HGCLAMIR: Hypergraph contrastive learning with attention mechanism and integrated multi-view representation for predicting miRNA-disease associations.

Existing studies have shown that the abnormal expression of microRNAs (miRNAs) usually leads to the occurrence and development of human diseases. Identifying disease-related miRNAs contributes to studying the pathogenesis of diseases at the molecular level. As traditional biological experiments are time-consuming and expensive, computational methods have been used as an effective complement to infer the potential associations between miRNAs and diseases. However, most of the existing computational methods still face three main challenges: (i) learning of high-order relations; (ii) insufficient representation learning ability; (iii) importance learning and integration of multi-view embedding representation. To this end, we developed a HyperGraph Contrastive Learning with view-aware Attention Mechanism and Integrated multi-view Representation (HGCLAMIR) model to discover potential miRNA-disease associations. First, hypergraph convolutional network (HGCN) was utilized to capture high-order complex relations from hypergraphs related to miRNAs and diseases. Then, we combined HGCN with contrastive learning to improve and enhance the embedded representation learning ability of HGCN. Moreover, we introduced view-aware attention mechanism to adaptively weight the embedded representations of different views, thereby obtaining the importance of multi-view latent representations. Next, we innovatively proposed integrated representation learning to integrate the embedded representation information of multiple views for obtaining more reasonable embedding information. Finally, the integrated representation information was fed into a neural network-based matrix completion method to perform miRNA-disease association prediction. Experimental results on the cross-validation set and independent test set indicated that HGCLAMIR can achieve better prediction performance than other baseline models. Furthermore, the results of case studies and enrichment analysis further demonstrated the accuracy of HGCLAMIR and unconfirmed potential associations had biological significance.

Inflammatory cell death PANoptosis is induced by the anti-cancer curaxin CBL0137 via eliciting the assembly of ZBP1-associated PANoptosome.

OBJECTIVE: PANoptosis, a new form of regulated cell death, concomitantly manifests hallmarks for pyroptosis, apoptosis, and necroptosis. It has been usually observed in macrophages, a class of widely distributed innate immune cells in various tissues, upon pathogenic infections. The second-generation curaxin, CBL0137, can trigger necroptosis and apoptosis in cancer-associated fibroblasts. This study aimed to explore whether CBL0137 induces PANoptosis in macrophages in vitro and in mouse tissues in vivo. METHODS: Bone marrow-derived macrophages and J774A.1 cells were treated with CBL0137 or its combination with LPS for indicated time periods. Cell death was assayed by propidium iodide staining and immunoblotting. Immunofluorescence microscopy was used to detect cellular protein distribution. Mice were administered with CBL0137 plus LPS and their serum and tissues were collected for biochemical and histopathological analyses, respectively. RESULTS: The results showed that CBL0137 alone or in combination with LPS induced time- and dose-dependent cell death in macrophages, which was inhibited by a combination of multiple forms of cell death inhibitors but not each alone. This cell death was independent of NLRP3 expression. CBL0137 or CBL0137 + LPS-induced cell death was characterized by simultaneously increased hallmarks for pyroptosis, apoptosis and necroptosis, indicating that this is PANoptosis. Induction of PANoptosis was associated with Z-DNA formation in the nucleus and likely assembly of PANoptosome. ZBP1 was critical in mediating CBL0137 + LPS-induced cell death likely by sensing Z-DNA. Moreover, intraperitoneal administration of CBL0137 plus LPS induced systemic inflammatory responses and caused multi-organ (including the liver, kidney and lung) injury in mice due to induction of PANoptosis in these organs. CONCLUSIONS: CBL0137 alone or plus inflammatory stimulation induces PANoptosis both in vitro and in vivo, which is associated with systemic inflammatory responses in mice.

Extracellular ATP contributes to the reactive oxygen species burst and exaggerated mitochondrial damage in D-galactosamine and lipopolysaccharide-induced fulminant hepatitis.

Fulminant hepatitis (FH) is a severe clinical syndrome leading to hepatic failure and even mortality. D-galactosamine (D-GalN) plus lipopolysaccharide (LPS) challenge is commonly used to establish an FH mouse model, but the mechanism underlying D-GalN/LPS-induced liver injury is incompletely understood. Previously, it has been reported that extracellular ATP that can be released under cytotoxic and inflammatory stresses serves as a damage signal to induce potassium ion efflux and trigger the NACHT, LRR and PYD domains-containing protein 3 (NLRP3) inflammasome activation through binding to P2X7 receptor. In this study, we tried to investigate whether it contributed to the fulminant hepatitis (FH) induced by D-GalN plus LPS. In an in vitro cellular model, D-GalN plus extracellular ATP, instead of D-GalN alone, induced pyroptosis and apoptosis, accompanied by mitochondrial reactive oxygen species (ROS) burst, and the oligomerization of Drp1, Bcl-2, and Bak, as well as the loss of mitochondrial membrane potential in LPS-primed macrophages, well reproducing the events induced by D-GalN and LPS in vivo. Moreover, these events in the cellular model were markedly suppressed by both A-804598 (an ATP receptor P2X7R inhibitor) and glibenclamide (an ATP-sensitive potassium ion channel inhibitor); in the FH mouse model, administration of A-804598 significantly mitigated D-GalN/LPS-induced hepatic injury, mitochondrial damage, and the activation of apoptosis and pyroptosis signaling, corroborating the contribution of extracellular ATP to the cell death. Collectively, our data suggest that extracellular ATP acts as an autologous damage-associated molecular pattern to augment mitochondrial damage, hepatic cell death, and liver injury in D-GalN/LPS-induced FH mouse model.

Anti-Necroptotic Effects of Itaconate and its Derivatives.

Itaconate is an unsaturated dicarboxylic acid that is derived from the decarboxylation of the Krebs cycle intermediate cis-aconitate and has been shown to exhibit anti-inflammatory and anti-bacterial/viral properties. But the mechanisms underlying itaconate's anti-inflammatory activities are not fully understood. Necroptosis, a lytic form of regulated cell death (RCD), is mediated by receptor-interacting protein kinase 1 (RIPK1), RIPK3, and mixed lineage kinase domain-like protein (MLKL) signaling. It has been involved in the pathogenesis of organ injury in many inflammatory diseases. In this study, we aimed to explore whether itaconate and its derivatives can inhibit necroptosis in murine macrophages, a mouse MPC-5 cell line and a human HT-29 cell line in response to different necroptotic activators. Our results showed that itaconate and its derivatives dose-dependently inhibited necroptosis, among which dimethyl itaconate (DMI) was the most effective one. Mechanistically, itaconate and its derivatives inhibited necroptosis by suppressing the RIPK1/RIPK3/MLKL signaling and the oligomerization of MLKL. Furthermore, DMI promoted the nuclear translocation of Nrf2 that is a critical regulator of intracellular redox homeostasis, and reduced the levels of intracellular reactive oxygen species (ROS) and mitochondrial superoxide (mtROS) that were induced by necroptotic activators. Consistently, DMI prevented the loss of mitochondrial membrane potential induced by the necroptotic activators. In addition, DMI mitigated caerulein-induced acute pancreatitis in mice accompanied by reduced activation of the necroptotic signaling in vivo. Collectively, our study demonstrates that itaconate and its derivatives can inhibit necroptosis by suppressing the RIPK1/RIPK3/MLKL signaling, highlighting their potential applications for treating necroptosis-associated diseases.

Potassium ion efflux induces exaggerated mitochondrial damage and non-pyroptotic necrosis when energy metabolism is blocked.

Damage-associated molecular patterns (DAMPs) such as extracellular ATP and nigericin (a bacterial toxin) not only act as potassium ion (K+) efflux inducers to activate NLRP3 inflammasome, leading to pyroptosis, but also induce cell death independently of NLRP3 expression. However, the roles of energy metabolism in determining NLRP3-dependent pyroptosis and -independent necrosis upon K+ efflux are incompletely understood. Here we established cellular models by pharmacological blockade of energy metabolism, followed by stimulation with a K+ efflux inducer (ATP or nigericin). Two energy metabolic inhibitors, namely CPI-613 that targets α-ketoglutarate dehydrogenase and pyruvate dehydrogenase (a rate-limiting enzyme) and 2-deoxy-d-glucose (2-DG) that targets hexokinase, are recruited in this study, and Nlrp3 gene knockout macrophages were used. Our data showed that CPI-613 and 2-DG dose-dependently inhibited NLRP3 inflammasome activation, but profoundly increased cell death in the presence of ATP or nigericin. The cell death was K+ efflux-induced but NLRP3-independent, which was associated with abrupt reactive oxygen species (ROS) production, reduction of mitochondrial membrane potential, and oligomerization of mitochondrial proteins, all indicating mitochondrial damage. Notably, the cell death induced by K+ efflux and blockade of energy metabolism was distinct from pyroptosis, apoptosis, necroptosis or ferroptosis. Furthermore, fructose 1,6-bisphosphate, a high-energy intermediate of glycolysis, significantly suppressed CPI-613+nigericin-induced mitochondrial damage and cell death. Collectively, our data show that energy deficiency diverts NLRP3 inflammasome activation-dependent pyroptosis to Nlrp3-independent necrosis upon K+ efflux inducers, which can be dampened by high-energy intermediate, highlighting a critical role of energy metabolism in cell survival and death under inflammatory conditions.

Predicting potential microbe-disease associations based on auto-encoder and graph convolution network.

The increasing body of research has consistently demonstrated the intricate correlation between the human microbiome and human well-being. Microbes can impact the efficacy and toxicity of drugs through various pathways, as well as influence the occurrence and metastasis of tumors. In clinical practice, it is crucial to elucidate the association between microbes and diseases. Although traditional biological experiments accurately identify this association, they are time-consuming, expensive, and susceptible to experimental conditions. Consequently, conducting extensive biological experiments to screen potential microbe-disease associations becomes challenging. The computational methods can solve the above problems well, but the previous computational methods still have the problems of low utilization of node features and the prediction accuracy needs to be improved. To address this issue, we propose the DAEGCNDF model predicting potential associations between microbes and diseases. Our model calculates four similar features for each microbe and disease. These features are fused to obtain a comprehensive feature matrix representing microbes and diseases. Our model first uses the graph convolutional network module to extract low-rank features with graph information of microbes and diseases, and then uses a deep sparse Auto-Encoder to extract high-rank features of microbe-disease pairs, after which the low-rank and high-rank features are spliced to improve the utilization of node features. Finally, Deep Forest was used for microbe-disease potential relationship prediction. The experimental results show that combining low-rank and high-rank features helps to improve the model performance and Deep Forest has better classification performance than the baseline model.

Inferring circRNA-drug sensitivity associations via dual hierarchical attention networks and multiple kernel fusion.

Increasing evidence has shown that the expression of circular RNAs (circRNAs) can affect the drug sensitivity of cells and significantly influence drug efficacy. Therefore, research into the relationships between circRNAs and drugs can be of great significance in increasing the comprehension of circRNAs function, as well as contributing to the discovery of new drugs and the repurposing of existing drugs. However, it is time-consuming and costly to validate the function of circRNA with traditional medical research methods. Therefore, the development of efficient and accurate computational models that can assist in discovering the potential interactions between circRNAs and drugs is urgently needed. In this study, a novel method is proposed, called DHANMKF , that aims to predict potential circRNA-drug sensitivity interactions for further biomedical screening and validation. Firstly, multimodal networks were constructed by DHANMKF using multiple sources of information on circRNAs and drugs. Secondly, comprehensive intra-type and inter-type node representations were learned using bi-typed multi-relational heterogeneous graphs, which are attention-based encoders utilizing a hierarchical process. Thirdly, the multi-kernel fusion method was used to fuse intra-type embedding and inter-type embedding. Finally, the Dual Laplacian Regularized Least Squares method (DLapRLS) was used to predict the potential circRNA-drug sensitivity associations using the combined kernel in circRNA and drug spaces. Compared with the other methods, DHANMKF obtained the highest AUC value on two datasets. Code is available at https://github.com/cuntjx/DHANMKF .

Triptolide induces PANoptosis in macrophages and causes organ injury in mice.

Macrophages represent the first lines of innate defense against pathogenic infections and are poised to undergo multiple forms of regulated cell death (RCD) upon infections or toxic stimuli, leading to multiple organ injury. Triptolide, an active compound isolated from Tripterygium wilfordii Hook F., possesses various pharmacological activities including anti-tumor and anti-inflammatory effects, but its applications have been hampered by toxic adverse effects. It remains unknown whether and how triptolide induces different forms of RCD in macrophages. In this study, we showed that triptolide exhibited significant cytotoxicity on cultured macrophages in vitro, which was associated with multiple forms of lytic cell death that could not be fully suppressed by any one specific inhibitor for a single form of RCD. Consistently, triptolide induced the simultaneous activation of pyroptotic, apoptotic and necroptotic hallmarks, which was accompanied by the co-localization of ASC specks respectively with RIPK3 or caspase-8 as well as their interaction with each other, indicating the formation of PANoptosome and thus the induction of PANoptosis. Triptolide-induced PANoptosis was associated with mitochondrial dysfunction and ROS production. PANoptosis was also induced by triptolide in mouse peritoneal macrophages in vivo. Furthermore, triptolide caused kidney and liver injury, which was associated with systemic inflammatory responses and the activation of hallmarks for PANoptosis in vivo. Collectively, our data reveal that triptolide induces PANoptosis in macrophages in vitro and exhibits nephrotoxicity and hepatotoxicity associated with induction of PANoptosis in vivo, suggesting a new avenue to alleviate triptolide's toxicity by harnessing PANoptosis.

Integration of multi-omics data using adaptive graph learning and attention mechanism for patient classification and biomarker identification.

With the rapid development and accumulation of high-throughput sequencing technology and omics data, many studies have conducted a more comprehensive understanding of human diseases from a multi-omics perspective. Meanwhile, graph-based methods have been widely used to process multi-omics data due to its powerful expressive ability. However, most existing graph-based methods utilize fixed graphs to learn sample embedding representations, which often leads to sub-optimal results. Furthermore, treating embedding representations of different omics equally usually cannot obtain more reasonable integrated information. In addition, the complex correlation between omics is not fully taken into account. To this end, we propose an end-to-end interpretable multi-omics integration method, named MOGLAM, for disease classification prediction. Dynamic graph convolutional network with feature selection is first utilized to obtain higher quality omic-specific embedding information by adaptively learning the graph structure and discover important biomarkers. Then, multi-omics attention mechanism is applied to adaptively weight the embedding representations of different omics, thereby obtaining more reasonable integrated information. Finally, we propose omic-integrated representation learning to capture complex common and complementary information between omics while performing multi-omics integration. Experimental results on three datasets show that MOGLAM achieves superior performance than other state-of-the-art multi-omics integration methods. Moreover, MOGLAM can identify important biomarkers from different omics data types in an end-to-end manner.

Associations between gut microbiota and sleep: a two-sample, bidirectional Mendelian randomization study.

Introduction: Previous research has reported that the gut microbiota performs an essential role in sleep through the microbiome-gut-brain axis. However, the causal association between gut microbiota and sleep remains undetermined. Methods: We performed a two-sample, bidirectional Mendelian randomization (MR) analysis using genome-wide association study summary data of gut microbiota and self-reported sleep traits from the MiBioGen consortium and UK Biobank to investigate causal relationships between 119 bacterial genera and seven sleep-associated traits. We calculated effect estimates by using the inverse-variance weighted (as the main method), maximum likelihood, simple model, weighted model, weighted median, and MR-Egger methods, whereas heterogeneity and pleiotropy were detected and measured by the MR pleiotropy residual sum and outlier method, Cochran's Q statistics, and MR-Egger regression. Results: In forward MR analysis, inverse-variance weighted estimates concluded that the genetic forecasts of relative abundance of 42 bacterial genera had causal effects on sleep-associated traits. In the reverse MR analysis, sleep-associated traits had a causal effect on 39 bacterial genera, 13 of which overlapped with the bacterial genera in the forward MR analysis. Discussion: In conclusion, our research indicates that gut microbiota may be involved in the regulation of sleep, and conversely, changes in sleep-associated traits may also alter the abundance of gut microbiota. These findings suggest an underlying reciprocal causal association between gut microbiota and sleep.

GDCL-NcDA: identifying non-coding RNA-disease associations via contrastive learning between deep graph learning and deep matrix factorization.

Non-coding RNAs (ncRNAs) draw much attention from studies widely in recent years because they play vital roles in life activities. As a good complement to wet experiment methods, computational prediction methods can greatly save experimental costs. However, high false-negative data and insufficient use of multi-source information can affect the performance of computational prediction methods. Furthermore, many computational methods do not have good robustness and generalization on different datasets. In this work, we propose an effective end-to-end computing framework, called GDCL-NcDA, of deep graph learning and deep matrix factorization (DMF) with contrastive learning, which identifies the latent ncRNA-disease association on diverse multi-source heterogeneous networks (MHNs). The diverse MHNs include different similarity networks and proven associations among ncRNAs (miRNAs, circRNAs, and lncRNAs), genes, and diseases. Firstly, GDCL-NcDA employs deep graph convolutional network and multiple attention mechanisms to adaptively integrate multi-source of MHNs and reconstruct the ncRNA-disease association graph. Then, GDCL-NcDA utilizes DMF to predict the latent disease-associated ncRNAs based on the reconstructed graphs to reduce the impact of the false-negatives from the original associations. Finally, GDCL-NcDA uses contrastive learning (CL) to generate a contrastive loss on the reconstructed graphs and the predicted graphs to improve the generalization and robustness of our GDCL-NcDA framework. The experimental results show that GDCL-NcDA outperforms highly related computational methods. Moreover, case studies demonstrate the effectiveness of GDCL-NcDA in identifying the associations among diversiform ncRNAs and diseases.

Celastrol inhibits necroptosis by attenuating the RIPK1/RIPK3/MLKL pathway and confers protection against acute pancreatitis in mice.

Necroptosis is a necrotic form of regulated cell death, which is primarily mediated by the receptor-interacting protein kinase 1 (RIPK1), RIPK3, and mixed lineage kinase domain-like (MLKL) pathway in a caspase-independent manner. Necroptosis has been found to occur in virtually all tissues and diseases evaluated, including pancreatitis. Celastrol, a pentacyclic triterpene extracted from the roots of Tripterygium wilfordii (thunder god vine), possesses potent anti-inflammatory and anti-oxidative activities. Yet, it is unclear whether celastrol has any effects on necroptosis and necroptotic-related diseases. Here we showed that celastrol significantly suppressed necroptosis induced by lipopolysaccharide (LPS) plus pan-caspase inhibitor (IDN-6556) or by tumor-necrosis factor-α in combination with LCL-161 (Smac mimetic) and IDN-6556 (TSI). In these in vitro cellular models, celastrol inhibited the phosphorylation of RIPK1, RIPK3, and MLKL and the formation of necrosome during necroptotic induction, suggesting its possible action on upstream signaling of the necroptotic pathway. Consistent with the known role of mitochondrial dysfunction in necroptosis, we found that celastrol significantly rescued TSI-induced loss of mitochondrial membrane potential. TSI-induced intracellular and mitochondrial reactive oxygen species (mtROS), which are involved in the autophosphorylation of RIPK1 and recruitment of RIPK3, were significantly attenuated by celastrol. Moreover, in a mouse model of acute pancreatitis that is associated with necroptosis, celastrol administration significantly reduced the severity of caerulein-induced acute pancreatitis accompanied by decreased phosphorylation of MLKL in pancreatic tissues. Collectively, celastrol can attenuate the activation of RIPK1/RIPK3/MLKL signaling likely by attenuating mtROS production, thereby inhibiting necroptosis and conferring protection against caerulein-induced pancreatitis in mice.

Dimethyl fumarate inhibits necroptosis and alleviates systemic inflammatory response syndrome by blocking the RIPK1-RIPK3-MLKL axis.

Necroptosis has been implicated in various inflammatory diseases including tumor-necrosis factor-α (TNF-α)-induced systemic inflammatory response syndrome (SIRS). Dimethyl fumarate (DMF), a first-line drug for treating relapsing-remitting multiple sclerosis (RRMS), has been shown to be effective against various inflammatory diseases. However, it is still unclear whether DMF can inhibit necroptosis and confer protection against SIRS. In this study, we found that DMF significantly inhibited necroptotic cell death in macrophages induced by different necroptotic stimulations. Both the autophosphorylation of receptor-interacting serine/threonine kinase 1 (RIPK1) and RIPK3 and the downstream phosphorylation and oligomerization of MLKL were robustly suppressed by DMF. Accompanying the suppression of necroptotic signaling, DMF blocked the mitochondrial reverse electron transport (RET) induced by necroptotic stimulation, which was associated with its electrophilic property. Several well-known anti-RET reagents also markedly inhibited the activation of the RIPK1-RIPK3-MLKL axis accompanied by decreased necrotic cell death, indicating a critical role of RET in necroptotic signaling. DMF and other anti-RET reagents suppressed the ubiquitination of RIPK1 and RIPK3, and they attenuated the formation of necrosome. Moreover, oral administration of DMF significantly alleviated the severity of TNF-α-induced SIRS in mice. Consistent with this, DMF mitigated TNF-α-induced cecal, uterine, and lung damage accompanied by diminished RIPK3-MLKL signaling. Collectively, DMF represents a new necroptosis inhibitor that suppresses the RIPK1-RIPK3-MLKL axis through blocking mitochondrial RET. Our study highlights DMF's potential therapeutic applications for treating SIRS-associated diseases.

["Component-target-efficacy" network analysis and experimental verification of Qingkailing Oral Preparation].

This study aims to explore the mechanism of Qingkailing(QKL) Oral Preparation's heat-clearing, detoxifying, mind-tranquilizing effects based on "component-target-efficacy" network. To be specific, the potential targets of the 23 major components in QKL Oral Preparation were predicted by the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform(TCMSP) and SwissTargetPrediction. The target genes were obtained based on UniProt. OmicsBean and STRING 10 were used for Gene Ontology(GO) term enrichment and Kyoto Encyclopedia of Genes and Genomes(KEGG) pathway enrichment of the targets. Cytoscape 3.8.2 was employed for visualization and construction of "component-target-pathway-pharmacological effect-efficacy" network, followed by molecular docking between the 23 main active components and 15 key targets. Finally, the lipopolysaccharide(LPS)-induced RAW264.7 cells were adopted to verify the anti-inflammatory effect of six monomer components in QKL Oral Preparation. It was found that the 23 compounds affected 33 key signaling pathways through 236 related targets, such as arachidonic acid metabolism, tumor necrosis factor α(TNF-α) signaling pathway, inflammatory mediator regulation of TRP channels, cAMP signaling pathway, cGMP-PKG signaling pathway, Th17 cell differentiation, interleukin-17(IL-17) signaling pathway, neuroactive ligand-receptor intera-ction, calcium signaling pathway, and GABAergic synapse. They were involved in the anti-inflammation, immune regulation, antipyretic effect, and anti-convulsion of the prescription. The "component-target-pathway-pharmacological effect-efficacy" network of QKL Oral Preparation was constructed. Molecular docking showed that the main active components had high binding affinity to the key targets. In vitro cell experiment indicated that the six components in the prescription(hyodeoxycholic acid, baicalin, chlorogenic acid, isochlorogenic acid C, epigoitrin, geniposide) can reduce the expression of nitric oxide(NO), TNF-α, and interleukin-6(IL-6) in cell supernatant(P<0.05). Thus, the above six components may be the key pharmacodynamic substances of QKL Oral Preparation. The major components in QKL Oral Prescription, including hyodeoxycholic acid, baicalin, chlorogenic acid, isochlorogenic acid C, epigoitrin, geniposide, cholic acid, isochlorogenic acid A, and γ-aminobutyric acid, may interfere with multiple biological processes related to inflammation, immune regulation, fever, and convulsion by acting on the key protein targets such as IL-6, TNF, prostaglandin-endoperoxide synthase 2(PTGS2), arachidonate 5-lipoxygenase(ALOX5), vascular cell adhesion molecule 1(VCAM1), nitric oxide synthase 2(NOS2), prostaglandin E2 receptor EP2 subtype(PTGER2), gamma-aminobutyric acid receptor subunit alpha(GABRA), gamma-aminobutyric acid type B receptor subunit 1(GABBR1), and 4-aminobutyrate aminotransferase(ABAT). This study reveals the effective components and mechanism of QKL Oral Prescription.

Improved Computational Drug-Repositioning by Self-Paced Non-Negative Matrix Tri-Factorization.

Drug repositioning (DR) is a strategy to find new targets for existing drugs, which plays an important role in reducing the costs, time, and risk of traditional drug development. Recently, the matrix factorization approach has been widely used in the field of DR prediction. Nevertheless, there are still two challenges: 1) Learning ability deficiencies, the model cannot accurately predict more potential associations. 2) Easy to fall into a bad local optimal solution, the model tends to get a suboptimal result. In this study, we propose a self-paced non-negative matrix tri-factorization (SPLNMTF) model, which integrates three types of different biological data from patients, genes, and drugs into a heterogeneous network through non-negative matrix tri-factorization, thereby learning more information to improve the learning ability of the model. In the meantime, the SPLNMTF model sequentially includes samples into training from easy (high-quality) to complex (low-quality) in the soft weighting way, which effectively alleviates falling into a bad local optimal solution to improve the prediction performance of the model. The experimental results on two real datasets of ovarian cancer and acute myeloid leukemia (AML) show that SPLNMTF outperforms the other eight state-of-the-art models and gets better prediction performance in drug repositioning. The data and source code are available at: https://github.com/qi0906/SPLNMTF.

Self-supervised learning-based Multi-Scale feature Fusion Network for survival analysis from whole slide images.

Understanding prognosis and mortality is critical for evaluating the treatment plan of patients. Advances in digital pathology and deep learning techniques have made it practical to perform survival analysis in whole slide images (WSIs). Current methods are usually based on a multi-stage framework which includes patch sampling, feature extraction and prediction. However, the random patch sampling strategy is highly unstable and prone to sampling non-ROI. Feature extraction typically relies on hand-crafted features or convolutional neural networks (CNNs) pre-trained on ImageNet, while the artificial error or domain gaps may affect the survival prediction performance. Besides, the limited information representation of local sampling patches will create a bottleneck limitation on the effectiveness of prediction. To address the above challenges, we propose a novel patch sampling strategy based on image information entropy and construct a Multi-Scale feature Fusion Network (MSFN) based on self-supervised feature extractor. Specifically, we adopt image information entropy as a criterion to select representative sampling patches, thereby avoiding the noise interference caused by random to blank regions. Meanwhile, we pretrain the feature extractor utilizing self-supervised learning mechanism to improve the efficiency of feature extraction. Furthermore, a global-local feature fusion prediction network based on the attention mechanism is constructed to improve the survival prediction effect of WSIs with comprehensive multi-scale information representation. The proposed method is validated by adequate experiments and achieves competitive results on both of the most popular WSIs survival analysis datasets, TCGA-GBM and TCGA-LUSC. Code and trained models are made available at: https://github.com/Mercuriiio/MSFN.

Prognostic analysis of pT1-T2aN0M0 cervical adenocarcinoma based on random survival forest analysis and the generation of a predictive nomogram.

Background: The efficacy of adjuvant radiotherapy for postoperative patients with early-stage cervical adenocarcinoma who are lymph node-negative is still inconclusive. Establishing a nomogram to predict the prognosis of such patients could facilitate clinical decision-making. Methods: We recruited 4636 eligible patients with pT1-T2aN0M0 cervical adenocarcinoma between 2004 and 2016 from the Surveillance, Epidemiology and End Results (SEER) database. Random survival forest (RSF) and conditional survival forest (CSF) model was used to assess the prognostic importance of each clinical characteristic variable. We identified independent prognostic factors associated with overall survival (OS) by univariate and multivariate Cox regression risk methods and then constructed a nomogram. We stratified patients based on nomogram risk scores and evaluated the survival benefit of different adjuvant therapies. To reduce confounding bias, we also used propensity score matching (PSM) to match the cohorts before performing survival analyses. Results: The RSF and CSF model identified several important variables that are associated with prognosis, including grade, age, radiotherapy and tumor size. Patients were randomly divided into training and validation groups at a ratio of 7:3. Multivariate cox analysis revealed that age, grade, tumor size, race, radiotherapy and histology were independent prognostic factors for overall survival. Using these variables, we then constructed a predictive nomogram. The C-index value for evaluating the prognostic nomogram fluctuated between 0.75 and 0.91. Patients were divided into three subgroups based on risk scores, and Kaplan-Meier (K-M) survival analysis revealed that in the low-risk group, postoperative chemotherapy alone was associated with a significantly worse OS than surgery alone. Following PSM, survival analysis showed that compared with surgery alone, radiotherapy was associated with a worse OS in the training group although there was no significant difference in the validation group. Conclusions: For patients with pT1-T2aN0M0 cervical adenocarcinoma, adjuvant treatments such as postoperative radiotherapy or chemotherapy, compared with surgery alone, are of no benefit with regards to patient survival. Our prognostic nomogram exhibits high accuracy for predicting the survival of patients with early-stage postoperative cervical adenocarcinoma.

Inferring human miRNA-disease associations via multiple kernel fusion on GCNII.

Increasing evidence shows that the occurrence of human complex diseases is closely related to the mutation and abnormal expression of microRNAs(miRNAs). MiRNAs have complex and fine regulatory mechanisms, which makes it a promising target for drug discovery and disease diagnosis. Therefore, predicting the potential miRNA-disease associations has practical significance. In this paper, we proposed an miRNA-disease association predicting method based on multiple kernel fusion on Graph Convolutional Network via Initial residual and Identity mapping (GCNII), called MKFGCNII. Firstly, we built a heterogeneous network of miRNAs and diseases to extract multi-layer features via GCNII. Secondly, multiple kernel fusion method was applied to weight fusion of embeddings at each layer. Finally, Dual Laplacian Regularized Least Squares was used to predict new miRNA-disease associations by the combined kernel in miRNA and disease spaces. Compared with the other methods, MKFGCNII obtained the highest AUC value of 0.9631. Code is available at https://github.com/cuntjx/bioInfo.

Corrigendum: Predicting multiple types of associations between miRNAs and diseases based on graph regularized weighted tensor decomposition.

[This corrects the article DOI: 10.3389/fbioe.2022.911769.].

Predicting multiple types of miRNA-disease associations using adaptive weighted nonnegative tensor factorization with self-paced learning and hypergraph regularization.

More and more evidence indicates that the dysregulations of microRNAs (miRNAs) lead to diseases through various kinds of underlying mechanisms. Identifying the multiple types of disease-related miRNAs plays an important role in studying the molecular mechanism of miRNAs in diseases. Moreover, compared with traditional biological experiments, computational models are time-saving and cost-minimized. However, most tensor-based computational models still face three main challenges: (i) easy to fall into bad local minima; (ii) preservation of high-order relations; (iii) false-negative samples. To this end, we propose a novel tensor completion framework integrating self-paced learning, hypergraph regularization and adaptive weight tensor into nonnegative tensor factorization, called SPLDHyperAWNTF, for the discovery of potential multiple types of miRNA-disease associations. We first combine self-paced learning with nonnegative tensor factorization to effectively alleviate the model from falling into bad local minima. Then, hypergraphs for miRNAs and diseases are constructed, and hypergraph regularization is used to preserve the high-order complex relations of these hypergraphs. Finally, we innovatively introduce adaptive weight tensor, which can effectively alleviate the impact of false-negative samples on the prediction performance. The average results of 5-fold and 10-fold cross-validation on four datasets show that SPLDHyperAWNTF can achieve better prediction performance than baseline models in terms of Top-1 precision, Top-1 recall and Top-1 F1. Furthermore, we implement case studies to further evaluate the accuracy of SPLDHyperAWNTF. As a result, 98 (MDAv2.0) and 98 (MDAv2.0-2) of top-100 are confirmed by HMDDv3.2 dataset. Moreover, the results of enrichment analysis illustrate that unconfirmed potential associations have biological significance.