Identification of a multi-cancer gene expression biomarker for cancer clinical outcomes using a network-based algorithm

Deep learning enabled integration of tumor microenvironment microbial profiles and host gene expressions for interpretable survival subtyping in diverse types of cancers

The tumor microbiome, a complex community of microbes found in tumors, has been found to be linked to cancer development, progression, and treatment outcome. However, it remains a bottleneck in distangling the relationship between the tumor microbiome and host gene expressions in tumor microenvironment, as well as their concert effects on patient survival. In this study, we aimed to decode this complex relationship by developing ASD-cancer (autoencoder-based subtypes detector for cancer), a semi-supervised deep learning framework that could extract survival-related features from tumor microbiome and transcriptome data, and identify patients' survival subtypes. By using tissue samples from The Cancer Genome Atlas database, we identified two statistically distinct survival subtypes across all 20 types of cancer Our framework provided improved risk stratification (e.g., for liver hepatocellular carcinoma, [LIHC], log-rank test, P = 8.12E-6) compared to PCA (e.g., for LIHC, log-rank test, P = 0.87), predicted survival subtypes accurately, and identified biomarkers for survival subtypes. Additionally, we identified potential interactions between microbes and host genes that may play roles in survival. For instance, in LIHC, Arcobacter, Methylocella, and Isoptericola may regulate host survival through interactions with host genes enriched in the HIF-1 signaling pathway, indicating these species as potential therapy targets. Further experiments on validation data sets have also supported these patterns. Collectively, ASD-cancer has enabled accurate survival subtyping and biomarker discovery, which could facilitate personalized treatment for broad-spectrum types of cancers.IMPORTANCEUnraveling the intricate relationship between the tumor microbiome, host gene expressions, and their collective impact on cancer outcomes is paramount for advancing personalized treatment strategies. Our study introduces ASD-cancer, a cutting-edge autoencoder-based subtype detector. ASD-cancer decodes the complexities within the tumor microenvironment, successfully identifying distinct survival subtypes across 20 cancer types. Its superior risk stratification, demonstrated by significant improvements over traditional methods like principal component analysis, holds promise for refining patient prognosis. Accurate survival subtype predictions, biomarker discovery, and insights into microbe-host gene interactions elevate ASD-cancer as a powerful tool for advancing precision medicine. These findings not only contribute to a deeper understanding of the tumor microenvironment but also open avenues for personalized interventions across diverse cancer types, underscoring the transformative potential of ASD-cancer in shaping the future of cancer care.

Construction of a prognostic model based on cuproptosis-related patterns for predicting survival, immune infiltration, and immunotherapy efficacy in breast cancer: Cuproptosis-based prognostic modeling in breast cancer

Breast cancer is the most common and lethal malignancy among women worldwide. Cuproptosis, a newly identified copper-dependent cell death, is closely associated with cancer development. However, its regulatory mechanisms in breast cancer are not well studied. This study aims to establish a prognostic model for breast cancer to improve risk stratification. The mRNA expression data was downloaded from The Cancer Genome Atlas and Gene Expression Omnibus databases. Consensus clustering identified patterns based on cuproptosis-related genes. Key genes were screened using Weighted Gene Co-Expression Network Analysis and differentially expressed gene analysis. A prognostic model was constructed using Cox regression and evaluated with time-dependent receiver operating characteristic and Kaplan-Meier analyses. Functional pathways, immune cell infiltration, and other tumor characteristics were also analyzed. Two distinct cuproptosis patterns were identified. The top 21 differentially expressed genes, significantly associated with survival, were used to construct the prognostic model. The risk score has a negative correlation with survival. Enrichment analysis showed immune-related pathways enriched in the low-risk group, which also had more immune cell infiltration, higher stromal component, lower tumor purity, and lower tumor heterogeneity. Finally, significant differences of half maximal inhibitory concentration were also observed between patients in high- and low-risk groups who received chemotherapy and targeted therapy drugs. These findings in our study may provide evidence for further research and individualized management of breast cancer.

Multi-Biomarker Profiling for Precision Diagnosis of Lung Cancer

Multi-biomarker analysis can enhance the accuracy of the single-biomarker analysis by reducing the errors caused by genetic and environmental differences. For this reason, multi-biomarker analysis shows higher accuracy in early and precision diagnosis. However, conventional analysis methods have limitations for multi-biomarker analysis because of their long pre-processing times, inconsistent results, and large sample requirements. To solve these, a fast and accurate precision diagnostic method is introduced for lung cancer by multi-biomarker profiling using a single drop of blood. For this, surface-enhanced Raman spectroscopic immunoassay (SERSIA) is employed for the accurate, quick, and reliable quantification of biomarkers. Then, it is checked the statistical relation of the multi-biomarkers to differentiate between healthy controls and lung cancer patients. This approach has proven effective; with 20 µL of blood serum, lung cancer is diagnosed with 92% accuracy. It also accurately identifies the type and stage of cancer with 87% and 85%, respectively. These results show the importance of multi-biomarker analysis in overcoming the challenges posed by single-biomarker diagnostics. Furthermore, it markedly improves multi-biomarker-based analysis methods, illustrating its important impact on clinical diagnostics.

Pan-Cancer, Genome-Scale Metabolic Network Analysis of over 10,000 Patients Elucidates Relationship between Metabolism and Survival

Despite the high variability in cancer biology, cancers nevertheless exhibit cohesive hallmarks across multiple cancer types, notably dysregulated metabolism. Metabolism plays a central role in cancer biology, and shifts in metabolic pathways have been linked to tumor aggressiveness and likelihood of response to therapy. We therefore sought to interrogate metabolism across cancer types and understand how intrinsic modes of metabolism vary within and across indications and how they relate to patient prognosis. We used context specific genome-scale metabolic modeling to simulate metabolism across 10,915 patients from 34 cancer types from The Cancer Genome Atlas and the MMRF-COMMPASS study. We found that cancer metabolism clustered into modes characterized by differential glycolysis, oxidative phosphorylation, and growth rate. We also found that the simulated activities of metabolic pathways are intrinsically prognostic across cancer types, especially tumor growth rate, fatty acid biosynthesis, folate metabolism, oxidative phosphorylation, steroid metabolism, and glutathione metabolism. This work shows the prognostic power of individual patient metabolic modeling across multiple cancer types. Additionally, it shows that analyzing large-scale models of cancer metabolism with survival information provides unique insights into underlying relationships across cancer types and suggests how therapies designed for one cancer type may be repurposed for use in others.

TFRC, associated with hypoxia and immune, is a prognostic factor and potential therapeutic target for bladder cancer

BACKGROUND

Bladder cancer is a common malignancy of the urinary system, and the survival rate and recurrence rate of patients with muscular aggressive (MIBC) bladder cancer are not ideal. Hypoxia is a pathological process in which cells acquire special characteristics to adapt to anoxic environment, which can directly affect the proliferation, invasion and immune response of bladder cancer cells. Understanding the exact effects of hypoxia and immune-related genes in BLCA is helpful for early assessment of the prognosis of BLCA. However, the prognostic model of BLCA based on hypoxia and immune-related genes has not been reported.

PURPOSE

Hypoxia and immune cell have important role in the prognosis of bladder cancer (BLCA). The aim of this study was to investigate whether hypoxia and immune related genes could be a novel tools to predict the overall survival and immunotherapy of BLCA patients.

METHODS

First, we downloaded transcriptomic data and clinical information of BLCA patients from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. A combined hypoxia and immune signature was then constructed on the basis of the training cohort via least absolute shrinkage and selection operator (LASSO) analysis and validated in test cohort. Afterwards, Kaplan-Meier curves, univariate and multivariate Cox and subgroup analysis were employed to assess the accuracy of our signature. Immune cell infiltration, checkpoint and the Tumor Immune Dysfunction and Exclusion (TIDE) algorithm were used to investigate the immune environment and immunotherapy of BLCA patients. Furthermore, we confirmed the role of TFRC in bladder cancer cell lines T24 and UMUC-3 through cell experiments.

RESULTS

A combined hypoxia and immune signature containing 8 genes were successfully established. High-risk group in both training and test cohorts had significantly poorer OS than low-risk group. Univariate and multivariate Cox analysis indicated our signature could be regarded as an independent prognostic factor. Different checkpoint was differently expressed between two groups, including CTLA4, HAVCR2, LAG3, PD-L1 and PDCD1. TIDE analysis indicated high-risk patients had poor response to immunotherapy and easier to have immune escape. The drug sensitivity analysis showed that high-risk group patients were more potentially sensitive to many drugs. Meanwhile, TFRC could inhibit the proliferation and invasion ability of T24 and UMUC-3 cells.

CONCLUSION

A combined hypoxia and immune-related gene could be a novel predictive model for OS and immunotherapy estimation of BLCA patients and TFRC could be used as a potential therapeutic target in the future.

Host-response transcriptional biomarkers accurately discriminate bacterial and viral infections of global relevance

Diagnostic limitations challenge management of clinically indistinguishable acute infectious illness globally. Gene expression classification models show great promise distinguishing causes of fever. We generated transcriptional data for a 294-participant (USA, Sri Lanka) discovery cohort with adjudicated viral or bacterial infections of diverse etiology or non-infectious disease mimics. We then derived and cross-validated gene expression classifiers including: 1) a single model to distinguish bacterial vs. viral (Global Fever-Bacterial/Viral [GF-B/V]) and 2) a two-model system to discriminate bacterial and viral in the context of noninfection (Global Fever-Bacterial/Viral/Non-infectious [GF-B/V/N]). We then translated to a multiplex RT-PCR assay and independent validation involved 101 participants (USA, Sri Lanka, Australia, Cambodia, Tanzania). The GF-B/V model discriminated bacterial from viral infection in the discovery cohort an area under the receiver operator curve (AUROC) of 0.93. Validation in an independent cohort demonstrated the GF-B/V model had an AUROC of 0.84 (95% CI 0.76-0.90) with overall accuracy of 81.6% (95% CI 72.7-88.5). Performance did not vary with age, demographics, or site. Host transcriptional response diagnostics distinguish bacterial and viral illness across global sites with diverse endemic pathogens.

Rapid in situ RNA imaging based on Cas12a thrusting strand displacement reaction

RNA In situ imaging through DNA self-assembly is advantaged in illustrating its structures and functions with high-resolution, while the limited reaction efficiency and time-consuming operation hinder its clinical application. Here, we first proposed a new strand displacement reaction (SDR) model (Cas12a thrusting SDR, CtSDR), in which Cas12a could overcome the inherent reaction limitation and dramatically enhance efficiency through energy replenishment and by-product consumption. The target-initiated CtSDR amplification was established for RNA analysis, with order of magnitude lower limit of detection (LOD) than the Cas13a system. The CtSDR-based RNA in situ imaging strategy was developed to monitor intra-cellular microRNA expression change and delineate the landscape of oncogenic RNA in 66 clinic tissue samples, possessing a clear advantage over classic in situ hybridization (ISH) in terms of operation time (1 h versus 14 h) while showing comparable sensitivity and specificity. This work presents a promising approach to developing advanced molecular diagnostic tools.

Debiased inference for heterogeneous subpopulations in a high-dimensional logistic regression model

Due to the prevalence of complex data, data heterogeneity is often observed in contemporary scientific studies and various applications. Motivated by studies on cancer cell lines, we consider the analysis of heterogeneous subpopulations with binary responses and high-dimensional covariates. In many practical scenarios, it is common to use a single regression model for the entire data set. To do this effectively, it is critical to quantify the heterogeneity of the effect of covariates across subpopulations through appropriate statistical inference. However, the high dimensionality and discrete nature of the data can lead to challenges in inference. Therefore, we propose a novel statistical inference method for a high-dimensional logistic regression model that accounts for heterogeneous subpopulations. Our primary goal is to investigate heterogeneity across subpopulations by testing the equivalence of the effect of a covariate and the significance of the overall effects of a covariate. To achieve overall sparsity of the coefficients and their fusions across subpopulations, we employ a fused group Lasso penalization method. In addition, we develop a statistical inference method that incorporates bias correction of the proposed penalized method. To address computational issues due to the nonlinear log-likelihood and the fused Lasso penalty, we propose a computationally efficient and fast algorithm by adapting the ideas of the proximal gradient method and the alternating direction method of multipliers (ADMM) to our settings. Furthermore, we develop non-asymptotic analyses for the proposed fused group Lasso and prove that the debiased test statistics admit chi-squared approximations even in the presence of high-dimensional variables. In simulations, the proposed test outperforms existing methods. The practical effectiveness of the proposed method is demonstrated by analyzing data from the Cancer Cell Line Encyclopedia (CCLE).

The endocytic receptor protein LRP-1 modulate P-glycoprotein mediated drug resistance in MCF-7 cells

Multidrug resistance (MDR) is a major obstacle to successful cancer chemotherapy. A typical form of MDR is due to the overexpression of membrane transport proteins., such as Glycoprotein-P (P-gp), resulting in an increased drug efflux preventing drug cytotoxicity. P-gp is mainly localized on the plasma membrane; however, it can also be endocytosed resulting in the trafficking of P-gp in endoplasmic reticulum, Golgi, endosomes, and lysosomes. The lysosomal P-gp has been found to be capable of transporting and sequestering P-gp substrates (e.g., Doxorubicin (Dox)) into lysosomes to protect cells against cytotoxic drugs. Many translational studies have shown that low-density lipoprotein receptor-related protein-1 (LRP-1) is involved in endocytosis and regulation of signalling pathways. LRP-1 mediates the endocytosis of a diverse set of extracellular ligands that play important roles in tumor progression. Here, we investigated the involvement of LRP-1 in P-gp expression and subcellular redistribution from the cell surface to the lysosomal membrane by endocytosis and its potential implication in P-gp-mediated multidrug resistance in MCF-7 cells. Our results showed that MCF-7 resistant cells (MCF-7R) overexpressed the P-gp, LRP-1 and LAMP-1 and were 11.66-fold resistant to Dox. Our study also revealed that in MCF-7R cells, lysosomes were predominantly high density compared to sensitized cells and P-gp was localized in the plasma membrane and lysosomes. LRP-1 blockade reduced lysosomes density and level of LAMP-1 and P-gp. It also affected the subcellular distribution of P-gp. Under these conditions, we restored Dox nuclear uptake and ERK 1/2 activation thus leading to MCF-7R cell sensitization to Dox. Our data suggest that LRP-1 is able to modulate the P-gp expression and subcellular redistribution by endocytosis and to potentiate the P-gp-acquired Dox resistance.

C-Biomarker.net: A Cytoscape app for the identification of cancer biomarker genes from cores of large biomolecular networks

Although there have been many studies revealing that biomarker genes for early cancer detection can be found in biomolecular networks, no proper tool exists to discover the cancer biomarker genes from various biomolecular networks. Accordingly, we developed a novel Cytoscape app called C-Biomarker.net, which can identify cancer biomarker genes from cores of various biomolecular networks. Derived from recent research, we designed and implemented the software based on parallel algorithms proposed in this study for working on high-performance computing devices. We tested our software on various network sizes and found the suitable size for each running mode on CPU or GPU. Interestingly, using the software for 17 cancer signaling pathways, we found that on average 70.59% of the top three nodes residing at the innermost core of each pathway are biomarker genes of the cancer respectively to the pathway. Similarly, by the software, we also found 100% of the top ten nodes at both cores of Human Gene Regulatory (HGR) network and Human Protein-Protein Interaction (HPPI) network are multi-cancer biomarkers. These case studies are reliable evidence for performance of cancer biomarker prediction function in the software. Through the case studies, we also suggest that true cores of directed complex networks should be identified by the algorithm of R-core rather than K-core as usual. Finally, we compared the prediction result of our software with those of other researchers and confirmed that our prediction method outperforms the other methods. Taken together, C-Biomarker.net is a reliable tool that efficiently detects biomarker nodes from cores of various large biomolecular networks. The software is available at https://github.com/trantd/C-Biomarker.net.

Identifying magnetosome-associated genes in the extended CtrA regulon in Magnetospirillum magneticum AMB-1 using a combinational approach

Magnetotactic bacteria (MTB) are worth studying because of magnetosome biomineralization. Magnetosome biogenesis in MTB is controlled by multiple genes known as magnetosome-associated genes. Recent advances in bioinformatics provide a unique opportunity for studying functions of magnetosome-associated genes and networks that they are involved in. Furthermore, various types of bioinformatics analyses can also help identify genes associated with magnetosome biogenesis. To predict novel magnetosome-associated genes in the extended CtrA regulon, we analyzed expression data of Magnetospirillum magneticum AMB-1 in the GSE35625 dataset in NCBI GEO. We identified 10 potential magnetosome-associated genes using a combinational approach of differential expression analysis, Gene ontology and Kyoto encyclopedia of genes and genomes pathway enrichment analysis, protein-protein interaction network analysis and weighted gene co-expression network analysis. Meanwhile, we also discovered and compared two co-expression modules that most known magnetosome-associated genes belong to. Our comparison indicated the importance of energy on regulating co-expression module structures for magnetosome biogenesis. At the last stage of our research, we predicted at least four real magnetosome-associated genes out of 10 potential genes, based on a comparison of evolutionary trees between known and potential magnetosome-associated genes. Because of the discovery of common subtrees that the stressed species are enriched in, we proposed a hypothesis that multiple types of environmental stress can trigger magnetosome evolution in different waters, and therefore its evolution can recur at different times in various locations on earth. Overall, our research provides useful information for identifying new MTB species and understanding magnetosome biogenesis.

A novel focal adhesion-related risk model predicts prognosis of bladder cancer —— a bioinformatic study based on TCGA and GEO database

Background

Bladder cancer (BLCA) is the ninth most common cancer globally, as well as the fourth most common cancer in men, with an incidence of 7%. However, few effective prognostic biomarkers or models of BLCA are available at present.

Methods

The prognostic genes of BLCA were screened from one cohort of The Cancer Genome Atlas (TCGA) database through univariate Cox regression analysis and functionally annotated by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis. The intersecting genes of the BLCA gene set and focal adhesion-related gene were obtained and subjected to the least absolute shrinkage and selection operator regression (LASSO) to construct a prognostic model. Gene set enrichment analysis (GSEA) of high- and low-risk patients was performed to explore further the biological process related to focal adhesion genes. Univariate and multivariate Cox analysis, receiver operating characteristic (ROC) curve analysis, and Kaplan–Meier survival analysis (KM) were used to evaluate the prognostic model. DNA methylation analysis was presented to explore the relationship between prognosis and gene methylation. Furthermore, immune cell infiltration was assessed by CIBERSORT, ESTIMATE, and TIMER. The model was verified in an external GSE32894 cohort of the Gene Expression Omnibus (GEO) database, and the Prognoscan database presented further validation of genes. The HPA database validated the related protein level, and functional experiments verified significant risk factors in the model.

Results

VCL, COL6A1, RAC3, PDGFD, JUN, LAMA2, and ITGB6 were used to construct a prognostic model in the TCGA-BLCA cohort and validated in the GSE32894 cohort. The 7-gene model successfully stratified the patients into both cohorts’ high- and low-risk groups. The higher risk score was associated with a worse prognosis.

Conclusions

The 7-gene prognostic model can classify BLCA patients into high- and low-risk groups based on the risk score and predict the overall survival, which may aid clinical decision-making.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-10264-5.

The Future of Biomarkers in Veterinary Medicine: Emerging Approaches and Associated Challenges

Simple Summary

In this review we seek to outline the role of new technologies in biomarker discovery, particularly within the veterinary field and with an emphasis on ‘omics’, as well as to examine why many biomarkers-despite much excitement-have not yet made it to clinical practice. Further we emphasise the critical need for close collaboration between clinicians, researchers and funding bodies and the need to set clear goals for biomarker requirements and realistic application in the clinical setting, ensuring that biomarker type, method of detection and clinical utility are compatible, and adequate funding, time and sample size are available for all phases of development.

Abstract

New biomarkers promise to transform veterinary practice through rapid diagnosis of diseases, effective monitoring of animal health and improved welfare and production efficiency. However, the road from biomarker discovery to translation is not always straightforward. This review focuses on molecular biomarkers under development in the veterinary field, introduces the emerging technological approaches transforming this space and the role of ‘omics platforms in novel biomarker discovery. The vast majority of veterinary biomarkers are at preliminary stages of development and not yet ready to be deployed into clinical translation. Hence, we examine the major challenges encountered in the process of biomarker development from discovery, through validation and translation to clinical practice, including the hurdles specific to veterinary practice and to each of the ‘omics platforms–transcriptomics, proteomics, lipidomics and metabolomics. Finally, recommendations are made for the planning and execution of biomarker studies with a view to assisting the success of novel biomarkers in reaching their full potential.

Screening of Prognostic Markers for Hepatocellular Carcinoma Patients Based on Multichip Combined Analysis

Methods

GSE (14520, 36376, 57957, 76427) datasets were accessed from GEO database. 55 differential mRNAs (DEGs) were obtained by differential analysis based on the datasets. GO and KEGG analysis results indicated that the DEGs were enriched in xenobiotic metabolic process and other pathways. Expression profiles and clinical data of TCGA-LIHC mRNAs were from TCGA database. We established a prognostic model of HCC through univariate and multivariate Cox risk regression analyses. ROC curve analysis was used to examine the prognostic model performance. GSEA analysis was performed between the high- and low-risk score sample groups.

Results

A 4-gene HCC prognostic model was constructed, in which the gene expressions correlated to HCC patients' survival. The AUC value presented 0.734 in the ROC analysis for the prognostic model.

Conclusion

The four-gene model could be introduced as an independent prognostic factors to assess HCC patients' survival status.

Gene and Protein Interaction Network Analysis in Epithelial-Mesenchymal Transition of Hertwig's Epithelial Root Sheath reveals periodontal regenerative drug targets - An in silico study

Background and aim

Hertwig’s Εpithelial Root Sheath (HΕRS) has a major function in the developing tooth roots. Earlier research revealed that it undergoes epithelial–mesenchymal transition, a vital process for the morphogenesis and complete development of the tooth and its surrounding periodontium. Few studies have demonstrated the role of HERS in cementogenesis through ΕMΤ. The background of this in-silico system biology approach is to find a hub protein and gene involved in the EMT of HERS that may uncover novel insights in periodontal regenerative drug targets.

Materials and methods

The protein and gene list involved in epithelial–mesenchymal transition were obtained from literature sources. The protein interaction was constructed using STRING software and the protein interaction network was analyzed. Molecular docking simulation checks the binding energy and stability of protein-ligand complex.

Results

Results revealed the hub gene to be DYRK1A(Hepcidin), and the ligand was identified as isoetharine. SΤRIΝG results showed a confidence cutoff of 0.9 in sensitivity analysis with a condensed protein interaction network. Overall, 98 nodes from 163 nodes of expected edges were found with an average node degree of 11.9. Docking results show binding energy of −4.70, and simulation results show an RMSD value of 5.6 Å at 50 ns.

Conclusion

Isoetharine could be a potential drug for periodontal regeneration.

A Three-mRNA Signature Associated with Pyrimidine Metabolism for Prognosis of BRCA

Objective

Breast invasive carcinoma (BRCA), as a systemic disease, is currently the most malignant tumor among women. Early detection of BRCA will increase the probability of cure. Pyrimidine metabolism (PyM) stands for an essential metabolic pathway related to DNA replication of cancer cells, which may also serve as a diagnostic marker and therapeutic target. Therefore, the aim of this research is to discover a prognostic signature associated with PyM for BRCA.

Methods

The BRCA mRNA sequencing data along with microarray data were obtained based on The Cancer Genome Atlas (TCGA) database. In addition, 4 PyM-related gene sets were profiled through gene set enrichment analysis (GSEA); it revealed the core genes differentially expressed in cancer and paracancerous tissue. Thereafter, genes were subjected to univariate as well as multivariate regression for constructing an mRNA signature to independently predict BRCA prognosis. Then, the Kaplan-Meier (KM) curve was applied for validation. The prognostic power of the signature was verified against the METABRIC (Molecular Taxonomy of Breast Cancer International Consortium) database.

Results

We constructed a three-mRNA (RRM2B, NME3, and POLD2) gene signature related to PyM to predict overall survival (OS) for BRCA. The as-constructed gene signature was adopted to classify cases as high- or low-risk group, identifying patients with BRCA with poor prognosis. Additionally, the risk score obtained using our constructed 3-mRNA prognosis signature is independent from other clinical variables.

Conclusion

Our findings suggested that PyM-related mRNA signature might be a combined prognostic biomarker for BRCA and can provide important reference that are useful for individualized treatment for BRCA patients.

Construction and validation of a metabolic gene-associated prognostic model for cervical carcinoma and the role on tumor microenvironment and immunity

Metabolic reprogramming is a common feature of tumor cells and is associated with tumorigenesis and progression. In this study, a metabolic gene-associated prognostic model (MGPM) was constructed using multiple bioinformatics analysis methods in cervical carcinoma (CC) tissues from The Cancer Genome Atlas (TCGA) database, which comprised fifteen differentially expressed metabolic genes (DEMGs). Patients were divided into a high-risk group with shorter overall survival (OS) and a low-risk group with better survival. Receiver operating characteristic (ROC) curve analysis showed that the MGPM precisely predicted the 1-, 3- and 5-year survival of CC patients. As expected, MGPM exhibited a favorable prognostic significance in the training and testing datasets of TCGA. And the clinicopathological parameters including stage, tumor (T) and metastasis (M) classifications had significant differences in low- and high-risk groups, which further demonstrated the MGPM had a favorite prognostic prediction ability. Additionally, patients with low-ESTMATEScore had a shorter OS and when those combined with high-risk scores presented a worse prognosis. Through “CIBERSORT” package and Wilcoxon rank-sum test, patients in the high-risk group with a poor prognosis showed lower levels of infiltration of T cell CD8 (P < 0.001), T cells memory activated (P = 0.010) and mast cells resting (P < 0.001), and higher levels of mast cells activated (P < 0.001), and we also found these patients had a worse response for immunosuppressive therapy. These findings demonstrate that MGPM accurately predicts survival outcomes in CC patients, which will be helpful for further optimizing immunotherapies for cancer by reprogramming its cell metabolism.

Screening Hub Genes of Hepatocellular Carcinoma Based on Public Databases

Tumor recurrence and metastasis often occur in HCC patients after surgery, and the prognosis is not optimistic. Hence, searching effective biomarkers for prognosis of is of great importance. Firstly, HCC-related data was acquired from the TCGA and GEO databases. Based on GEO data, 256 differentially expressed genes (DEGs) were obtained firstly. Subsequently, to clarify function of DEGs, clusterProfiler package was used to conduct functional enrichment analyses on DEGs. Protein-protein interaction (PPI) network analysis screened 20 key genes. The key genes were filtered via GEPIA database, by which 11 hub genes (F9, CYP3A4, ASPM, AURKA, CDC20, CDCA5, NCAP, PRC1, PTTG1, TOP2A, and KIFC1) were screened out. Then, univariate Cox analysis was applied to construct a prognostic model, followed by a prediction performance validation. With the risk score calculated by the model and common clinical features, univariate and multivariate analyses were carried out to assess whether the prognostic model could be used independently for prognostic prediction. In conclusion, the current study screened HCC prognostic gene signature based on public databases.

Convolutional neural network for human cancer types prediction by integrating protein interaction networks and omics data

Many studies have proven the power of gene expression profile in cancer identification, however, the explosive growth of genomics data increasing needs of tools for cancer diagnosis and prognosis in high accuracy and short times. Here, we collected 6136 human samples from 11 cancer types, and integrated their gene expression profiles and protein-protein interaction (PPI) network to generate 2D images with spectral clustering method. To predict normal samples and 11 cancer tumor types, the images of these 6136 human cancer network were separated into training and validation dataset to develop convolutional neural network (CNN). Our model showed 97.4% and 95.4% accuracies in identification of normal versus tumors and 11 cancer types, respectively. We also provided the results that tumors located in neighboring tissues or in the same cell types, would induce machine make error classification due to the similar gene expression profiles. Furthermore, we observed some patients may exhibit better prognosis if their tumors often misjudged into normal samples. As far as we know, we are the first to generate thousands of cancer networks to predict and classify multiple cancer types with CNN architecture. We believe that our model not only can be applied to cancer diagnosis and prognosis, but also promote the discovery of multiple cancer biomarkers.

Expression Profile and Prognostic Value of Wnt Signaling Pathway Molecules in Colorectal Cancer

Colorectal cancer (CRC) is a heterogeneous disease with changes in the genetic and epigenetic levels of various genes. The molecular assessment of CRC is gaining increasing attention, and furthermore, there is an increase in biomarker use for disease prognostication. Therefore, the identification of different gene biomarkers through messenger RNA (mRNA) abundance levels may be useful for capturing the complex effects of CRC. In this study, we demonstrate that the high mRNA levels of 10 upregulated genes (DPEP1, KRT80, FABP6, NKD2, FOXQ1, CEMIP, ETV4, TESC, FUT1, and GAS2) are observed in CRC cell lines and public CRC datasets. Moreover, we find that a high mRNA expression of DPEP1, NKD2, CEMIP, ETV4, TESC, or FUT1 is significantly correlated with a worse prognosis in CRC patients. Further investigation reveals that CTNNB1 is the key factor in the interaction of the canonical Wnt signaling pathway with 10 upregulated CRC-associated genes. In particular, we identify NKD2, FOXQ1, and CEMIP as three CTNNB1-regulated genes. Moreover, individual inhibition of the expression of three CTNNB1-regulated genes can cause the growth inhibition of CRC cells. This study reveals efficient biomarkers for the prognosis of CRC and provides a new molecular interaction network for CRC.

DeepProg: an ensemble of deep-learning and machine-learning models for prognosis prediction using multi-omics data

Multi-omics data are good resources for prognosis and survival prediction; however, these are difficult to integrate computationally. We introduce DeepProg, a novel ensemble framework of deep-learning and machine-learning approaches that robustly predicts patient survival subtypes using multi-omics data. It identifies two optimal survival subtypes in most cancers and yields significantly better risk-stratification than other multi-omics integration methods. DeepProg is highly predictive, exemplified by two liver cancer (C-index 0.73-0.80) and five breast cancer datasets (C-index 0.68-0.73). Pan-cancer analysis associates common genomic signatures in poor survival subtypes with extracellular matrix modeling, immune deregulation, and mitosis processes. DeepProg is freely available at https://github.com/lanagarmire/DeepProg.

Identification of Five Glycolysis-Related Gene Signature and Risk Score Model for Colorectal Cancer

Metabolic changes, especially in glucose metabolism, are widely established during the occurrence and development of tumors and regarded as biological markers of pan-cancer. The well-known ‘Warburg effect’ demonstrates that cancer cells prefer aerobic glycolysis even if there is sufficient ambient oxygen. Accumulating evidence suggests that aerobic glycolysis plays a pivotal role in colorectal cancer (CRC) development. However, few studies have examined the relationship of glycolytic gene clusters with prognosis of CRC patients. Here, our aim is to build a glycolysis-associated gene signature as a biomarker for colorectal cancer. The mRNA sequencing and corresponding clinical data were downloaded from TCGA and GEO databases. Gene set enrichment analysis (GSEA) was performed, indicating that four gene clusters were significantly enriched, which revealed the inextricable relationship of CRC with glycolysis. By comparing gene expression of cancer and adjacent samples, 236 genes were identified. Univariate, multivariate, and LASSO Cox regression analyses screened out five prognostic-related genes (ENO3, GPC1, P4HA1, SPAG4, and STC2). Kaplan–Meier curves and receiver operating characteristic curves (ROC, AUC = 0.766) showed that the risk model could become an effective prognostic indicator (P < 0.001). Multivariate Cox analysis also revealed that this risk model is independent of age and TNM stages. We further validated this risk model in external cohorts (GES38832 and GSE39582), showing these five glycolytic genes could emerge as reliable predictors for CRC patients’ outcomes. Lastly, based on five genes and risk score, we construct a nomogram model assessed by C-index (0.7905) and calibration plot. In conclusion, we highlighted the clinical significance of glycolysis in CRC and constructed a glycolysis-related prognostic model, providing a promising target for glycolysis regulation in CRC.

Transcriptome analysis uncovers the diagnostic value of miR-192-5p/HNF1A-AS1/VIL1 panel in cervical adenocarcinoma

Despite the fact that the incidence of cervical squamous cell carcinoma has decreased, there is an increase in the incidence of cervical adenocarcinoma. However, our knowledge on cervical adenocarcinoma is largely unclear. Transcriptome sequencing was conducted to compare 4 cervical adenocarcinoma tissue samples with 4 normal cervical tissue samples. mRNA, lncRNA, and miRNA signatures were identified to discriminate cervical adenocarcinoma from normal cervix. The expression of VIL1, HNF1A-AS1, MIR194-2HG, SSTR5-AS1, miR-192-5p, and miR-194-5p in adenocarcinoma were statistically significantly higher than that in normal control samples. The Receiver Operating Characteristic (ROC) curve analysis indicated that combination of miR-192-5p, HNF1A-AS1, and VIL1 yielded a better performance (AUC = 0.911) than any single molecule -and could serve as potential biomarkers for cervical adenocarcinoma. Of note, the combination model also gave better performance than TCT test for cervical adenocarcinoma diagnosis. However, there was no correlation between miR-192-5p or HNF1A-AS1 and HPV16/18 E6 or E7. VIL1 was weakly correlated with HPV18 E7 expression. In summary, our study has identified miR-192-5p/HNF1A-AS1/VIL1 panel that accurately discriminates adenocarcinoma from normal cervix. Detection of this panel may provide considerable clinical value in the diagnosis of cervical adenocarcinoma.

Overall survival prediction of non-small cell lung cancer by integrating microarray and clinical data with deep learning

Non-small cell lung cancer (NSCLC) is one of the most common lung cancers worldwide. Accurate prognostic stratification of NSCLC can become an important clinical reference when designing therapeutic strategies for cancer patients. With this clinical application in mind, we developed a deep neural network (DNN) combining heterogeneous data sources of gene expression and clinical data to accurately predict the overall survival of NSCLC patients. Based on microarray data from a cohort set (614 patients), seven well-known NSCLC biomarkers were used to group patients into biomarker- and biomarker+ subgroups. Then, by using a systems biology approach, prognosis relevance values (PRV) were then calculated to select eight additional novel prognostic gene biomarkers. Finally, the combined 15 biomarkers along with clinical data were then used to develop an integrative DNN via bimodal learning to predict the 5-year survival status of NSCLC patients with tremendously high accuracy (AUC: 0.8163, accuracy: 75.44%). Using the capability of deep learning, we believe that our prediction can be a promising index that helps oncologists and physicians develop personalized therapy and build the foundation of precision medicine in the future.

Network Properties of Cancer Prognostic Gene Signatures in the Human Protein Interactome

Prognostic gene signatures are critical in cancer prognosis assessments and their pinpoint treatments. However, their network properties remain unclear. Here, we obtained nine prognostic gene sets including 1439 prognostic genes of different cancers from related publications. Four network centralities were used to examine the network properties of prognostic genes (PG) compared with other gene sets based on the Human Protein Reference Database (HPRD) and String networks. We also proposed three novel network measures for further investigating the network properties of prognostic gene sets (PGS) besides clustering coefficient. The results showed that PG did not occupy key positions in the human protein interaction network and were more similar to essential genes rather than cancer genes. However, PGS had significantly smaller intra-set distance (IAD) and inter-set distance (IED) in comparison with random sets (p-value < 0.001). Moreover, we also found that PGS tended to be distributed within network modules rather than between modules (p-value < 0.01), and the functional intersection of the modules enriched with PGS was closely related to cancer development and progression. Our research reveals the common network properties of cancer prognostic gene signatures in the human protein interactome. We argue that these are biologically meaningful and useful for understanding their molecular mechanism.

Glycolysis gene expression profilings screen for prognostic risk signature of hepatocellular carcinoma

Metabolic changes are the markers of cancer and have attracted wide attention in recent years. One of the main metabolic features of tumor cells is the high level of glycolysis, even if there is oxygen. The transformation and preference of metabolic pathways is usually regulated by specific gene expression. The aim of this study is to develop a glycolysis-related risk signature as a biomarker via four common cancer types. Only hepatocellular carcinoma was shown the strong relationship with glycolysis. The mRNA sequencing and chip data of hepatocellular carcinoma, breast invasive carcinoma, renal clear cell carcinoma, colorectal adenocarcinoma were included in the study. Gene set enrichment analysis was performed, profiling three glycolysis-related gene sets, it revealed genes associated with the biological process. Univariate and multivariate Cox proportional regression models were used to screen out prognostic-related gene signature. We identified six mRNAs (DPYSL4, HOMER1, ABCB6, CENPA, CDK1, STMN1) significantly associated with overall survival in the Cox proportional regression model for hepatocellular carcinoma. Based on this gene signature, we were able to divide patients into high-risk and low-risk subgroups. Multivariate Cox regression analysis showed that prognostic power of this six gene signature is independent of clinical variables. Further, we validated this data in our own 55 paired hepatocellular carcinoma and adjacent tissues. The results showed that these proteins were highly expressed in hepatocellular carcinoma tissues compared with adjacent tissue. The survival time of high-risk group was significantly shorter than that of low-risk group, indicating that high-risk group had poor prognosis. We calculated the correlation coefficients between six proteins and found that these six proteins were independent of each other. In conclusions, we developed a glycolysis-related gene signature that could predict survival in hepatocellular carcinoma patients. Our findings provide novel insight to the mechanisms of glycolysis and it is useful for identifying patients with hepatocellular carcinoma with poor prognoses.

Expression of EEF1A1 Is Associated with Prognosis of Patients with Colon Adenocarcinoma

Background: The prognostic role of the translational factor, elongation factor-1 alpha 1 (EEF1A1), in colon cancer is unclear. Objectives: The present study aimed to investigate the expression of EEF1A in tissues obtained from patients with stage II and III colon cancer and analyze its association with patient prognosis. Methods: A total of 281 patients with colon cancer who underwent curative resection were analyzed according to EEF1A1 expression. Results: The five-year overall survival in the high-EEF1A1 group was 87.7%, whereas it was 65.6% in the low-EEF1A1 expression group (hazard ratio (HR) 2.47, 95% confidence interval (CI) 1.38–4.44, p = 0.002). The five-year disease-free survival of patients with high EEF1A1 expression was 82.5%, which was longer than the rate of 55.4% observed for patients with low EEF1A1 expression (HR 2.94, 95% CI 1.72–5.04, p < 0.001). Univariate Cox regression analysis indicated that age, preoperative carcinoembryonic antigen level, adjuvant treatment, total number of metastatic lymph nodes, and EEF1A1 expression level were significant prognostic factors for death. In multivariate analysis, expression of EEF1A1 was an independent prognostic factor associated with death (HR 3.01, 95% CI 1.636–5.543, p < 0.001). EEF1A1 expression was also an independent prognostic factor for disease-free survival in multivariate analysis (HR 2.54, 95% CI 1.459–4.434, p < 0.001). Conclusions: Our study demonstrated that high expression of EEF1A1 has a favorable prognostic effect on patients with colon adenocarcinoma.

Sex-biased differences in the correlation between epithelial-to-mesenchymal transition-associated genes in cancer cell lines

There is a wide disparity in the incidence, malignancy and mortality of different types of cancer between each sex. The sex-specificity of cancer seems to be dependent on the type of cancer. Cancer incidence and mortality have been demonstrated as sex-specific in a number of different types of cancer, such as liver cancer, whereas sex-specificity is not noticeable in certain other types of cancer, including colon and lung cancer. The present study aimed to elucidate the molecular basis for sex-biased gene expression in cancer. The mRNA expression of the epithelial-to-mesenchymal transition-associated genes was investigated, including E-cadherin (also termed CDH1), vimentin (VIM), discoidin domain receptor 1 (DDR1) and zinc finger E-box binding homeobox 1 (ZEB1) in female- and male-derived cancer cell lines by reverse transcription (RT)-PCR and the Broad-Novartis Cancer Cell Line Encyclopedia (CCLE) database analysis. A negative correlation was observed between DDR1 and ZEB1 only in the female-derived cancer cell lines via RT-PCR analysis. A negative correlation between DDR1 index (defined by the logarithmic value of DDR1 divided by ZEB1, based on the mRNA data from the RT-PCR analysis) and an invasive phenotype was observed in cancer cell lines in a sex-specific manner. Analysis of the CCLE database demonstrated that DDR1 and ZEB1, which are already known to be sex-biased, were negatively correlated in female-derived liver cancer cell lines, but not in male-derived liver cancer cell lines. In contrast, cell lines of colon and lung cancer did not reveal any sex-dependent difference in the correlation between DDR1 and ZEB1. Kaplan-Meier survival curves using the transcriptomic datasets such as Gene Expression Omnibus, European Genome-phenome Archiva and The Cancer Genome Atlas databases suggested a sex-biased difference in the correlation between DDR1 expression pattern and overall survival in patients with liver cancer. The results of the present study indicate that sex factors may affect the regulation of gene expression, contributing to the sex-biased progression of the different types of cancer, particularly liver cancer. Overall, these findings suggest that analyses of the correlation between DDR1 and ZEB1 may prove useful when investigating sex-biased cancers.

Integrative Analysis of Cancer Omics Data for Prognosis Modeling

Prognosis modeling plays an important role in cancer studies. With the development of omics profiling, extensive research has been conducted to search for prognostic markers for various cancer types. However, many of the existing studies share a common limitation by only focusing on a single cancer type and suffering from a lack of sufficient information. With potential molecular similarity across cancer types, one cancer type may contain information useful for the analysis of other types. The integration of multiple cancer types may facilitate information borrowing so as to more comprehensively and more accurately describe prognosis. In this study, we conduct marginal and joint integrative analysis of multiple cancer types, effectively introducing integration in the discovery process. For accommodating high dimensionality and identifying relevant markers, we adopt the advanced penalization technique which has a solid statistical ground. Gene expression data on nine cancer types from The Cancer Genome Atlas (TCGA) are analyzed, leading to biologically sensible findings that are different from the alternatives. Overall, this study provides a novel venue for cancer prognosis modeling by integrating multiple cancer types.

Wx: a neural network-based feature selection algorithm for transcriptomic data

Next-generation sequencing (NGS), which allows the simultaneous sequencing of billions of DNA fragments simultaneously, has revolutionized how we study genomics and molecular biology by generating genome-wide molecular maps of molecules of interest. However, the amount of information produced by NGS has made it difficult for researchers to choose the optimal set of genes. We have sought to resolve this issue by developing a neural network-based feature (gene) selection algorithm called Wx. The Wx algorithm ranks genes based on the discriminative index (DI) score that represents the classification power for distinguishing given groups. With a gene list ranked by DI score, researchers can institutively select the optimal set of genes from the highest-ranking ones. We applied the Wx algorithm to a TCGA pan-cancer gene-expression cohort to identify an optimal set of gene-expression biomarker candidates that can distinguish cancer samples from normal samples for 12 different types of cancer. The 14 gene-expression biomarker candidates identified by Wx were comparable to or outperformed previously reported universal gene expression biomarkers, highlighting the usefulness of the Wx algorithm for next-generation sequencing data. Thus, we anticipate that the Wx algorithm can complement current state-of-the-art analytical applications for the identification of biomarker candidates as an alternative method. The stand-alone and web versions of the Wx algorithm are available at https://github.com/deargen/DearWXpub and https://wx.deargendev.me/, respectively.

Salivary Extracellular Vesicle-Associated exRNA as Cancer Biomarker

Extracellular vesicles (EVs) secreted in biological fluids contain several transcripts of the cell of origin, which may modify the functions and phenotype of proximal and distant cells. Cancer-derived EVs may promote a favorable microenvironment for cancer growth and invasion by acting on stroma and endothelial cells and may favor metastasis formation. The transcripts contained in cancer EVs may be exploited as biomarkers. Protein and extracellular RNA (exRNA) profiling in patient bio-fluids, such as blood and urine, was performed to identify molecular features with potential diagnostic and prognostic values. EVs are concentrated in saliva, and salivary EVs are particularly enriched in exRNAs. Several studies were focused on salivary EVs for the detection of biomarkers either of non-oral or oral cancers. The present paper provides an overview of the available studies on the diagnostic potential of exRNA profiling in salivary EVs.

Systems Biology Approaches to Investigate Genetic and Epigenetic Molecular Progression Mechanisms for Identifying Gene Expression Signatures in Papillary Thyroid Cancer

Thyroid cancer is the most common endocrine cancer. Particularly, papillary thyroid cancer (PTC) accounts for the highest proportion of thyroid cancer. Up to now, there are few researches discussing the pathogenesis and progression mechanisms of PTC from the viewpoint of systems biology approaches. In this study, first we constructed the candidate genetic and epigenetic network (GEN) consisting of candidate protein–protein interaction network (PPIN) and candidate gene regulatory network (GRN) by big database mining. Secondly, system identification and system order detection methods were applied to prune candidate GEN via next-generation sequencing (NGS) and DNA methylation profiles to obtain the real GEN. After that, we extracted core GENs from real GENs by the principal network projection (PNP) method. To investigate the pathogenic and progression mechanisms in each stage of PTC, core GEN was denoted in respect of KEGG pathways. Finally, by comparing two successive core signaling pathways of PTC, we not only shed light on the causes of PTC progression, but also identified essential biomarkers with specific gene expression signature. Moreover, based on the identified gene expression signature, we suggested potential candidate drugs to prevent the progression of PTC with querying Connectivity Map (CMap).

An Ensemble Strategy to Predict Prognosis in Ovarian Cancer Based on Gene Modules

Due to the high heterogeneity and complexity of cancer, it is still a challenge to predict the prognosis of cancer patients. In this work, we used a clustering algorithm to divide patients into different subtypes in order to reduce the heterogeneity of the cancer patients in each subtype. Based on the hypothesis that the gene co-expression network may reveal relationships among genes, some communities in the network could influence the prognosis of cancer patients and all the prognosis-related communities could fully reveal the prognosis of cancer patients. To predict the prognosis for cancer patients in each subtype, we adopted an ensemble classifier based on the gene co-expression network of the corresponding subtype. Using the gene expression data of ovarian cancer patients in TCGA (The Cancer Genome Atlas), three subtypes were identified. Survival analysis showed that patients in different subtypes had different survival risks. Three ensemble classifiers were constructed for each subtype. Leave-one-out and independent validation showed that our method outperformed control and literature methods. Furthermore, the function annotation of the communities in each subtype showed that some communities were cancer-related. Finally, we found that the current drug targets can partially support our method.

Identifying cancer prognostic modules by module network analysis

Background

The identification of prognostic genes that can distinguish the prognostic risks of cancer patients remains a significant challenge. Previous works have proven that functional gene sets were more reliable for this task than the gene signature. However, few works have considered the cross-talk among functional gene sets, which may result in neglecting important prognostic gene sets for cancer.

Results

Here, we proposed a new method that considers both the interactions among modules and the prognostic correlation of the modules to identify prognostic modules in cancers. First, dense sub-networks in the gene co-expression network of cancer patients were detected. Second, cross-talk between every two modules was identified by a permutation test, thus generating the module network. Third, the prognostic correlation of each module was evaluated by the resampling method. Then, the GeneRank algorithm, which takes the module network and the prognostic correlations of all the modules as input, was applied to prioritize the prognostic modules. Finally, the selected modules were validated by survival analysis in various data sets. Our method was applied in three kinds of cancers, and the results show that our method succeeded in identifying prognostic modules in all the three cancers. In addition, our method outperformed state-of-the-art methods. Furthermore, the selected modules were significantly enriched with known cancer-related genes and drug targets of cancer, which may indicate that the genes involved in the modules may be drug targets for therapy.

Conclusions

We proposed a useful method to identify key modules in cancer prognosis and our prognostic genes may be good candidates for drug targets.

Electronic supplementary material

The online version of this article (10.1186/s12859-019-2674-z) contains supplementary material, which is available to authorized users.

EMT network-based feature selection improves prognosis prediction in lung adenocarcinoma

Various feature selection algorithms have been proposed to identify cancer prognostic biomarkers. In recent years, however, their reproducibility is criticized. The performance of feature selection algorithms is shown to be affected by the datasets, underlying networks and evaluation metrics. One of the causes is the curse of dimensionality, which makes it hard to select the features that generalize well on independent data. Even the integration of biological networks does not mitigate this issue because the networks are large and many of their components are not relevant for the phenotype of interest. With the availability of multi-omics data, integrative approaches are being developed to build more robust predictive models. In this scenario, the higher data dimensions create greater challenges. We proposed a phenotype relevant network-based feature selection (PRNFS) framework and demonstrated its advantages in lung cancer prognosis prediction. We constructed cancer prognosis relevant networks based on epithelial mesenchymal transition (EMT) and integrated them with different types of omics data for feature selection. With less than 2.5% of the total dimensionality, we obtained EMT prognostic signatures that achieved remarkable prediction performance (average AUC values >0.8), very significant sample stratifications, and meaningful biological interpretations. In addition to finding EMT signatures from different omics data levels, we combined these single-omics signatures into multi-omics signatures, which improved sample stratifications significantly. Both single- and multi-omics EMT signatures were tested on independent multi-omics lung cancer datasets and significant sample stratifications were obtained.

Prognostic Gene Discovery in Glioblastoma Patients using Deep Learning

This study aims to discover genes with prognostic potential for glioblastoma (GBM) patients’ survival in a patient group that has gone through standard of care treatments including surgeries and chemotherapies, using tumor gene expression at initial diagnosis before treatment. The Cancer Genome Atlas (TCGA) GBM gene expression data are used as inputs to build a deep multilayer perceptron network to predict patient survival risk using partial likelihood as loss function. Genes that are important to the model are identified by the input permutation method. Univariate and multivariate Cox survival models are used to assess the predictive value of deep learned features in addition to clinical, mutation, and methylation factors. The prediction performance of the deep learning method was compared to other machine learning methods including the ridge, adaptive Lasso, and elastic net Cox regression models. Twenty-seven deep-learned features are extracted through deep learning to predict overall survival. The top 10 ranked genes with the highest impact on these features are related to glioblastoma stem cells, stem cell niche environment, and treatment resistance mechanisms, including POSTN, TNR, BCAN, GAD1, TMSB15B, SCG3, PLA2G2A, NNMT, CHI3L1 and ELAVL4.

Lipids as Biomarkers of Cancer and Bacterial Infections

Lipids are ubiquitous molecules, known to play important roles in various cellular processes. Alterations to the lipidome can therefore be used as a read-out of the signs of disease, highlighting the importance to consider lipids as biomarkers in addition of nucleic acid and proteins. Lipids are among the primary structural and functional constituents of biological tissues, especially cell membranes. Along with membrane formation, lipids play also a crucial role in cell signalling, inflammation and energy storage. It was shown recently that lipid metabolism disorders play an important role in carcinogenesis and development. As well, the role of lipids in disease is particularly relevant for bacterial infections, during which several lipid bacterial virulence factors are recognized by the human innate immune response, such as lipopolysaccharide in Gram-negative bacteria, lipoteichoic acid in Gram-positive bacteria, and lipoglycans in mycobacteria. Compared to nucleic acids and proteins, a complete analysis of the lipidome, which is the comprehensive characterization of different lipid families, is usually very challenging due to the heterogeneity of lipid classes and their intrinsic physicoproperties caused by variations in the constituents of each class. Understanding the chemical diversity of lipids is therefore crucial to understanding their biological relevance and, as a consequence, their use as potential biomarkers for non-infectious and infectious diseases. This mini-review exposes the current knowledge and limitations of the use of lipids as biomarkers of the top global killers which are cancer and bacterial infections.

Cross-cancer Prediction: A Novel Machine Learning Approach to Discover Molecular Targets for Development of Treatments for Multiple Cancers

Conventional cancer drug development has long been limited to organ- or tissue-specific cancer types. However, it has become increasingly known that specific genetic abnormalities are responsible for the carcinogenesis of multiple cancers. The recent US Food and Drug Administration (FDA) approval of the first multi-cancer drug, Keytruda, has demonstrated the feasibility of developing new drugs that target multiple cancers. Despite a promising future, methodological development for identifying multi-cancer molecular targets remains encumbered. This study developed a novel machine learning approach to identify such genes responsible for multiple cancers by synthesizing salient genomic information from cancer-specific classification models. This approach centered on the cross-cancer prediction method for identifying groups of cancers with high cross-cancer predictability. Furthermore, a robust hybrid classifier, comprising Prediction Analysis for Microarrays and Random Forest, was developed to integrate predictive models for gene inference. This approach has successfully identified key genes shared by endometrial cancer, mammary gland ductal carcinoma, and small cell lung cancer. The results are supported by published experimental evidence. This framework holds potential to transform the current methods of discovering multi-cancer molecular targets for clinical oncology.

A data-driven, knowledge-based approach to biomarker discovery: application to circulating microRNA markers of colorectal cancer prognosis

Recent advances in high-throughput technologies have provided an unprecedented opportunity to identify molecular markers of disease processes. This plethora of complex-omics data has simultaneously complicated the problem of extracting meaningful molecular signatures and opened up new opportunities for more sophisticated integrative and holistic approaches. In this era, effective integration of data-driven and knowledge-based approaches for biomarker identification has been recognised as key to improving the identification of high-performance biomarkers, and necessary for translational applications. Here, we have evaluated the role of circulating microRNA as a means of predicting the prognosis of patients with colorectal cancer, which is the second leading cause of cancer-related death worldwide. We have developed a multi-objective optimisation method that effectively integrates a data-driven approach with the knowledge obtained from the microRNA-mediated regulatory network to identify robust plasma microRNA signatures which are reliable in terms of predictive power as well as functional relevance. The proposed multi-objective framework has the capacity to adjust for conflicting biomarker objectives and to incorporate heterogeneous information facilitating systems approaches to biomarker discovery. We have found a prognostic signature of colorectal cancer comprising 11 circulating microRNAs. The identified signature predicts the patients' survival outcome and targets pathways underlying colorectal cancer progression. The altered expression of the identified microRNAs was confirmed in an independent public data set of plasma samples of patients in early stage vs advanced colorectal cancer. Furthermore, the generality of the proposed method was demonstrated across three publicly available miRNA data sets associated with biomarker studies in other diseases.

The role of glycosyltransferase enzyme GCNT3 in colon and ovarian cancer prognosis and chemoresistance

Glycosyltransferase enzyme GCNT3, has been proposed as a biomarker for prognosis in colorectal cancer (CRC). Our study goes in depth into the molecular basis of GCNT3 role in tumorigenesis and drug resistance, and it explores its potential role as biomarker in epithelial ovarian cancer (EOC). High levels of GCNT3 are associated with increased sensibility to 5-fluoracil in metastatic cells. Accordingly, GCNT3 re-expression leads to the gain of anti-carcinogenic cellular properties by reducing cell growth, invasion and by changing metabolic capacities. Integrated transcriptomic and proteomic analyses reveal that GCNT3 is linked to cellular cycle, mitosis and proliferation, response to drugs and metabolism pathways. The vascular epithelial growth factor A (VEGFA) arises as an attractive partner of GCNT3 functions in cell invasion and resistance. Finally, GCNT3 expression was analyzed in a cohort of 56 EOC patients followed by a meta-analysis of more than one thousand patients. This study reveals that GCNT3 might constitute a prognostic factor also in EOC, since its overexpression is associated with better clinical outcome and response to initial therapy. GCNT3 emerges as an essential glycosylation-related molecule in CRC and EOC progression, with potential interest as a predictive biomarker of response to chemotherapy.

LYL1 gene amplification predicts poor survival of patients with uterine corpus endometrial carcinoma: analysis of the Cancer genome atlas data

BACKGROUND

Somatic amplifications of the LYL1 gene are relatively common occurrences in patients who develop uterine corpus endometrial carcinoma (UCEC) as opposed to other cancers. This study was undertaken to determine whether such genetic alterations affect survival outcomes of UCEC.

METHODS

In 370 patients with UCEC, we analysed clinicopathologic characteristics and corresponding genomic data from The Cancer Genome Atlas database. Patients were stratified according to LYL1 gene status, grouped as amplification or non-amplification. Heightened levels of cancer-related genes expressed in concert with LYL1 amplification were similarly investigated through differentially expressed gene and gene set enrichment analyses. Factors associated with survival outcomes were also identified.

RESULTS

Somatic LYL1 gene amplification was observed in 22 patients (5.9%) with UCEC. Patients displaying amplification (vs. non-amplification) were significantly older at the time of diagnosis and more often were marked by non-endometrioid, high-grade, or advanced disease. In survival analysis, the amplification subset showed poorer progression-free survival (PFS) and overall survival (OS) rates (3-year PFS: 34.4% vs. 79.9%, P = 0.031; 5-year OS: 25.1% vs. 84.9%, P = 0.014). However, multivariate analyses adjusted for tumor histologic type, grade, and stage did not confirm LYL1 gene amplification as an independent prognostic factor for either PFS or OS. Nevertheless, MAPK, WNT, and cell cycle pathways were significantly enriched by LYL1 gene amplification (P < 0.001, P = 0.002, and P = 0.004, respectively).

CONCLUSIONS

Despite not being identified as an independent prognostic factor in UCEC, LYL1 gene amplification is associated with other poor prognostic factors and correlated with upregulation of cancer-related pathways.

Genes and functions from breast cancer signatures

BACKGROUND

Breast cancer is a heterogeneous disease and personalized medicine is the hope for the improvement of the clinical outcome. Multi-gene signatures for breast cancer stratification have been extensively studied in the past decades and more than 30 different signatures have been reported. A major concern is the minimal overlap of genes among the reported signatures. We investigated the breast cancer signature genes to address our hypothesis that the genes of different signature may share common functions, as well as to use these previously reported signature genes to build better prognostic models.

METHODS

A total of 33 signatures and the corresponding gene lists were investigated. We first examined the gene frequency and the gene overlap in these signatures. Then the gene functions of each signature gene list were analysed and compared by the KEGG pathways and gene ontology (GO) terms. A classifier built using the common genes was tested using the METABRIC (Molecular Taxonomy of Breast Cancer International Consortium) data. The common genes were also tested for building the Yin Yang gene mean expression ratio (YMR) signature using public datasets (GSE1456 and GSE2034).

RESULTS

Among a total of 2239 genes collected from the 33 breast cancer signatures, only 238 genes overlapped in at least two signatures; while from a total of 1979 function terms enriched in the 33 signature gene lists, 429 terms were common in at least two signatures. Most of the common function terms were involved in cell cycle processes. While there is almost no common overlapping genes between signatures developed for ER-positive (e.g. 21-gene signature) and those developed for ER-negative (e.g. basal signatures) tumours, they have common function terms such as cell death, regulation of cell proliferation. We used the 62 genes that were common in at least three signatures as a classifier and subtyped 1141 METABRIC cases including 144 normal samples into nine subgroups. These subgroups showed different clinical outcome. Among the 238 common genes, we selected those genes that are more highly expressed in normal breast tissue than in tumours as Yang genes and those more highly expressed in tumours than in normal as Yin genes and built a YMR model signature. This YMR showed significance in risk stratification in two datasets (GSE1456 and GSE2034).

CONCLUSIONS

The lack of significant numbers of overlapping genes among most breast cancer signatures can be partially explained by our discovery that these signature genes represent groups with similar functions. The genes collected from these previously reported signatures are valuable resources for new model development. The subtype classifier and YMR signature built from the common genes showed promising results.

Identifying statistically significant combinatorial markers for survival analysis

Background

Survival analysis methods have been widely applied in different areas of health and medicine, spanning over varying events of interest and target diseases. They can be utilized to provide relationships between the survival time of individuals and factors of interest, rendering them useful in searching for biomarkers in diseases such as cancer. However, some disease progression can be very unpredictable because the conventional approaches have failed to consider multiple-marker interactions. An exponential increase in the number of candidate markers requires large correction factor in the multiple-testing correction and hide the significance.

Methods

We address the issue of testing marker combinations that affect survival by adapting the recently developed Limitless Arity Multiple-testing Procedure (LAMP), a p-value correction technique for statistical tests for combination of markers. LAMP cannot handle survival data statistics, and hence we extended LAMP for the log-rank test, making it more appropriate for clinical data, with newly introduced theoretical lower bound of the p-value.

Results

We applied the proposed method to gene combination detection for cancer and obtained gene interactions with statistically significant log-rank p-values. Gene combinations with orders of up to 32 genes were detected by our algorithm, and effects of some genes in these combinations are also supported by existing literature.

Conclusion

The novel approach for detecting prognostic markers presented here can identify statistically significant markers with no limitations on the order of interaction. Furthermore, it can be applied to different types of genomic data, provided that binarization is possible.

Biomarker Identification for Cancer Disease Using Biclustering Approach: An Empirical Study

This paper presents an exhaustive empirical study to identify biomarkers using two approaches: frequency-based and network-based, over seventeen different biclustering algorithms and six different cancer expression datasets. To systematically analyze the biclustering algorithms, we perform enrichment analysis, subtype identification and biomarker identification. Biclustering algorithms such as C&C, SAMBA and Plaid are useful to detect biomarkers by both approaches for all datasets except prostate cancer. We detect a total of 102 gene biomarkers using frequency-based method out of which 19 are for blood cancer, 36 for lung cancer, 25 for colon cancer, 13 for multi-tissue cancer and 9 for prostate cancer. Using the network-based approach we detect a total of 41 gene biomarkers of which 15 are from blood cancer, 12 from lung cancer, 6 from colon cancer, 7 from multi-tissue cancer and 1 from prostate cancer dataset. We further extend our network analysis over some biclusters and detect some gene biomarkers not detected earlier by both frequency-based or network-based approach. We expand our work on breast cancer miRNA expression data to evaluate the performance of the biclustering algorithms. We detect 19 breast cancer biomarkers by frequency-based method and 5 by network-based method for the miRNA dataset.

Methylation-to-Expression Feature Models of Breast Cancer Accurately Predict Overall Survival, Distant-Recurrence Free Survival, and Pathologic Complete Response in Multiple Cohorts

Prognostic biomarkers serve a variety of purposes in cancer treatment and research, such as prediction of cancer progression, and treatment eligibility. Despite growing interest in multi-omic data integration for defining prognostic biomarkers, validated methods have been slow to emerge. Given that breast cancer has been the focus of intense research, it is amenable to studying the benefits of multi-omic prognostic models due to the availability of datasets. Thus, we examined the efficacy of our methylation-to-expression feature model (M2EFM) approach to combining molecular and clinical predictors to create risk scores for overall survival, distant metastasis, and chemosensitivity in breast cancer. Gene expression, DNA methylation, and clinical variables were integrated via M2EFM to build models of overall survival using 1028 breast tumor samples and applied to validation cohorts of 61 and 327 samples. Models of distant recurrence-free survival and pathologic complete response were built using 306 samples and validated on 182 samples. Despite different populations and assays, M2EFM models validated with good accuracy (C-index or AUC ≥ 0.7) for all outcomes and had the most consistent performance compared to other methods. Finally, we demonstrated that M2EFM identifies functionally relevant genes, which could be useful in translating an M2EFM biomarker to the clinic.

LRP1 expression in colon cancer predicts clinical outcome

LRP1 (low-density lipoprotein receptor-related protein 1), a multifunctional endocytic receptor, has recently been identified as a hub within a biomarker network for multi-cancer clinical outcome prediction. As its role in colon cancer has not yet been characterized, we here investigate the relationship between LRP1 and outcome.

MATERIALS AND METHODS

LRP1 mRNA expression was determined in colon adenocarcinoma and paired colon mucosa samples, as well as in stromal and tumor cells obtained after laser capture microdissection. Clinical potential was further investigated by immunohistochemistry in a population-based colon cancer series (n = 307). LRP1 methylation, mutation and miR-205 expression were evaluated and compared with LRP1 expression levels.

RESULTS

LRP1 mRNA levels were significantly lower in colon adenocarcinoma cells compared with colon mucosa and stromal cells obtained after laser capture microdissection. Low LRP1 immunohistochemical expression in adenocarcinomas was associated with higher age, right-sided tumor, loss of CDX2 expression, Annexin A10 expression, CIMP-H, MSI-H and BRAFV600E mutation. Low LRP1 expression correlated with poor clinical outcome, especially in stage IV patients. While LRP1 expression was downregulated by LRP1 mutation, LRP1 promoter was never methylated.

CONCLUSIONS

Loss of LRP1 expression is associated with worse colon cancer outcomes. Mechanistically, LRP1 mutation modulates LRP1 expression.