Yuehan He

Identification of breast cancer patients based on human signaling network motifs

Identifying breast cancer patients is crucial to the clinical diagnosis and therapy for this disease. Conventional gene-based methods for breast cancer diagnosis ignore gene-gene interactions and thus may lead to loss of power. In this study, we proposed a novel method to select classification features, called “Selection of Significant Expression-Correlation Differential Motifs” (SSECDM). This method applied a network motif-based approach, combining a human signaling network and high-throughput gene expression data to distinguish breast cancer samples from normal samples. Our method has higher classification performance and better classification accuracy stability than the mutual information (MI) method or the individual gene sets method. It may become a useful tool for identifying and treating patients with breast cancer and other cancers, thus contributing to clinical diagnosis and therapy for these diseases.

Uncovering packaging features of co-regulated modules based on human protein interaction and transcriptional regulatory networks

BackgroundNetwork co-regulated modules are believed to have the functionality of packaging multiple biological entities, and can thus be assumed to coordinate many biological functions in their network neighbouring regions.ResultsHere, we weighted edges of a human protein interaction network and a transcriptional regulatory network to construct an integrated network, and introduce a probabilistic model and a bipartite graph framework to exploit human co-regulated modules and uncover their specific features in packaging different biological entities (genes, protein complexes or metabolic pathways). Finally, we identified 96 human co-regulated modules based on this method, and evaluate its effectiveness by comparing it with four other methods.ConclusionsDysfunctions in co-regulated interactions often occur in the development of cancer. Therefore, we focussed on an example co-regulated module and found that it could integrate a number of cancer-related genes. This was extended to causal dysfunctions of some complexes maintained by several physically interacting proteins, thus coordinating several metabolic pathways that directly underlie cancer.

Unraveling the characteristics of microRNA regulation in the developmental and aging process of the human brain

BackgroundStructure and function of the human brain are subjected to dramatic changes during its development and aging. Studies have demonstrated that microRNAs (miRNAs) play an important role in the regulation of brain development and have a significant impact on brain aging and neurodegeneration. However, the underling molecular mechanisms are not well understood. In general, development and aging are conventionally studied separately, which may not completely address the physiological mechanism over the entire lifespan. Thus, we study the regulatory effect between miRNAs and mRNAs in the developmental and aging process of the human brain by integrating miRNA and mRNA expression profiles throughout the lifetime.MethodsIn this study, we integrated miRNA and mRNA expression profiles in the human brain across lifespan from the network perspective. First, we chose the age-related miRNAs by polynomial regression models. Second, we constructed the bipartite miRNA-mRNA regulatory network by pair-wise correlation coefficient analysis between miRNA and mRNA expression profiles. At last, we constructed the miRNA-miRNA synergistic network from the miRNA-mRNA network, considering not only the enrichment of target genes but also GO function enrichment of co-regulated target genes.ResultsWe found that the average degree of age-related miRNAs was significantly higher than that of non age-related miRNAs in the miRNA-mRNA regulatory network. The topological features between age-related and non age-related miRNAs were significantly different, and 34 reliable age-related miRNA synergistic modules were identified using Cfinder in the miRNA-miRNA synergistic network. The synergistic regulations of module genes were verified by reviewing miRNA target databases and previous studies.ConclusionsAge-related miRNAs play a more important role than non age-related mrRNAs in the developmental and aging process of the human brain. The age-related miRNAs have synergism, which tend to work together as small modules. These results may provide a new insight into the regulation of miRNAs in the developmental and aging process of the human brain.

Identification of cancer risk lncRNAs and cancer risk pathways regulated by cancer risk lncRNAs based on genome sequencing data in human cancers

Cancer is a group of diseases involving abnormal cell growth with the potential to invade or spread to other parts of the body. The complexity of cancer can be reduced to a small number of underlying principles like cancer hallmarks which could govern the transformation of normal cells to cancer. Besides, the growth and metastasis of cancer often relate to combined effects of long non-coding RNAs (lncRNAs). Here, we performed comprehensive analysis for lncRNA expression profiles and clinical data of six types of human cancer patients from The Cancer Genome Atlas (TCGA), and identified six risk pathways and twenty three lncRNAs. In addition, twenty three cancer risk lncRNAs which were closely related to the occurrence or development of cancer had a good classification performance for samples of testing datasets of six cancer datasets. More important, these lncRNAs were able to separate samples in the entire cancer dataset into high-risk group and low-risk group with significantly different overall survival (OS), which was further validated in ten validation datasets. In our study, the robust and effective cancer biomarkers were obtained from cancer datasets which had information of normal-tumor samples. Overall, our research can provide a new perspective for the further study of clinical diagnosis and treatment of cancer.

Cancer-risk module identification and module-based disease risk evaluation: a case study on lung cancer.

Gene expression profiles have drawn broad attention in deciphering the pathogenesis of human cancers. Cancer-related gene modules could be identified in co-expression networks and be applied to facilitate cancer research and clinical diagnosis. In this paper, a new method was proposed to identify lung cancer-risk modules and evaluate the module-based disease risks of samples. The results showed that thirty one cancer-risk modules were closely related to the lung cancer genes at the functional level and interactional level, indicating that these modules and genes might synergistically lead to the occurrence of lung cancer. Our method was proved to have good robustness by evaluating the disease risk of samples in eight cancer expression profiles (four for lung cancer and four for other cancers), and had better performance than the WGCNA method. This method could provide assistance to the diagnosis and treatment of cancers and a new clue for explaining cancer mechanisms.

Identification of polycystic ovary syndrome potential drug targets based on pathobiological similarity in the protein-protein interaction network

Polycystic ovary syndrome (PCOS) is one of the most common endocrinological disorders in reproductive aged women. PCOS and Type 2 Diabetes (T2D) are closely linked in multiple levels and possess high pathobiological similarity. Here, we put forward a new computational approach based on the pathobiological similarity to identify PCOS potential drug target modules (PPDT-Modules) and PCOS potential drug targets in the protein-protein interaction network (PPIN). From the systems level and biological background, 1 PPDT-Module and 22 PCOS potential drug targets were identified, 21 of which were verified by literatures to be associated with the pathogenesis of PCOS. 42 drugs targeting to 13 PCOS potential drug targets were investigated experimentally or clinically for PCOS. Evaluated by independent datasets, the whole PPDT-Module and 22 PCOS potential drug targets could not only reveal the drug response, but also distinguish the statuses between normal and disease. Our identified PPDT-Module and PCOS potential drug targets would shed light on the treatment of PCOS. And our approach would provide valuable insights to research on the pathogenesis and drug response of other diseases.

A Novel Prioritization Method in Identifying Recurrent Venous Thromboembolism-Related Genes.

Identifying the genes involved in venous thromboembolism (VTE) recurrence is important not only for understanding the pathogenesis but also for discovering the therapeutic targets. We proposed a novel prioritization method called Function-Interaction-Pearson (FIP) by creating gene-disease similarity scores to prioritize candidate genes underling VTE. The scores were calculated by integrating and optimizing three types of resources including gene expression, gene ontology and protein-protein interaction. As a result, 124 out of top 200 prioritized candidate genes had been confirmed in literature, among which there were 34 antithrombotic drug targets. Compared with two well-known gene prioritization tools Endeavour and ToppNet, FIP was shown to have better performance. The approach provides a valuable alternative for drug targets discovery and disease therapy.

Identifying colon cancer risk modules with better classification performance based on human signaling network.

Identifying differences between normal and tumor samples from a modular perspective may help to improve our understanding of the mechanisms responsible for colon cancer. Many cancer studies have shown that signaling transduction and biological pathways are disturbed in disease states, and expression profiles can distinguish variations in diseases. In this study, we integrated a weighted human signaling network and gene expression profiles to select risk modules associated with tumor conditions. Risk modules as classification features by our method had a better classification performance than other methods, and one risk module for colon cancer had a good classification performance for distinguishing between normal/tumor samples and between tumor stages. All genes in the module were annotated to the biological process of positive regulation of cell proliferation, and were highly associated with colon cancer. These results suggested that these genes might be the potential risk genes for colon cancer.

Identification of Modules Related to Programmed Cell Death in CHD Based on EHEN

The formation and death of macrophages and foam cells are one of the major factors that cause coronary heart disease (CHD). In our study, based on the Edinburgh Human Metabolic Network (EHMN) metabolic network, we built an enzyme network which was constructed by enzymes (nodes) and reactions (edges) called the Edinburgh Human Enzyme Network (EHEN). By integrating the subcellular location information for the reactions and refining the protein-reaction relationships based on the location information, we proposed a computational approach to select modules related to programmed cell death. The identified module was in the EHEN-mitochondria (EHEN-M) and was confirmed to be related to programmed cell death, CHD pathogenesis, and lipid metabolism in the literature. We expected this method could analyze CHD better and more comprehensively from the point of programmed cell death in subnetworks.

Identification of breast cancer prognostic modules based on weighted protein‑protein interaction networks

Breast cancer is one of the leading causes of mortality in females. A number of prognostic markers have been identified, including single genes, multi-gene signatures and network modules; however, the robustness of these prognostic markers is insufficient. Thus, the present study proposed a more robust method to identify breast cancer prognostic modules based on weighted protein-protein interaction networks, by integrating four sets of disease-associated expression profiles. Three identified prognostic modules were closely associated with prognosis-associated functions and survival time, as determined by Cox regression and Kaplan-Meier survival analyses. The robustness of these modules was verified with an independent profile from another platform. Genes from these modules may be useful as breast cancer prognostic markers. The prognostic modules could be used to determine the prognoses of patients with breast cancer and characterize patient recovery.