scHiCStackL: a stacking ensemble learning-based method for single-cell Hi-C classification using cell embedding.

Single-cell Hi-C data are a common data source for studying the differences in the three-dimensional structure of cell chromosomes. The development of single-cell Hi-C technology makes it possible to obtain batches of single-cell Hi-C data. How to quickly and effectively discriminate cell types has become one hot research field. However, the existing computational methods to predict cell types based on Hi-C data are found to be low in accuracy. Therefore, we propose a high accuracy cell classification algorithm, called scHiCStackL, based on single-cell Hi-C data. In our work, we first improve the existing data preprocessing method for single-cell Hi-C data, which allows the generated cell embedding better to represent cells. Then, we construct a two-layer stacking ensemble model for classifying cells. Experimental results show that the cell embedding generated by our data preprocessing method increases by 0.23, 1.22, 1.46 and 1.61$\%$ comparing with the cell embedding generated by the previously published method scHiCluster, in terms of the Acc, MCC, F1 and Precision confidence intervals, respectively, on the task of classifying human cells in the ML1 and ML3 datasets. When using the two-layer stacking ensemble framework with the cell embedding, scHiCStackL improves by 13.33, 19, 19.27 and 14.5 over the scHiCluster, in terms of the Acc, ARI, NMI and F1 confidence intervals, respectively. In summary, scHiCStackL achieves superior performance in predicting cell types using the single-cell Hi-C data. The webserver and source code of scHiCStackL are freely available at http://hww.sdu.edu.cn:8002/scHiCStackL/ and https://github.com/HaoWuLab-Bioinformatics/scHiCStackL, respectively.

CLNN-loop: a deep learning model to predict CTCF-mediated chromatin loops in the different cell lines and CTCF-binding sites (CBS) pair types.

MOTIVATION: Three-dimensional (3D) genome organization is of vital importance in gene regulation and disease mechanisms. Previous studies have shown that CTCF-mediated chromatin loops are crucial to studying the 3D structure of cells. Although various experimental techniques have been developed to detect chromatin loops, they have been found to be time-consuming and costly. Nowadays, various sequence-based computational methods can capture significant features of 3D genome organization and help predict chromatin loops. However, these methods have low performance and poor generalization ability in predicting chromatin loops. RESULTS: Here, we propose a novel deep learning model, called CLNN-loop, to predict chromatin loops in different cell lines and CTCF-binding sites (CBS) pair types by fusing multiple sequence-based features. The analysis of a series of examinations based on the datasets in the previous study shows that CLNN-loop has satisfactory performance and is superior to the existing methods in terms of predicting chromatin loops. In addition, we apply the SHAP framework to interpret the predictions of different models, and find that CTCF motif and sequence conservation are important signs of chromatin loops in different cell lines and CBS pair types. AVAILABILITY AND IMPLEMENTATION: The source code of CLNN-loop is freely available at https://github.com/HaoWuLab-Bioinformatics/CLNN-loop and the webserver of CLNN-loop is freely available at http://hwclnn.sdu.edu.cn. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Integrating Protein-Protein Interaction Networks and Somatic Mutation Data to Detect Driver Modules in Pan-Cancer.

With the constant update of large-scale sequencing data and the continuous improvement of cancer genomics data, such as International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA), it gains increasing importance to detect the functional high-frequency mutation gene set in cells that causes cancer in the field of medicine. In this study, we propose a new recognition method of driver modules, named ECSWalk to solve the issue of mutated gene heterogeneity and improve the accuracy of driver modules detection, based on human protein-protein interaction networks and pan-cancer somatic mutation data. This study first utilizes high mutual exclusivity and high coverage between mutation genes and topological structure similarity of the nodes in complex networks to calculate interaction weights between genes. Second, the method of random walk with restart is utilized to construct a weighted directed network, and the strong connectivity principle of the directed graph is utilized to create the initial candidate modules with a certain number of genes. Finally, the large modules in the candidate modules are split using induced subgraph method, and the small modules are expanded using a greedy strategy to obtain the optimal driver modules. This method is applied to TCGA pan-cancer data and the experimental results show that ECSWalk can detect driver modules more effectively and accurately, and can identify new candidate gene sets with higher biological relevance and statistical significance than MEXCOWalk and HotNet2. Thus, ECSWalk is of theoretical implication and practical value for cancer diagnosis, treatment and drug targets.

Integrating multi-omics data to identify dysregulated modules in endometrial cancer.

Cancer is generally caused by genetic mutations, and differentially expressed genes are closely associated with genetic mutations. Therefore, mutated genes and differentially expressed genes can be used to study the dysregulated modules in cancer. However, it has become a big challenge in cancer research how to accurately and effectively detect dysregulated modules that promote cancer in massive data. In this study, we propose a network-based method for identifying dysregulated modules (Netkmeans). Firstly, the study constructs an undirected-weighted gene network based on the characteristics of high mutual exclusivity, high coverage and complex network topology among genes widely existed in the genome. Secondly, the study constructs a comprehensive evaluation function to select the number of clusters scientifically and effectively. Finally, the K-means clustering method is applied to detect the dysregulated modules. Compared with the results detected by IBA and CCEN methods, the results of Netkmeans proposed in this study have higher statistical significance and biological relevance. Besides, compared with the dysregulated modules detected by MCODE, CFinder and ClusterONE, the results of Netkmeans have higher accuracy, precision and F-measure. The experimental results show that the multiple dysregulated modules detected by Netkmeans are essential in the generation, development and progression of cancer, and thus they play a vital role in the precise diagnosis, treatment and development of new medications for cancer patients.

Identifying driver modules based on multi-omics biological networks in prostate cancer.

The development of sequencing technology has promoted the expansion of cancer genome data. It is necessary to identify the pathogenesis of cancer at the molecular level and explore reliable treatment methods and precise drug targets in cancer by identifying carcinogenic functional modules in massive multi-omics data. However, there are still limitations to identifying carcinogenic driver modules by utilising genetic characteristics simply. Therefore, this study proposes a computational method, NetAP, to identify driver modules in prostate cancer. Firstly, high mutual exclusivity, high coverage, and high topological similarity between genes are integrated to construct a weight function, which calculates the weight of gene pairs in a biological network. Secondly, the random walk method is utilised to reevaluate the strength of interaction among genes. Finally, the optimal driver modules are identified by utilising the affinity propagation algorithm. According to the results, the authors' method identifies more validated driver genes and driver modules compared with the other previous methods. Thus, the proposed NetAP method can identify carcinogenic driver modules effectively and reliably, and the experimental results provide a powerful basis for cancer diagnosis, treatment and drug targets.

Senescence and cell death in chronic liver injury: roles and mechanisms underlying hepatocarcinogenesis.

Chronic liver injury (CLI) is a complex pathological process typically characterized by progressive destruction and regeneration of liver parenchymal cells due to diverse risk factors such as alcohol abuse, drug toxicity, viral infection, and genetic metabolic disorders. When the damage to hepatocytes is mild, the liver can regenerate itself and restore to the normal state; when the damage is irreparable, hepatocytes would undergo senescence or various forms of death including apoptosis, necrosis and necroptosis. These pathological changes not only promote the progression of the existing hepatopathies via various underlying mechanisms but are closely associated with hepatocarcinogenesis. In this review, we discuss the pathological changes that hepatocytes undergo during CLI, and their roles and mechanisms in the progression of hepatopathies and hepatocarcinogenesis. We also give a brief introduction about some animal models currently used for the research of CLI and progress in the research of CLI.

The extent of liver injury determines hepatocyte fate toward senescence or cancer.

It is well known that induction of hepatocyte senescence could inhibit the development of hepatocellular carcinoma (HCC). Until now, it is still unclear how the degree of liver injury dictates hepatocyte senescence and carcinogenesis. In this study, we investigated whether the severity of injury determines cell fate decisions between hepatocyte senescence and carcinogenesis. After testing of different degrees of liver injury, we found that hepatocyte senescence is strongly induced in the setting of severe acute liver injury. Longer-term, moderate liver injury, on the contrary did not result into hepatocyte senescence, but led to a significant incidence of HCC instead. In addition, carcinogenesis was significantly reduced by the induction of severe acute injury after chronic moderate liver injury. Meanwhile, immune surveillance, especially the activations of macrophages, was activated after re-induction of senescence by severe acute liver injury. We conclude that severe acute liver injury leads to hepatocyte senescence along with activating immune surveillance and a low incidence of HCC, whereas chronic moderate injury allows hepatocytes to proliferate rather than to enter into senescence, and correlates with a high incidence of HCC. This study improves our understanding in hepatocyte cell fate decisions and suggests a potential clinical strategy to induce senescence to treat HCC.