Abstract The advances in single-cell RNA sequencing (scRNA-seq) technologies enable the characterization of transcriptomic profiles at cellular level and demonstrate great promise bulk sample analysis thereby offering opportunities to transfer gene signature from scRNA-seq data. However, expression signatures identified single cells are typically inapplicable RNA-seq data due profiling differences distinct technologies. Here, we propose pair-wise (scPAGE), a novel method develop pair...

Hepatocellular carcinoma (HCC) is a primary malignancy with poor prognosis. Recently, multi-omics molecular-level measurement enables HCC diagnosis and prognosis prediction, which crucial for early intervention of personalized therapy to diminish mortality. Here, we introduce novel strategy utilizing DNA methylation RNA expression data achieve gene pair signature (GPS) discrimination.The immune genes negative correlations between promoter are enriched in the highly connected cancer-related...

Accurate detection of pathogens, particularly distinguishing between Gram-positive and Gram-negative bacteria, could improve disease treatment. Host gene expression can capture the immune system's response to infections caused by various pathogens. Here, we present a deep neural network model, bvnGPS2, which incorporates attention mechanism based on large-scale integrated host transcriptome dataset precisely identify bacterial as well viral infections. We performed analysis 4,949 blood...

Abstract Background Sepsis is the major cause of death in Intensive Care Unit (ICU) globally. Molecular detection enables rapid diagnosis that allows early intervention to minimize rate. Recent studies showed long non-coding RNAs (lncRNAs) regulate proinflammatory genes and are related dysfunction organs sepsis. Identifying lncRNA signature with absolute abundance challenging because technical variation systematic experimental bias. Results Cohorts ( n = 768) containing whole blood profiling...

Abstract Motivation Many studies have shown that IDH mutation and 1p/19q co-deletion can serve as prognostic signatures of glioma. Although these genetic variations affect the expression one or more genes, value gene related to status is still unclear. Results We constructed an ensemble pair signature for risk evaluation survival prediction glioma based on prior knowledge status. First, we separately built two IDH-GPS 1p/19q-GPS elucidated they were useful transcriptome markers projecting...

Abstract Motivation The confusion of acute inflammation infected by virus and bacteria or noninfectious will lead to missing the best therapy occasion resulting in poor prognoses. diagnostic model based on host gene expression has been widely used diagnose infections, but clinical usage was hindered capability across different samples cohorts due small sample size for signature training discovery. Results Here, we construct a large-scale dataset integrating multiple transcriptomic data...

Most graph contrastive learning (GCL) methods heavily rely on cross-view contrast, thus facing several concomitant challenges, such as the complexity of designing effective augmentations, potential for information loss between views, and increased computational costs. To mitigate reliance contrasts, we propose SIGNA, a novel single-view framework. Regarding inconsistency structural connection semantic similarity neighborhoods, resort to soft neighborhood awareness GCL. Specifically, leverage...

Spatial multi-modal omics technology, highlighted by Nature Methods as an advanced biological technique in 2023, plays a critical role resolving regulatory processes with spatial context. Recently, graph neural networks based on K-nearest neighbor (KNN) graphs have gained prominence methods due to their ability model semantic relations between sequencing spots. However, the fixed KNN fails capture latent hidden inevitable data perturbations during process, resulting loss of information. In...

Dear Editor, Sepsis, the highest mortality disease in critically ill patients, is clinically diagnosed through dysregulated systemic inflammatory response of patients to infection presence organ dysfunction.1-3 No effective biomarkers and approved molecular therapies have been developed for sepsis diagnose treat immune state leading management these only relies on early recognition by experience supportive care.4, 5 Long noncoding RNAs (lncRNAs) are implicated a wide variety biological...

High-throughput sequencing can detect tens of thousands genes in parallel, providing opportunities for improving the diagnostic accuracy multiple diseases including sepsis, which is an aggressive inflammatory response to infection that cause organ failure and death. Early screening sepsis essential clinic, but no effective biomarkers are available yet. Here, we present a novel method, Recurrent Logistic Regression, identify from blood transcriptome data. A panel five immune-related genes,...

The non-coding RNA (ncRNA) regulation appears to be associated the diagnosis and targeted therapy of complex diseases. Motifs RNAs genes in competing endogenous (ceRNA) network would probably contribute accurate prediction serous ovarian carcinoma (SOC). We conducted a microarray study profiling whole transcriptomes eight human SOCs controls constructed ceRNA including mRNAs, long ncRNAs, circular (circRNAs). Novel form motifs (mRNA-ncRNA-mRNA) were identified from defined as RNA's gene...

Abstract Background Portal vein thrombosis (PVT) is a significant issue in cirrhotic patients, necessitating early detection. This study aims to develop data-driven predictive model for PVT diagnosis chronic hepatitis liver cirrhosis patients. Methods We employed data from total of 816 patients with PVT, divided into the Lanzhou cohort (n = 468) training and Jilin 348) validation. dataset encompassed wide range variables, including general characteristics, blood parameters, ultrasonography...

Spatial multi-modal omics technology, highlighted by Nature Methods as an advanced biological technique in 2023, plays a critical role resolving regulatory processes with spatial context. Recently, graph neural networks based on K-nearest neighbor (KNN) graphs have gained prominence methods due to their ability model semantic relations between sequencing spots. However, the fixed KNN fails capture latent hidden inevitable data perturbations during process, resulting loss of information. In...

Bisulfite sequencing is considered as the gold standard approach for measuring DNA methylation, which acts a pivotal part in regulating variety of biological processes without changes sequences. In this study, we introduced most prevalent methods processing bisulfite data and evaluated consistency acquired from different measurements liver cancer. Firstly, three commonly used assays, i.e., reduced-representation (RRBS), whole-genome (WGBS), targeted (targeted BS). Next, discussed principles...

Sepsis is a life-threatening condition characterized by an exaggerated immune response to pathogens, leading organ damage and high mortality rates in the intensive care unit. Although deep learning has achieved impressive performance on prediction classification tasks medicine, it requires large amounts of data lacks explainability, which hinder its application sepsis diagnosis. We introduce framework, called scCaT, blends capsulating architecture with Transformer develop diagnostic model...

Abstract Motivation Human gut microbiota plays a vital role in maintaining body health. The dysbiosis of is associated with variety diseases. It critical to uncover the associations between and disease states as well other intrinsic or environmental factors. However, inferring alterations individual microbial taxa based on relative abundance data likely leads false conflicting discoveries different studies. Moreover, effects underlying factors microbe–microbe interactions could lead...

Abstract Sepsis is a life-threatening condition characterized by an exaggerated immune response to pathogens, leading organ damage and high mortality rates in the intensive care unit. Although deep learning has achieved impressive performance on prediction classification tasks medicine, it requires large amounts of data lacks explainability, which hinder its application sepsis diagnosis. We introduce framework, called scCaT, blends capsulating architecture with Transformer develop diagnostic...

Single-cell multi-omics (scMulti-omics) refers to the paired multimodal data, such as Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE-seq), where regulation each cell was measured from different modalities, i.e. genes proteins. scMulti-omics can reveal heterogeneity inside tumors understand distinct genetic properties diverse types, which is crucial targeted therapy. Currently, deep learning methods based on attention structures in bioinformatics area face two...

Knowledge Distillation (KD) transfers knowledge from a large pre-trained teacher network to compact and efficient student network, making it suitable for deployment on resource-limited media terminals. However, traditional KD methods require balanced data ensure robust training, which is often unavailable in practical applications. In such scenarios, few head categories occupy substantial proportion of examples. This imbalance biases the trained towards categories, resulting severe...

Percocypris pingi (Tchang) was classified as Endangered on the Red List of China′s Vertebrates in 2015 and is widely distributed Upper Yangtze River. Although breeding release into wild habitats have been performed for this commercially important fish recent years, low genetic diversity has found populations. Genomic resources are strongly recommended before formulating carrying out conservation strategies P. pingi. Thus, there an urgent need to conserve germplasm improve population To date,...

Most graph contrastive learning (GCL) methods heavily rely on cross-view contrast, thus facing several concomitant challenges, such as the complexity of designing effective augmentations, potential for information loss between views, and increased computational costs. To mitigate reliance contrasts, we propose \ttt{SIGNA}, a novel single-view framework. Regarding inconsistency structural connection semantic similarity neighborhoods, resort to soft neighborhood awareness GCL. Specifically,...