Integration Of Multiple Information Research Articles

THU500 Geospatial Analysis To Quantify Spatial Heterogeneity Of Tumor Microenvironment

Abstract Disclosure: J. Yoo: None. C. Mitra: None. Y. Yue: None. A. Soliman: None. Z. Madak Erdogan: None. Metabolic heterogeneity, a concept describing distinct differences in nutrient metabolism within the cells of a healthy or diseased tissue, is a common challenge in development and cancer research. Current methods, such as histology and Magnetic Resonance Imaging (MRI), cannot visualize nutrient uptake into single cells, making it difficult to understand the heterogeneity that affects cellular growth and metabolism. Visium Spatial Transcriptomics, a method that shows regional gene expression, is emerging as a powerful spatial analysis tool, and we aim to integrate this transcriptomic method with other biologically relevant datasets to highlight the importance of cell-microenvironment interactions. We propose to leverage geostatistical tools to model the multivariate covariance structure of genes expression profiles to allow: 1) understanding of their spatial characteristics and covariance changes under certain conditions (e.g. drug treatment or damage), 2) using these covariance structures to simulate synthetic samples at a scale sufficient to train the state of art neural network architectures, and 3) incorporating spatial awareness directly in the machine learning/deep learning models by enriching the spatial transcriptomics data using distance proximity functions. We will quantify the tumor microenvironment spatial heterogeneity using different approaches. First, we will analyze the degree of spatial auto-correlation to estimate the ranges of spatial heterogeneity within a tumor. We also plan to quantify the degree of anisotropy to identify if the tumor heterogeneity is happening along specific axis. Different local clustering indicators will be also calculated, such as the Getis-Ord (Gi*) statistic and Local Moran's I to identify the prevalence of hot and cold spots in the tumor. These hot and cold spots will be used to segment the tumor area into sub-regions and further quantify the local heterogeneity with a single tumor. We hypothesize that transcriptional heterogeneity reflects phenotypic differences in different regions of healthy or diseased tissue, rendering cells in certain regions to display a differential response to external stimuli (e.g. diet or drugs). Integration of multiple information (e.g. retrospective patient data) with different spatial heterogeneity indicators identified from the geospatial analyses could aid in identification of patients who are resistant to cancer therapies and further advance personalized therapy. Presentation: Thursday, June 15, 2023

MFIDMA: A Multiple Information Integration Model for the Prediction of Drug-miRNA Associations.

Abnormal microRNA (miRNA) functions play significant roles in various pathological processes. Thus, predicting drug-miRNA associations (DMA) may hold great promise for identifying the potential targets of drugs. However, discovering the associations between drugs and miRNAs through wet experiments is time-consuming and laborious. Therefore, it is significant to develop computational prediction methods to improve the efficiency of identifying DMA on a large scale. In this paper, a multiple features integration model (MFIDMA) is proposed to predict drug-miRNA association. Specifically, we first formulated known DMA as a bipartite graph and utilized structural deep network embedding (SDNE) to learn the topological features from the graph. Second, the Word2vec algorithm was utilized to construct the attribute features of the miRNAs and drugs. Third, two kinds of features were entered into the convolution neural network (CNN) and deep neural network (DNN) to integrate features and predict potential target miRNAs for the drugs. To evaluate the MFIDMA model, it was implemented on three different datasets under a five-fold cross-validation and achieved average AUCs of 0.9407, 0.9444 and 0.8919. In addition, the MFIDMA model showed reliable results in the case studies of Verapamil and hsa-let-7c-5p, confirming that the proposed model can also predict DMA in real-world situations. The model was effective in analyzing the neighbors and topological features of the drug-miRNA network by SDNE. The experimental results indicated that the MFIDMA is an accurate and robust model for predicting potential DMA, which is significant for miRNA therapeutics research and drug discovery.

Identification of miRNA-disease associations via multiple information integration with Bayesian ranking.

In recent years, increasing microRNA (miRNA)-disease associations were identified through traditionally biological experiments. These associations contribute to revealing molecular mechanism of diseases and preventing and curing diseases. To improve the efficiency of miRNA-disease association discovery, some calculation methods were developed as auxiliary tools for researchers. In the current study, we raised a novel model named Bayesian Ranking for MiRNA-Disease Association prediction (BRMDA) by improving Bayesian Personalized Ranking from three aspects: (i) taking advantage of similarity of diseases and miRNAs; (ii) incorporating miRNA bias for miRNAs associated with different number of diseases; and (iii) implementing neighborhood-based approach for new miRNAs and diseases. For each investigated disease, BRMDA used the set of triples (i.e. disease, labeled miRNA, unlabeled miRNA) that reflected association preference of the disease to miRNAs as training set, which made full use of unknown samples rather than simply considering them as negative samples. To investigate the predictive performance of BRMDA, we employed leave-one-out cross-validation and obtained Area Under the Curve of 0.8697, which outperformed many classical methods. Besides, we further implemented three distinct classes of case studies for three common Neoplasms. As a result, there are 44 (Colon Neoplasms), 49 (Esophageal Neoplasms) and 49 (Lung Neoplasms) among the top 50 predicted miRNAs validated through experiments. In short, BRMDA would be a trustable tool for inferring valuable associations.

Prediction of lncRNA-Protein Interactions via the Multiple Information Integration.

The long non-coding RNA (lncRNA)–protein interaction plays an important role in the post-transcriptional gene regulation, such as RNA splicing, translation, signaling, and the development of complex diseases. The related research on the prediction of lncRNA–protein interaction relationship is beneficial in the excavation and the discovery of the mechanism of lncRNA function and action occurrence, which are important. Traditional experimental methods for detecting lncRNA–protein interactions are expensive and time-consuming. Therefore, computational methods provide many effective strategies to deal with this problem. In recent years, most computational methods only use the information of the lncRNA–lncRNA or the protein–protein similarity and cannot fully capture all features to identify their interactions. In this paper, we propose a novel computational model for the lncRNA–protein prediction on the basis of machine learning methods. First, a feature method is proposed for representing the information of the network topological properties of lncRNA and protein interactions. The basic composition feature information and evolutionary information based on protein, the lncRNA sequence feature information, and the lncRNA expression profile information are extracted. Finally, the above feature information is fused, and the optimized feature vector is used with the recursive feature elimination algorithm. The optimized feature vectors are input to the support vector machine (SVM) model. Experimental results show that the proposed method has good effectiveness and accuracy in the lncRNA–protein interaction prediction.

Assessing the impacts of climate change and habitat suitability on the distribution and quality of medicinal plant using multiple information integration: Take Gentiana rigescens as an example

Gentiana rigescens is an economically important medicinal plant in the subtropical zone in Asia. However, it is unclear how its distribution responds to climate change in the future. This study aims to simulate the effect of climate change on the distribution of G. rigescens by using the maximum entropy (MaxEnt) model and to evaluate the potential impact of habitat suitability on the quality of medicinal materials vis multiple chemical information. In the study, we investigated the suitable habitat fluctuation ranges of G. rigescens under different future climate scenarios at first. Subsequently, the chemical geographical variation of medicinal materials collected from different suitable areas was evaluated based on high-performance liquid chromatography (HPLC) analysis and Fourier-transform mid-infrared (FT-MIR) spectroscopy. The results showed that the areas of highly suitable habitat for G. rigescens was about 23.17 × 104 km2, and more than 98.88% of highly suitable habitat was distributed in the southwest of China. The survival of G. rigescens may be threatened by global climate change in the future. Under the Representative Concentration Pathways 8.5 (RCP8.5) climate scenario, the change ratio of the highly suitable habitat will be −99.22% in 2070. Meanwhile, the centroids of the highly suitable area showed a westward-shifting trend under future emission trajectories. HPLC analysis showed that medicinal materials from highly suitable areas contained higher concentrations of total iridoid glycoside. Analysis results of FT-MIR data combined with OPLS-DA modeling indicated that the overall chemical composition of samples collected from different suitable habitats were different, and the habitat suitability had a stronger effect on the chemical composition in leaf than that in stem and root. Based on the concentration of iridoids and FT-MIR fingerprint, our research demonstrated there was a certain correlation of habitat suitability with the overall chemical variation of G. rigescens. The highly suitable habitat was better for the cultivation of high-quality medicinal materials. In summary, these findings would improve our understanding of the distribution and quality variation of medicinal Gentiana under future climate change. Meanwhile, the study approach based on multiple information integration could provide a useful case for research of other medicinal plants.

A network embedding-based multiple information integration method for the MiRNA-disease association prediction

BackgroundMiRNAs play significant roles in many fundamental and important biological processes, and predicting potential miRNA-disease associations makes contributions to understanding the molecular mechanism of human diseases. Existing state-of-the-art methods make use of miRNA-target associations, miRNA-family associations, miRNA functional similarity, disease semantic similarity and known miRNA-disease associations, but the known miRNA-disease associations are not well exploited.ResultsIn this paper, a network embedding-based multiple information integration method (NEMII) is proposed for the miRNA-disease association prediction. First, known miRNA-disease associations are formulated as a bipartite network, and the network embedding method Structural Deep Network Embedding (SDNE) is adopted to learn embeddings of nodes in the bipartite network. Second, the embedding representations of miRNAs and diseases are combined with biological features about miRNAs and diseases (miRNA-family associations and disease semantic similarities) to represent miRNA-disease pairs. Third, the prediction models are constructed based on the miRNA-disease pairs by using the random forest. In computational experiments, NEMII achieves high-accuracy performances and outperforms other state-of-the-art methods: GRNMF, NTSMDA and PBMDA. The usefulness of NEMII is further validated by case studies. The studies demonstrate the great potential of network embedding method for the miRNA-disease association prediction, and SDNE outperforms other popular network embedding methods: DeepWalk, High-Order Proximity preserved Embedding (HOPE) and Laplacian Eigenmaps (LE).ConclusionWe propose a new method, named NEMII, for predicting miRNA-disease associations, which has great potential to benefit the field of miRNA-disease association prediction.

Identification of drug-side effect association via multiple information integration with centered kernel alignment

Abstract In medicine research, drug discovery aims to develop a drug to patients who will benefit from it and try to avoid some side effects. However, the tradition experiment is time consuming and expensive. In recent years, computational approaches provide many effective strategies to deal with this issue. In fact, the known associations between drugs and side-effects are less than unknown associations, thus it can be seen as an imbalance classification problem. Although several classification methods have been developed to predict drug-side effect associations, the performance of predictors could also be further improved. In this paper, we propose a novel predictor of drug-side effect associations. First, we construct multiple kernels from drug space and side-effect space, respectively. Then, these corresponding kernels are linear weighted by optimized Centered Kernel Alignment-based Multiple Kernel Learning (CKA-MKL) algorithm in two different spaces. At last, Kronecker Regularized Least Squares (Kronecker RLS) is employed to fuse drug kernel and side-effect kernel, further identify drug-side effect associations. Compared with many existing methods, our proposed approach achieves better results on three benchmark datasets of drug side-effect associations. The values of Area Under the Precision Recall curve (AUPR) are 0.672, 0.679 and 0.675 on Pauwels’s dataset, Mizutani’s dataset and Liu’s dataset, respectively. The AUPRs are improved by at least 0.012, 0.013 and 0.014 on three different datasets. Experimental results show that our method has outstanding performance among other excellent approaches on identifying drug-side effect associations.

Indoor person position estimation system based on integration of multiple information

Indoor person position estimation system based on integration of multiple information

Influence of neural adaptation on dynamics and equilibrium state of neural activities in a ring neural network

How neural adaptation affects neural information processing (i.e. the dynamics and equilibrium state of neural activities) is a central question in computational neuroscience. In my previous works, I analytically clarified the dynamics and equilibrium state of neural activities in a ring-type neural network model that is widely used to model the visual cortex, motor cortex, and several other brain regions. The neural dynamics and the equilibrium state in the neural network model corresponded to a Bayesian computation and statistically optimal multiple information integration, respectively, under a biologically inspired condition. These results were revealed in an analytically tractable manner; however, adaptation effects were not considered. Here, I analytically reveal how the dynamics and equilibrium state of neural activities in a ring neural network are influenced by spike-frequency adaptation (SFA). SFA is an adaptation that causes gradual inhibition of neural activity when a sustained stimulus is applied, and the strength of this inhibition depends on neural activities. I reveal that SFA plays three roles: (1) SFA amplifies the influence of external input in neural dynamics; (2) SFA allows the history of the external input to affect neural dynamics; and (3) the equilibrium state corresponds to the statistically optimal multiple information integration independent of the existence of SFA. In addition, the equilibrium state in a ring neural network model corresponds to the statistically optimal integration of multiple information sources under biologically inspired conditions, independent of the existence of SFA.

Analysis of early-warning threshold for metro construction collapse risk based on D-S evidence theory and rough set

The existing early-warning system in metro construction are generally based on traditional single-sensor data and simple analytic model, which makes it difficult to deal with the complex and comprehensive environment in metro construction. In this paper, the framework of early-warning threshold for metro construction collapse risk based on D-S evidence theory and rough set is built. By combining the primary data fusion collected based on rough set with the secondary data fusion which is based on D-S evidence theory, the integration of multiple information in metro construction is realized and the risk assessment methods are optimized. A case trial based on Hangzhou metro construction collapse accident is also carried out to exemplify the framework. The empirical analysis guarantees the completeness and independence of the prediction information, and realizes the dynamic prediction of the variation trend of metro construction collapse risk.

Identification of drug-target interactions via multiple information integration

Identifying Drug-Target Interactions (DTIs) is an important process in drug discovery. Traditional experimental methods are expensive and time-consuming for detecting DTIs. Therefore, computational approaches provide many effective strategies to deal with this issue. In recent years, most of computational methods only use the information of drug-drug similarity or target-target similarity, which cannot perfectly capture all characteristics to identify DTIs. In this paper, we propose a novel computational model of DTIs prediction, based on machine learning methods. To improve the performance of prediction, we further use molecular substructure fingerprints, Multivariate Mutual Information (MMI) of proteins and network topology to represent drugs, targets and relationship between them. Moreover, we employ Support Vector Machine (SVM) and Feature Selection (FS) to construct model for predicting DTIs. Experiments of evaluation show that proposed approach achieves better results than other outstanding methods for feature-based DTIs prediction. The proposed approach achieves AUPRs of 0.899, 0.929, 0.821 and 0.655 on Enzyme, Ion Channel (IC), GPCR and Nuclear Receptor datasets, respectively. Compared with existing best methods, AUPRs are increased by 0.016 on Ion Channel datasets. In addition, our method obtains the second best performance on GPCR and Enzyme datasets. The source code and all datasets are available at https://figshare.com/s/53bf5a6065f3911d46f6.

To Resolve or Not To Resolve, that Is the Question: The Dual-Path Model of Incongruity Resolution and Absurd Verbal Humor by fMRI.

It is well accepted that the humor comprehension processing involves incongruity detection and resolution and then induces a feeling of amusement. However, this three-stage model of humor processing does not apply to absurd humor (so-called nonsense humor). Absurd humor contains an unresolvable incongruity but can still induce a feeling of mirth. In this study, we used functional magnetic resonance imaging (fMRI) to identify the neural mechanisms of absurd humor. Specifically, we aimed to investigate the neural substrates associated with the complete resolution of incongruity resolution humor and partial resolution of absurd humor. Based on the fMRI data, we propose a dual-path model of incongruity resolution and absurd verbal humor. According to this model, the detection and resolution for the incongruity of incongruity resolution humor activate brain regions involved in the temporo-parietal lobe (TPJ) implicated in the integration of multiple information and precuneus, likely to be involved in the ability of perspective taking. The appreciation of incongruity resolution humor activates regions the posterior cingulate cortex (PCC), implicated in autobiographic or event memory retrieval, and parahippocampal gyrus (PHG), implying the funny feeling. By contrast, the partial resolution of absurd humor elicits greater activation in the fusiform gyrus which have been implicated in word processing, inferior frontal gyrus (IFG) for the process of incongruity resolution and superior temporal gyrus (STG) for the pragmatic awareness.

Drug-pathway interaction prediction via multiple feature fusion.

Predicting new drug-pathway interactions from heterogeneous biological data is important not only for the understanding of various drug response and molecular interaction processes, but also for the development of novel drugs and the therapy of human diseases. In this paper, three different learning methods including the Bipartite Local Models method (BLM), Gaussian Interaction Profiles kernels (GIP) method and Graph-based Semi-supervised Learning method (GBSSL) were used to predict drug-pathway interactions. To realize the purpose, drugs were firstly represented by functional groups and chemical structure similarity, and pathways were represented by their related gene expressions and semantic similarity based features. Then, the parameter optimization procedures were further adopted to deal with heterogeneous data sources. As a result, the proposed methods achieved a high ROC curve score (AUC score) over 0.95, which validated the effectiveness of multiple information integration. Moreover, several new potential drug-pathway interactions were identified for further biological function research.

Structure-centered portal for child psychiatry research.

The real world needs of the clinical community require a domain-specific solution to integrate disparate information available from various web-based resources for data, materials, and tools into routine clinical and clinical research setting. We present a child-psychiatry oriented portal as an effort to deliver a knowledge environment wrapper that provides organization and integration of multiple information and data sources. Organized semantically by resource context, the portal groups information sources by context type, and permits the user to interactively “narrow” or “broaden” the scope of the information resources that are available and relevant to the specific context. The overall objective of the portal is to bring information from multiple complex resources into a simple single uniform framework and present it to the user in a single window format.

Predicting residue contacts for protein-protein interactions by integration of multiple information

Detailed knowledge of interfacial region between interacting proteins is not only helpful in annotating function for proteins, but also very important for structure-based drug design and disease treatment. However, this is one of the most difficult tasks and current methods are constrained by some factors. In this study, we developed a new method to predict residue-residue contacts of two interacting protein domains by integrating information about evolutionary couplings andamino acid pairwise contact potentials, as well as domain-domain interaction interfaces. The experimental results showed that our proposed method outperformed the previous method with the same datasets. Moreover, the method promises an improvement in the source of template-based protein docking.

Assessing and predicting protein interactions by combining manifold embedding with multiple information integration.

BackgroundProtein-protein interactions (PPIs) play crucial roles in virtually every aspect of cellular function within an organism. Over the last decade, the development of novel high-throughput techniques has resulted in enormous amounts of data and provided valuable resources for studying protein interactions. However, these high-throughput protein interaction data are often associated with high false positive and false negative rates. It is therefore highly desirable to develop scalable methods to identify these errors from the computational perspective.ResultsWe have developed a robust computational technique for assessing the reliability of interactions and predicting new interactions by combining manifold embedding with multiple information integration. Validation of the proposed method was performed with extensive experiments on densely-connected and sparse PPI networks of yeast respectively. Results demonstrate that the interactions ranked top by our method have high functional homogeneity and localization coherence.ConclusionsOur proposed method achieves better performances than the existing methods no matter assessing or predicting protein interactions. Furthermore, our method is general enough to work over a variety of PPI networks irrespectively of densely-connected or sparse PPI network. Therefore, the proposed algorithm is a much more promising method to detect both false positive and false negative interactions in PPI networks.

224 情報統合のための不整合情報の棄却方法(機械の知能化)

224 情報統合のための不整合情報の棄却方法(機械の知能化)