Consensus clustering algorithm based on the automatic partitioning similarity graph

The highest level of interface design for online games is to achieve invisibility, which means that the interface is very natural and easy to use. Players are proficient in using it and fully immerse themselves in the world of online games. It seems that the interface is transparent and they no longer feel its existence. Excellent online game interface design is the requirement of users that runs through the entire design process. But this is not to say that the usability of online game software can surpass other factors. All great designs seek a balance and harmony between artistic beauty, reliability, security, usability, cost, and performance. The quality evaluation of online games interface design is a multiple-attribute group decision-making (MAGDM) problem. Recently, the Evaluation Based on Distance from Average Solution (EDAS) technique, cosine similarity measure (CSM), Dice similarity measure (DSM), Jaccard similarity measure (JSM) and entropy technique has been separately employed to cope with MAGDM issues. The probabilistic linguistic term sets (PLTSs) are employed as a tool for conveying uncertain information during the quality evaluation of online games interface design. In this paper, the EDAS technique is expanded to the PLTSs and the probabilistic linguistic EDAS (PL-EDAS) technique based on cosine similarity measure (CSM), Dice similarity measure (DSM) and Jaccard similarity measure (JSM) is constructed to manage MAGDM issue. The information entropy technique is employed to implement the weight values based on CSM technique, DSM technique and JSM technique under PLTSs. Finally, the quality evaluation of online games interface design is employed to demonstrate the PL-EDAS technique and some comparative analysis is employed to demonstrate the PL-EDAS technique. Thus, the main research contribution of this work is constructed: (1) the information entropy technique is constructed to implement the attribute weight values based on CSM technique, DSM technique and JSM technique; (2) the PL-EDAS technique is constructed under PLTSs based on CSM technique, DSM technique and JSM technique; (3) an example for quality evaluation of online games interface design is employed to verify PL-EDAS technique and several decision comparative analysis is employed to verify the PL-EDAS technique.

We present superpixel-based segmentation frameworks for unsupervised and semi-supervised epithelium-stroma identification in histopathological images or oropharyngeal tissue micro arrays. A superpixel segmentation algorithm is initially used to split-up the image into binary regions (superpixels) and their colour features are extracted and fed into several base clustering algorithms with various parameter initializations. Two Consensus Clustering (CC) formulations are then used: the Evidence Accumulation Clustering (EAC) and the voting-based consensus function. These combine the base clustering outcomes to obtain a more robust detection of tissue compartments than the base clustering methods on their own. For the voting-based function, a technique is introduced to generate consistent labellings across the base clustering results. The obtained CC result is then utilized to build a self-training Semi-Supervised Classification (SSC) model. Unlike supervised segmentations, which rely on large number of labelled training images, our SSC approach performs a quality segmentation while relying on few labelled samples. Experiments conducted on forty-five hand-annotated images of oropharyngeal cancer tissue microarrays show that (a) the CC algorithm generates more accurate and stable results than individual clustering algorithms; (b) the clustering performance of the voting-based function outperforms the existing EAC; and (c) the proposed SSC algorithm outperforms the supervised methods, which is trained with only a few labelled instances.

Fuzzy set (FS) theory is one of the most important tool to deasl with complicated and difficult information in real-world. Now FS has many extensions and hesitant fuzzy set (HFS) is one of them. Further generalization of FS is complex fuzzy set (CFS), which contains only the membership grade, whose range is unit disc instead of [0, 1]. The aim of this paper is to present the idea of complex hesitant fuzzy set (CHFS) and to introduce its basic properties. Basically, CHFS is the combination of CFS and HFS to deal with two dimension information in a single set. Further, the vector similarity measures (VSMs) such as Jaccard similarity measures (JSMs), Dice similarity measures (DSMs) and Cosine similarity measures (CSMs) for CHFSs are discussed. The special cases of the proposed measures are also discussed. Then, the notion of complex hesitant fuzzy hybrid vector similarity measures are utilized in the environment of pattern recognition and medical diagnosis. Further, based on these distance measures, a decision-making method has been presented for finding the best alternative under the set of the feasible one. Illustrative examples from the field of pattern recognition as well as medical diagnosis have been taken to validate the approach. Finally, the comparison between proposed approaches with existing approaches are also discussed to find the reliability and proficiency of the elaborated measures for complex hesitant fuzzy elements.

Consensus clustering methods are motivated by the success of combining multiple classifiers in many areas. In this paper, graph-based consensus clustering is used to improve the quality of chemical compound clustering by enhancing the robustness, novelty, consistency and stability of individual clusterings. For this purpose, Hyper-Graph Partitioning Algorithm (HGPA) [1], was applied. The clustering is evaluated based on the ability to separate actives from inactives molecules in each cluster and the results were compared with the Ward's clustering method. The chemical dataset MDL Drug Data Report (MDDR) database has been used for experiments. The MDL Drug Data Report (MDDR) database consists of 102516 molecules. For the experiments, the dataset DS1 was chosen from the MDDR database. This dataset has been used for many virtual screening experiments [2-4]. The dataset DS1contains 10 heterogeneous activity classes (8568 molecules). For the clustering experiments, two 2D fingerprint descriptors will be used which are developed by Scitegic's Pipeline Pilot [5]. These are 120-bit ALOGP and 1024-bit extended connectivity fingerprints (ECFP_4). The results were evaluated based on the effectiveness of the methods to separate actives from non-actives molecules using QPI- (for quality partition index) measure, which was devised by Varin et al. [6]. As defined by [7], an active cluster as a non-singleton cluster for which the percentage of active molecules in the cluster is greater than the percentage of active molecules in the dataset as a whole. Let p be the number of actives in active clusters, q the number of inactives in active clusters, r the number of actives in inactive clusters (i.e., clusters that are not active clusters) and s the number of singleton actives. The high value occurs when the actives are clustered tightly together and separated from the inactive molecules. Then the quality partition index, QPI, is defined to be: QPI=pp+q+r+s (1) Then, the results will be compared with Ward's individual clustering method, the standard clustering method for chemoinformatics applications. The generation process has been done by multiple run of K-means algorithms, each with random initialization of cluster centroids. The number of partitions generated in this step is ranged between n = 5 to n = 50, with 5-times step. Then, all the generated partitions were combined using HGPA to obtain the consensus partition. This process is done for each fingerprint (ALOGP and ECFP_4). The mean of QPI values are averaged over the ten activity classes of the datasets. Tables ​Tables1,1, ​,22 show the effectiveness of MDDR dataset clustering using ALOGP and ECFP_4 fingerprints. The best PQI value of consensus clustering methods for each column has been bold-faced for ease of reference. Table 1 Effectivenss of clustering of high diverse MDDR dataset: ALOGP Fingerprint. Table 2 Effectivenss of clustering of high diverse MDDR dataset: ECFP_4 Fingerprint. Visual inspection of the results enables comparisons to be made between the effectiveness of clustering of MDDR datasets and Ward's method, the best of choice clustering method for chemoinformatics applications. In addition, ten times of consensus clustering, for each fingerprint were observed in order to study the effectiveness of consensus clustering with different ensemble sizes. The results show that HGPA consensus clustering gives robust and novel result when K-means algorithm is run 20-50 times using ALOGP. The performance of consensus clustering outperforms the Wards' method. For consensus of dataset which represented by ECFP_4 fingerprint, the best QPI values of consensus clustering are obtained from ensembles of size n = 20-50. The performance of consensus clustering gives robust results which are better than overall performance of individual clusterings. The values of QPI in both datasets for consensus clustering are close to the Wards method. The consensus clustering, HGPA, provide stable clusters by decreasing the sensitivity to noise and outliers. The average percentages of singleton clusters of individual clusterings compared with consensus clustering for both fingerprints. The results show that consensus clustering partition the datasets with average percentage of singleton equal to zero, which is much better than individual clusterings and Wards' method. For example, 16.72% of molecules of DS1 are clustered as singletons when Wards method is applied on ALOGP fingerprint with number of clusters equal to 1000 clusters. Finally we conclude that graph-based consensus clustering can improve the effectiveness of chemical compounds clustering. The performance of consensus clustering is more robust, novel, stable, consistent, and out-perform Wards' method in case of using ALOGP fingerprint. By using ECFP_4 fingerprint, consensus clustering methods provide more robust, stable, and consistent clustering and close to the Wards clustering results. The experiments reported here suggest that graph-based consensus clustering can improve the quality of individual clustering by using the efficient algorithm, K-means algorithm, to generate the ensemble with size (20-50) for both structurally diverse chemical datasets.

Phenotypic Consensus Clustering and Treatment Heterogeneity Analysis in Critically Ill Patients with Comorbid Type 2 Diabetes Mellitus.

Atherosclerosis (AS) is the leading cause of coronary heart disease, which is the primary cause of death worldwide. Recent studies have identified disulfidptosis as a new type of cell death that may be involved in onset and development of many diseases. However, the role of disulfidptosis in AS is not clear. In this study, bioinformatics analysis and experiments in vivo and in vitro were performed to evaluate the potential relationship between disulfidptosis and AS. AS-related sequencing data were obtained from the Gene Expression Omnibus (GEO). Bioinformatics techniques were used to evaluate differentially expressed genes (DEGs) associated with disulfidptosis-related AS. Hub genes were screened using least absolute shrinkage and selection operator (LASSO) and random forests (RF) methods. In addition, we established a foam cell model in vitro and an AS mouse model in vivo to verify the expressions of hub genes. In addition, we constructed a diagnostic nomogram with hub genes to predict progression of AS. Finally, the consensus clustering method was used to establish two different subtypes, and associations between subtypes and immunity were explored. As the results, 9 disulfidptosis-related AS DEGs were identified from GSE28829 and GSE43292 datasets. Evaluation of DEGs using LASSO and RF methods resulted in identification of 4 hub genes (CAPZB, DSTN, MYL6, PDLIM1), which were analyzed for diagnostic value using ROC curve analysis and verified in vitro and in vivo. Furthermore, a nomogram including hub genes was established that accurately predicted the occurrence of AS. The consensus clustering algorithm was used to separate patients with early atherosclerotic plaques and patients with advanced atherosclerotic plaques into two disulfidptosis subtypes. Cluster B displayed higher levels of infiltrating immune cells, which indicated that patients in cluster B may have a positive immune response for progression of AS. In summary, disulfidptosis-related genes including CAPZB, DSTN, MYL6, and PDLIM1 may be diagnostic markers and therapeutic targets for AS. In addition, these genes are closely related to immune cells, which may inform immunotherapy for AS.

The accuracy of similarity measurement between sentences is critical to the performance of several applications such as text mining, question answering, and text summarization. This paper focuses on calculating semantic similarities between sentences and performing a comparative analysis among identified similarity measurement techniques. Comparison between three popular similarity measurements which are Jaccard, Cosine and Dice similarity measures has been conducted. The performance of each identified measurement was evaluated and recorded. In this paper, we use a large lexical database of English known as WordNet to calculate the word-to-word semantic similarity. The result of this research concludes that the Jaccard and Dice performs better in measuring the semantic similarity between sentences.

Accurate determination of joint roughness coefficient (JRC) of rock joints is essential for evaluating the influence of surface roughness on the shear behavior of rock joints. The JRC values of rock joints are typically measured by visual comparison against Barton’s standard JRC profiles. However, its accuracy is strongly affected by personal bias. In the present study, a new comparison method is proposed for JRC evaluation to overcome the drawback of conventional visual comparison methods based on vector similarity measures (VSMs). The feature vectors are obtained by analyzing the angular variation of line segments of both standard JRC profiles and test profiles obtained from three kinds of natural rocks with a sampling interval of 0.5 mm. The roughness similarity degrees between test profiles and standard profiles are evaluated by the Jaccard, Dice, and cosine similarity measures. The JRC values of the test profiles are then determined according to the maximum relation index based on the similarity degrees. In the present study, a comparative analysis between the VSMs method and the JRC evaluation method using different roughness parameters demonstrated that the VSMs method is effective and accurate for JRC measurement.

Consensus clustering provides a framework to ensemble multiple clustering results to obtain a consensus and robust result. Most existing consensus clustering methods usually apply all data to ensemble learning, whereas ignoring the side effects caused by some difficult or unreliable instances. To tackle this problem, we propose a novel self-paced consensus clustering method to gradually involve instances from more reliable to less reliable ones into the ensemble learning. We first construct an initial bipartite graph from the multiple base clustering results, where the nodes represent the instances and clusters and the edges indicate that an instance belongs to a cluster. Then, we learn a structured bipartite graph from the initial one by self-paced learning, i.e., we automatically decide the reliability of each edge and involves the edges into graph learning in order of their reliability. At last, we obtain the final consensus clustering result from the learned bipartite graph. The extensive experimental results demonstrate the effectiveness and superiority of the proposed method.

BackgroundAlternative splicing (AS) plays an essential role in tumorigenesis and progression. This study intended to construct an innovative prognostic model based on AS events to gain more precise survival prediction and search for potential therapeutic targets in ovarian cancer.MethodsSeven types of AS events in ovarian serous cystadenocarcinoma (OV) patients with RNA-seq were obtained using The Cancer Genome Atlas (TCGA) SpliceSeq tool and database. Cox and Kaplan-Meier curve analyses were employed to establish the prognostic models. Relying on drug sensitivity data from the CellMiner database, Genomics of Drug Sensitivity (GDS) was adopted to estimate the platinum-sensitive analysis. Furthermore, a prognostic splicing factor (SF)-AS network was constructed using Cytoscape. Finally, in order to explore the influence of the tumor microenvironment on the prognosis of OV patients, we first combined a similar network fusion and consensus clustering (SNF-CC) algorithm to identify three OV subtypes based on survival-related AS events and then utilized single-sample Gene Set Enrichment Analysis (ssGSEA) method to perform immune cell infiltration analysis.ResultsA total of 48,049 AS events and 21,841 related genes were selected from 318 OV samples, and 2,206 AS events associated with disease-free survival (DFS) were identified. Multivariate Cox and Kaplan-Meier curve analyses were then employed to establish the prognostic models. Receiver operating characteristic (ROC) analysis from 0.59 to 0.75 showed that these models were highly efficient in distinguishing patient survival. GDS was adopted with the CellMiner database to provide some insights for platinum-sensitive analysis of OV. Furthermore, a prognostic SF-AS network, which discovered a significant connection between SFs and prognostic AS genes, was constructed using Cytoscape. The combined SNF-CC algorithm revealed three distinct OV subtypes based on the prognostic AS events, and the associations between this novel molecular classification and immune cell infiltration were further explored.ConclusionsWe developed a powerful prognostic AS signature for OV and provided a deeper understanding of SF-AS network regulatory mechanisms, as well as platinum-sensitive and cancer immune microenvironments. These results revealed various candidate biomarkers and potential targets for OV treatment strategies.

BackgroundGrowing evidence demonstrated that m6A modification in cardiovascular diseases. However, how it is involved in the intracranial aneurysm (IA) is still unclear. This study aimed to identify the role of m6A modification in IA.MethodsThree datasets downloaded from the Gene Expression Omnibus (GEO) database were used, including GSE122897, GSE15629, and GSE3679. The landscapes of 24 m6A regulators were depicted using the STRING database, Pearson’s correlation analysis, and Wilcoxon test. The targets of differentially expressed m6A (DEm6A) were predicted in the m6A2Target database and the modification m6A sites of hub targets were identified in SRAMP online tool. A diagnostic model based on DEm6A was constructed and verified in training and test databases. A consensus clustering algorithm was performed to classify IA patients into distinct m6A-related clusters. Functional analyses including gene ontology, Kyoto Encyclopedia of Genes and Genomes (KEGG), gene set variation analysis, and gene set enrichment analysis analyses were conducted to elucidate the underlying mechanisms. ssGSEA algorithm was performed to uncover the immune characteristics. A PCA method was adopted to quantify the m6A score.ResultsNine DEm6A (IGF2BP1, IGF2BP3, YTHDF2, ZNF217, RBM15, YTHDF3, YTHDC1, FTO, and LRPPRC) significantly differed between IA and controls. Biological annotations showed that immune-related pathways (such as complement activation, inflammatory response, and interleukin signaling) and apoptosis were more enriched in IAs than in controls. Immune analyses indicate that the abundance of immune cells, immune responses, and HLA gene expression were elevated in IA samples than in controls. PCA results showed that IA has a lower m6A score than controls. An immune/apoptosis-related network modified by DEm6A was constructed. The m6A sites of six hub targets (CDK1, ASPM, AURKB, BUB1B, MKI67, and TPX2) were predicted with very high confidence. A diagnostic model with four genes (LRPPRC, YTHDF3, IGF2BP1, and ZNF217) was constructed and verified. Two m6A modification subtypes were identified with unsupervised cluster analysis. Immune infiltration analysis revealed that cluster 1 had higher immune activation than cluster 2. Further study showed that cluster 1 had a larger proportion of ruptured IAs.ConclusionThe m6A modification may shape the IAs microenvironment and participates in the formation and rupture of IAs by regulating immune infiltration.

Prognosis prediction and drug guidance of ovarian serous cystadenocarcinoma through mitochondria gene-based model.

Background: Bladder urothelial carcinoma (BLCA) is the most common type of bladder cancer. In this study, the correlation between the metabolic status and the outcome of patients with BLCA was evaluated using data from the Cancer Genome Atlas and Gene Expression Omnibus datasets. Methods: The clinical and transcriptomic data of patients with BLCA were downloaded from the Cancer Genome Atlas and cBioPortal datasets, and energy metabolism-related gene sets were obtained from the Molecular Signature Database. A consensus clustering algorithm was then conducted to classify the patients into two clusters. Tumor prognosis, clinicopathological features, mutations, functional analysis, ferroptosis status analysis, immune infiltration, immune checkpoint-related gene expression level, chemotherapy resistance, and tumor stem cells were analyzed between clusters. An energy metabolism-related signature was further developed and verified using data from cBioPortal datasets. Results: Two clusters (C1 and C2) were identified using a consensus clustering algorithm based on an energy metabolism-related signature. The patients with subtype C1 had more metabolism-related pathways, different ferroptosis status, higher cancer stem cell scores, higher chemotherapy resistance, and better prognosis. Subtype C2 was characterized by an increased number of advanced BLCA cases and immune-related pathways. Higher immune and stromal scores were also observed for the C2 subtype. A signature containing 16 energy metabolism-related genes was then identified, which can accurately predict the prognosis of patients with BLCA. Conclusion: We found that the energy metabolism-associated subtypes of BLCA are closely related to the immune microenvironment, immune checkpoint-related gene expression, ferroptosis status, CSCs, chemotherapy resistance, prognosis, and progression of BLCA patients. The established energy metabolism-related gene signature was able to predict survival in patients with BLCA.

In the pursuit of enhancing yield efficiency, mitigating environmental impact, and reducing costs of fertilizers and fuel, precision farming emerges as a pivotal strategy. The application of nutrients must be tailored according to spatial and temporal variations. This requires a comprehensive understanding of the nutrient composition of organic fertilizers, the nutrient supply of plants, and the soil’s capacity. To optimize fertilizer application in the field, it is recommendable to subdivide the field into management zones, thereby identifying distinct zones characterized by uniform growing conditions. To establish management zones, we combined satellite-based phenological-dependent timeseries of vegetation, with proximal soil sensor data from a multi-sensor platform. The zones were generated through a multi-step clustering algorithm, based on hierarchical clustering, which results were combined by a consensus clustering algorithm. Four different scenarios of input datasets were tested. The first scenario incorporates all scenes during the timeseries, followed by the one with selected scenes during specific phenology stages. Another scenario was based solely on soil information. The fourth scenario involves phenologically distributed vegetation and soil information. For the validation we calculated the variance for the input datasets per cluster, lying under one scenario. Our hypothesis that the clustering based on soil and phenology separated vegetation data would improve the management zones was refuted. The vegetation cluster, which was based on the entire Sentinel-2 timeseries, exhibited optimal performance, for one field in Groß Kreutz, Germany. The management zones are interpreted as recommendations for farmers to adapt the management practices within the framework of possibilities.

BackgroundRNA N6-methyladenosine (m6A) methylation, the most abundant and prominent form of epigenetic modification, is involved in hepatocellular carcinoma (HCC) initiation and progression. However, the role of m6A methylation in HCC tumor microenvironment (TME) formation is unexplored. This study aimed to reveal the TME features of HCC patients with distinct m6A expression patterns and establish a prognostic model based on m6A signatures for HCC cohorts.Material/MethodsWe classified the m6A methylation patterns in 365 HCC samples based on 21 m6A modulators using a consensus clustering algorithm. Single-sample gene set enrichment analysis algorithm was used to quantify the abundance of immune cell infiltration. Gene set variation analysis revealed the biological characteristics between the m6A modification patterns. The m6A-based prognostic model was constructed using a training set with least absolute shrinkage and selection operator regression and validated in internal and external datasets.ResultsTwo distinct m6A modification patterns exhibiting different TME immune-infiltrating characteristics, heterogeneity, and prognostic variations were identified in the HCC cohort. After depicting the immune landscape of TME in HCC, we found patients with high LRPPRC m6A modulator expression had depletion of T cells, cytotoxic cells, dendritic cells, and cytolytic activity response. A high m6A score, characterized by suppression of immunity, indicated an immune-excluded TME phenotype, with poor survival. A nomogram was developed to facilitate HCC clinical decision making.ConclusionsOur results highlight the nonnegligible role of m6A methylation in TME formation and reveal a potential clinical application of the m6A-associated prognostic model for patients with HCC.