Integrative computational and experimental approaches to establish a post-myocardial infarction knowledge map

Classification of Cancer Types Using Graph Convolutional Neural Networks

BACKGROUND

Cancer has been a leading cause of death in the United States with significant health care costs. Accurate prediction of cancers at an early stage and understanding the genomic mechanisms that drive cancer development are vital to the improvement of treatment outcomes and survival rates, thus resulting in significant social and economic impacts. Attempts have been made to classify cancer types with machine learning techniques during the past two decades and deep learning approaches more recently.

RESULTS

In this paper, we established four models with graph convolutional neural network (GCNN) that use unstructured gene expressions as inputs to classify different tumor and non-tumor samples into their designated 33 cancer types or as normal. Four GCNN models based on a co-expression graph, co-expression+singleton graph, protein-protein interaction (PPI) graph, and PPI+singleton graph have been designed and implemented. They were trained and tested on combined 10,340 cancer samples and 731 normal tissue samples from The Cancer Genome Atlas (TCGA) dataset. The established GCNN models achieved excellent prediction accuracies (89.9-94.7%) among 34 classes (33 cancer types and a normal group). In silico gene-perturbation experiments were performed on four models based on co-expression graph, co-expression+singleton, PPI graph, and PPI+singleton graphs. The co-expression GCNN model was further interpreted to identify a total of 428 markers genes that drive the classification of 33 cancer types and normal. The concordance of differential expressions of these markers between the represented cancer type and others are confirmed. Successful classification of cancer types and a normal group regardless of normal tissues' origin suggested that the identified markers are cancer-specific rather than tissue-specific.

CONCLUSION

Novel GCNN models have been established to predict cancer types or normal tissue based on gene expression profiles. We demonstrated the results from the TCGA dataset that these models can produce accurate classification (above 94%), using cancer-specific markers genes. The models and the source codes are publicly available and can be readily adapted to the diagnosis of cancer and other diseases by the data-driven modeling research community.

Corin Overexpression Reduces Myocardial Infarct Size and Modulates Cardiomyocyte Apoptotic Cell Death

Altered expression of corin, a cardiac transmembrane serine protease, has been linked to dilated and ischemic cardiomyopathy. However, the potential role of corin in myocardial infarction (MI) is lacking. This study examined the outcomes of MI in wild-type vs. cardiac-specific overexpressed corin transgenic (Corin-Tg) mice during pre-MI, early phase (3, 24, 72 h), and late phase (1, 4 weeks) post-MI. Corin overexpression significantly reduced cardiac cell apoptosis (p < 0.001), infarct size (p < 0.001), and inhibited cleavage of procaspases 3, 9, and 8 (p < 0.05 to p < 0.01), as well as altered the expression of Bcl2 family proteins, Bcl-xl, Bcl2 and Bak (p < 0.05 to p < 0.001) at 24 h post-MI. Overexpressed cardiac corin also significantly modulated heart function (ejection fraction, p < 0.0001), lung congestion (lung weight to body weight ratio, p < 0.0001), and systemic extracellular water (edema, p < 0.05) during late phase post-MI. Overall, cardiac corin overexpression significantly reduced apoptosis, infarct size, and modulated cardiac expression of key members of the apoptotic pathway in early phase post-MI; and led to significant improvement in heart function and reduced congestion in late phase post-MI. These findings suggest that corin may be a useful target to protect the heart from ischemic injury and subsequent post-infarction remodeling.

Integrative network analysis reveals time-dependent molecular events underlying left ventricular remodeling in post-myocardial infarction patients

To elucidate the time-resolved molecular events underlying the LV remodeling (LVR) process, we developed a large-scale network model that integrates the 24 molecular variables (plasma proteins and non-coding RNAs) collected in the REVE-2 study at four time points (baseline, 1month, 3months and 1year) after MI. The REVE-2 network model was built by extending the set of REVE-2 variables with their mechanistic context based on known molecular interactions (1310 nodes and 8639 edges). Changes in the molecular variables between the group of patients with high LVR (>20%) and low LVR (<20%) were used to identify active network modules within the clusters associated with progression of LVR, enabling assessment of time-resolved molecular changes. Although the majority of molecular changes occur at the baseline, two network modules specifically show an increasing number of active molecules throughout the post-MI follow up: one involved in muscle filament sliding, containing the major troponin forms and tropomyosin proteins, and the other associated with extracellular matrix disassembly, including matrix metalloproteinases, tissue inhibitors of metalloproteinases and laminin proteins. For the first time, integrative network analysis of molecular variables collected in REVE-2 patients with known molecular interactions allows insight into time-dependent mechanisms associated with LVR following MI, linking specific processes with LV structure alteration. In addition, the REVE-2 network model provides a shortlist of prioritized putative novel biomarker candidates for detection of LVR after MI event associated with a high risk of heart failure and is a valuable resource for further hypothesis generation.

Knowledge gaps to understanding cardiac macrophage polarization following myocardial infarction

Following myocardial infarction (MI), macrophages coordinate both pro-inflammatory and reparative responses of the left ventricle (LV) by reacting to and secreting cytokines, chemokines, and growth factors and by stimulating endothelial cells and fibroblasts to modulate neovascularization and scar formation. Healing of the infarcted LV can be divided into three distinct, but overlapping phases: inflammatory, proliferative, and maturation. Macrophages are involved in all phases. Despite macrophages being a major leukocyte cell type in the post-MI LV, how this cell type regulates LV remodeling over the post-MI time continuum is not completely understood. In this review, we summarize the current literature as a foundation to discuss the major knowledge gaps that remain. Defining the post-MI temporal macrophage phenotypes to establish a classification system is the first step in exploring how macrophage phenotypes are regulated, how temporal stimulation and secretion profiles evolve, and how best to modify stimuli to yield predictable cell responses. This article is part of a Special Issue entitled: The role of post-translational protein modifications on heart and vascular metabolism edited by Jason R.B. Dyck & Jan F.C. Glatz.

Transformative Impact of Proteomics on Cardiovascular Health and Disease: A Scientific Statement From the American Heart Association

The year 2014 marked the 20th anniversary of the coining of the term proteomics. The purpose of this scientific statement is to summarize advances over this period that have catalyzed our capacity to address the experimental, translational, and clinical implications of proteomics as applied to cardiovascular health and disease and to evaluate the current status of the field. Key successes that have energized the field are delineated; opportunities for proteomics to drive basic science research, facilitate clinical translation, and establish diagnostic and therapeutic healthcare algorithms are discussed; and challenges that remain to be solved before proteomic technologies can be readily translated from scientific discoveries to meaningful advances in cardiovascular care are addressed. Proteomics is the result of disruptive technologies, namely, mass spectrometry and database searching, which drove protein analysis from 1 protein at a time to protein mixture analyses that enable large-scale analysis of proteins and facilitate paradigm shifts in biological concepts that address important clinical questions. Over the past 20 years, the field of proteomics has matured, yet it is still developing rapidly. The scope of this statement will extend beyond the reaches of a typical review article and offer guidance on the use of next-generation proteomics for future scientific discovery in the basic research laboratory and clinical settings.

Systems analysis of gene ontology and biological pathways involved in post-myocardial infarction responses

Background

Pathway analysis has been widely used to gain insight into essential mechanisms of the response to myocardial infarction (MI). Currently, there exist multiple pathway databases that organize molecular datasets and manually curate pathway maps for biological interpretation at varying forms of organization. However, inconsistencies among different databases in pathway descriptions, frequently due to conflicting results in the literature, can generate incorrect interpretations. Furthermore, although pathway analysis software provides detailed images of interactions among molecules, it does not exhibit how pathways interact with one another or with other biological processes under specific conditions.

Methods

We propose a novel method to standardize descriptions of enriched pathways for a set of genes/proteins using Gene Ontology terms. We used this method to examine the relationships among pathways and biological processes for a set of condition-specific genes/proteins, represented as a functional biological pathway-process network. We applied this algorithm to a set of 613 MI-specific proteins we previously identified.

Results

A total of 96 pathways from Biocarta, KEGG, and Reactome, and 448 Gene Ontology Biological Processes were enriched with these 613 proteins. The pathways were represented as Boolean functions of biological processes, delivering an interactive scheme to organize enriched information with an emphasis on involvement of biological processes in pathways. We extracted a network focusing on MI to demonstrate that tyrosine phosphorylation of Signal Transducer and Activator of Transcription (STAT) protein, positive regulation of collagen metabolic process, coagulation, and positive/negative regulation of blood coagulation have immediate impacts on the MI response.

Conclusions

Our method organized biological processes and pathways in an unbiased approach to provide an intuitive way to identify biological properties of pathways under specific conditions. Pathways from different databases have similar descriptions yet diverse biological processes, indicating variation in their ability to share similar functional characteristics. The coverages of pathways can be expanded with the incorporation of more biological processes, predicting involvement of protein members in pathways. Further, detailed analyses of the functional biological pathway-process network will allow researchers and scientists to explore critical routes in biological systems in the progression of disease.

Unraveling the exercise-related proteome signature in heart

Exercise training is a well-known non-pharmacological strategy for the prevention and treatment of cardiovascular diseases. Despite the established phenotypic knowledge, the molecular signature of exercise-induced cardiac remodeling remains poorly characterized. The great majority of studies dedicated to this topic use conventional reductionist methods, which only allow analyzing individual protein candidates. Nowadays, several methodologies based on mass spectrometry are available and have been successfully applied for the characterization of heart proteome, representing an attractive approach for the wide characterization of the complex molecular networks that underlie exercise-induced cardiac remodeling. Still, few studies have used these methodologies to understand the impact of exercise training on the remodeling of cardiac proteome. The present study analyzes the few available data obtained from mass spectrometry (MS)-based proteomic studies assessing the impact of distinct types of exercise training on the protein profile of heart (left ventricle and isolated mitochondria) and the potential cross-tolerance between exercise training and diseases as myocardial infarction and obesity. Network analysis was performed with bioinformatics to integrate data from distinct research papers, based on distinct exercise training protocols, animal models and methodological approaches applied in the characterization of heart proteome. The analysis revealed that exercise training confers a unique proteome signature characterized by the up-regulation of lipid and organic metabolic processes, vasculogenesis and tissue regeneration. Data retrieved from this analysis also suggested that cardiac mitochondrial proteome is highly dynamic in response to exercise training due, in part, to the action of specific kinases as PKA and PKG. Regarding to the type of exercise, treadmill training seems to have a greater effect on the modulation of cardiac proteome than swimming. Data from the present review will certainly open new perspectives on cardiac proteomics and will help to envisage future studies targeting the identification of the regulatory mechanisms underlying cardiac adaptive and maladaptive remodeling.

Integrative analysis of ocular complications in atherosclerosis unveils pathway convergence and crosstalk

Atherosclerosis is a life-threatening disease and a major cause of mortalities worldwide. While many of the atherosclerotic sequelae are reflected as microvascular effects in the eye, the molecular mechanisms of their development is not yet known. In this study, we employed a systems biology approach to unveil the most significant events and key molecular mediators of ophthalmic sequelae caused by atherosclerosis. Literature mining was used to identify the proteins involved in both atherosclerosis and ophthalmic diseases. A protein-protein interaction (PPI) network was prepared using the literature-mined seed nodes. Network topological analysis was carried out using Cytoscape, while network nodes were annotated using database for annotation, visualization and integrated discovery in order to identify the most enriched pathways and processes. Network analysis revealed that mitogen-activated protein kinase 1 (MAPK1) and protein kinase C occur with highest betweenness centrality, degree and closeness centrality, thus reflecting their functional importance to the network. Our analysis shows that atherosclerosis-associated ophthalmic complications are caused by the convergence of neurotrophin signaling pathways, multiple immune response pathways and focal adhesion pathway on the MAPK signaling pathway. The PPI network shares features with vasoregression, a process underlying multiple vascular eye diseases. Our study presents a first clear and composite picture of the components and crosstalk of the main pathways of atherosclerosis-induced ocular diseases. The hub bottleneck nodes highlight the presence of molecules important for mediating the ophthalmic complications of atherosclerosis and contain five established drug targets for future therapeutic modulation efforts.