysis is widely used in single-cell ribonucleic acid (RNA)-sequencing (scRNA-seq) data to discover cell heterogeneity and cell states. While many clustering methods have been developed for scRNA-seq analysis, most of these methods require to provide the number of clusters. However, it is not easy to know the exact number of cell types in advance, and experienced determination is not always reliable. Here, we have developed ADClust, an automatic deep embedding clustering method for scRNA-seq data, which can accurately cluster cells without requiring a predefined number of clusters. Specifically, ADClust first obtains low-dimensional representation through pre-trained autoencoder and uses the representations to cluster cells into initial micro-clusters. The clusters are then compared in between by a statistical test, and similar micro-clusters are merged into larger clusters. According to the clustering, cell representations are updated so that each cell will be pulled toward centers of its assigned cluster and similar clusters, while cells are separated to keep distances between clusters. This is accomplished through jointly optimizing the carefully designed clustering and autoencoder loss functions. This merging process continues until convergence. ADClust was tested on 11 real scRNA-seq datasets and was shown to outperform existing methods in terms of both clustering performance and the accuracy on the number of the determined clusters. More importantly, our model provides high speed and scalability for large datasets.

A robust and scalable graph neural network for accurate single-cell classification.

Single-cell RNA sequencing (scRNA-seq) techniques provide high-resolution data on cellular heterogeneity in diverse tissues, and a critical step for the data analysis is cell type identification. Traditional methods usually cluster the cells and manually identify cell clusters through marker genes, which is time-consuming and subjective. With the launch of several large-scale single-cell projects, millions of sequenced cells have been annotated and it is promising to transfer labels from the annotated datasets to newly generated datasets. One powerful way for the transferring is to learn cell relations through the graph neural network (GNN), but traditional GNNs are difficult to process millions of cells due to the expensive costs of the message-passing procedure at each training epoch. Here, we have developed a robust and scalable GNN-based method for accurate single-cell classification (GraphCS), where the graph is constructed to connect similar cells within and between labelled and unlabeled scRNA-seq datasets for propagation of shared information. To overcome the slow information propagation of GNN at each training epoch, the diffused information is pre-calculated via the approximate Generalized PageRank algorithm, enabling sublinear complexity over cell numbers. Compared with existing methods, GraphCS demonstrates better performance on simulated, cross-platform, cross-species and cross-omics scRNA-seq datasets. More importantly, our model provides a high speed and scalability on large datasets, and can achieve superior performance for 1 million cells within 50 min.

Cancer survival prognosis with Deep Bayesian Perturbation Cox Network.

BACKGROUND: The Cox proportional hazards model with neural networks is widely used to accurately predict survival outcome for choosing cancer treatment strategies. Although this method has shown outstanding performance in many tasks, it has encountered challenges when dealing with high-dimensional datasets. In this study, we point out that the Cox network has estimation bias in processing such datasets with a large number of censored samples. The estimation bias is composed of censored estimation bias and variance estimation bias, which limit the prediction performance of the model. In order to correct this bias, this paper proposes the Deep Bayesian Perturbation Cox Network (DBP), which introduces Bayesian prior knowledge about censored samples to optimize the training process of the neural network. Specifically, the model uses a sampling module called Bayesian Perturbation to approximate the prior knowledge, which can be used as a component for other Cox-based neural networks. RESULTS: The comparison between DBP and the previous model in different kinds of genomic datasets demonstrates that our model has made significant improvements over previous state-of-the-art methods. In addition, the simulation experiments are performed to illustrate how the DBP method addresses the bias caused by Cox Network. In the case study, based on the predicted risks in BRCA data from TCGA, we identify 400 differential expressed genes and 20 KEGG pathways that are associated with breast cancer prognosis, among which 65% of the top 20 genes have been proved by literature review. CONCLUSION: Overall, these results demonstrate that our proposed method is advanced and robust in datasets with a large proportion of censored samples. Besides, it can guide to discover disease-related genes.

Recent Advances of Deep Learning in Bioinformatics and Computational Biology.

Extracting inherent valuable knowledge from omics big data remains as a daunting problem in bioinformatics and computational biology. Deep learning, as an emerging branch from machine learning, has exhibited unprecedented performance in quite a few applications from academia and industry. We highlight the difference and similarity in widely utilized models in deep learning studies, through discussing their basic structures, and reviewing diverse applications and disadvantages. We anticipate the work can serve as a meaningful perspective for further development of its theory, algorithm and application in bioinformatic and computational biology.