DriverDBv3: a multi-omics database for cancer driver gene research

Gene Expression Profiling of Tricarboxylic Acid Cycle and One Carbon Metabolism Related Genes for Prognostic Risk Signature of Colon Carcinoma

Colorectal cancer (CRC) is one of the most prevalent malignant tumors worldwide. Colon adenocarcinoma (COAD) is the most common pathological type of CRC and several biomarkers related to survival have been confirmed. Yet, the predictive effect of a single gene biomarker is not enough. The tricarboxylic acid (TCA) cycle and carbon metabolism play an important role in tumors. Thus, we aimed to identify new gene signatures from the TCA cycle and carbon metabolism to better predict the survival of COAD. This study performed mRNA expression profiling in large COAD cohorts (n = 417) from The Cancer Genome Atlas (TCGA) database. Univariate Cox regression and multivariate Cox regression analysis were performed, and receiver operating characteristic (ROC) curve was used to screen the variable combinations model which is most relevant to patient prognosis survival mostly. Univariable or multivariate analysis results showed that SUCLG2, SUCLG1, ACLY, SUCLG2P2, ATIC and ACO2 have associations with survival in COAD. Combined with clinical variables, we confirmed model 1 (AUC = 0.82505), most relevant to patient prognosis survival. Model 1 contains three genes: SUCLG2P2, SUCLG2 and ATIC, in which SUCLG2P2 and SUCLG2 were low-expressed in COAD, however, ATIC was highly expressed, and the expressions above are related to stages of CRC. Pearson analysis showed that SUCLG2P2, SUCLG2 and ATIC were correlated in normal COAD tissues, while only SUCLG2P2 and SUCLG2 were correlated in tumor tissues. Finally, we verified the expressions of these three genes in COAD samples. Our study revealed a possible connection between the TCA cycle and carbon metabolism and prognosis and showed a TCA cycle and carbon metabolism related gene signature which could better predict survival in COAD patients.

Effect of MAP3K8 on Prognosis and Tumor-Related Inflammation in Renal Clear Cell Carcinoma

Background: MAPK kinase kinase 8 (MAP3K8) is involved in the regulation of MAPK cascades and immune responses. Differential expression of MAP3K8 is closely correlated with tumorigenesis. In this study, we used bioinformatics tools to explore expression level, prognostic values, and interactive networks of MAP3K8 in renal clear cell carcinoma (ccRCC).

Methods: Differential expression of MAP3K8 was determined by TIMER2.0, UALCAN, and Oncomine Platform. For exploration of MAP3K8 mutation profile, TIMER2.0, DriverDBv3, and cBioPortal were used. The survival module of GEPIA, UALCAN, and DriverDBv3 was used to examine the prognostic value of MAP3K8. Immune infiltration was estimated by TIMER, TIDE, CIBERSORT, CIBERSORT-ABS, QUANTISEQ, XCELL, MCPCOUNTER, and EPIC algorithms. PPI networks and functional enrichment analysis were constructed using GeneMANIA, Cytoscape, and Metascape. The co-expression module in cBioPortal was used to find genes that are correlated with MAP3K8 in mRNA expression.

Results: Compared to normal renal samples, ccRCC (3.08-fold change, P = 1.50E-7; 1.10-fold change, P = 3.00E-3), papillary RCC (2.24-fold change, P = 1.86E-4), and hereditary ccRCC (1.98-fold change, P = 1.69E-9) have significantly higher levels of MAP3K8 expression. Compared to Grade 1 ccRCC samples, Grade 2 (P = 1.28E-3) and Grade 3 (P = 7.41E-4) cases have higher levels of MAP3K8 methylation. Percentage of patients harboring MAP3K8 mutation is 0.3% from TIMER2.0 and 0.2 to 11.5% from cBioPortal. High levels of MAP3K8 expression were associated with poorer overall survival (OS) in ccRCC (GEPIA: Log-rank P = 0.60E-2, HR = 1.5; DriverDBv3: Log-rank P = 1.68E-7, HR = 2.21; UALCAN: P = 0.20E-2). MAP3K8 was positively correlated with the presence of T cell regulatory (Tregs) (QUANTISEQ: Rho = 0.33, P = 1.59E-13). PPI network and functional enrichment analyses revealed that MAP3K8 correlated with NFKBIZ, MIAT, PARP15, CHFR, MKNK1, and ERMN, which was mainly involved in I-kappaB kinase/NF-kappaB and toll-like receptor signaling pathways.

Conclusion: MAP3K8 overexpression was correlated with damaged survival in ccRC and may play a crucial role in cancer-related inflammation via I-kappaB kinase/NF-kappaB and toll-like receptor signaling pathways.

DAGM: A novel modelling framework to assess the risk of HER2-negative breast cancer based on germline rare coding mutations

BACKGROUND

Breast cancers can be divided into HER2-negative and HER2-positive subtypes according to different status of HER2 gene. Despite extensive studies connecting germline mutations with possible risk of HER2-negative breast cancer, the main category of breast cancer, it remains challenging to obtain accurate risk assessment and to understand the potential underlying mechanisms.

METHODS

We developed a novel framework named Damage Assessment of Genomic Mutations (DAGM), which projects rare coding mutations and gene expressions into Activity Profiles of Signalling Pathways (APSPs).

FINDINGS

We characterized and validated DAGM framework at multiple levels. Based on an input of germline rare coding mutations, we obtained the corresponding APSP spectrum to calculate the APSP risk score, which was capable of distinguish HER2-negative from HER2-positive cases. These findings were validated using breast cancer data from TCGA (AUC = 0.7). DAGM revealed that HER2 signalling pathway was up-regulated in germline of HER2-negative patients, and those with high APSP risk scores had exhibited immune suppression. These findings were validated using RNA sequencing, phosphoproteome analysis, and CyTOF. Moreover, using germline mutations, DAGM could evaluate the risk for HER2-negative breast cancer, not only in women carrying BRCA1/2 mutations, but also in those without known disease-associated mutations.

INTERPRETATION

The DAGM can facilitate the screening of subjects at high risk of HER2-negative breast cancer for primary prevention. This study also provides new insights into the potential mechanisms of developing HER2-negative breast cancer. The DAGM has the potential to be applied in the prevention, diagnosis, and treatment of HER2-negative breast cancer.

FUNDING

This work was supported by the National Key Research and Development Program of China (grant no. 2018YFC0910406 and 2018AAA0103302 to CZ); the National Natural Science Foundation of China (grant no. 81202076 and 82072939 to MY, 81871513 to KW); the Guangzhou Science and Technology Program key projects (grant no. 2014J2200007 to MY, 202002030236 to KW); the National Key R&D Program of China (grant no. 2017YFC1309100 to CL); Shenzhen Science and Technology Planning Project (grant no. JCYJ20170817095211560 574 to YN); and the Natural Science Foundation of Guangdong Province (grant no. 2017A030313882 to KW and S2013010012048 to MY); Hefei National Laboratory for Physical Sciences at the Microscale (grant no. KF2020009 to GN); and RGC General Research Fund (grant no. 17114519 to YQS).

Uncovering the Subtype-Specific Molecular Characteristics of Breast Cancer by Multiomics Analysis of Prognosis-Associated Genes, Driver Genes, Signaling Pathways, and Immune Activity

Breast cancer is a heterogeneous malignant disease with different prognoses and has been divided into four molecular subtypes. It is believed that molecular events occurring in breast stem/progenitor cells contribute to the carcinogenesis and development of different breast cancer subtypes. However, these subtype-specific molecular characteristics are largely unknown. In this study, we employed 1217 breast cancer samples from The Cancer Genome Atlas (TCGA) database for a multiomics analysis of the molecular characteristics of different breast cancer subtypes based on PAM50 algorithms. We detected the expression changes of subtype-specific genes and revealed that the expression of particular subtype-specific genes significantly affected prognosis. We also investigated the mutations and copy number variations (CNVs) of breast cancer driver genes and the representative genes of ten signaling pathways in different subtypes and revealed several subtype-specifically altered genes. Moreover, we detected the infiltration of various immune cells in different subtypes of breast cancer and showed that the infiltration levels of major immune cell types are different among these subtypes. Additionally, we investigated the factors affecting the immune infiltration level and the immune cytolytic activity in different breast cancer subtypes, namely, the mutation burden, genome instability and cancer-associated fibroblast (CAF) infiltration. This study may shed light on the molecular events contributing to carcinogenesis and development and provide potential markers and targets for the clinical diagnosis and treatment of different breast cancer subtypes.

CYP2J2 Is a Diagnostic and Prognostic Biomarker Associated with Immune Infiltration in Kidney Renal Clear Cell Carcinoma

Cytochrome P450 family 2 subfamily J member 2 (CYP2J2), a member of the monooxygenase cytochrome P450 (CYP) family and the only member of the human CYP2J subfamily, has many functions, including regulation of oxidative stress, inflammation, apoptosis, and immune responses. However, its role in cancer development has not been clearly elucidated. In this study, expression levels of CYP2J2 in various cancer types were determined using the Oncomine, the Gene Expression Profiling Interactive Analysis (GEIPA), DriverDBv3, UALCAN, and Tumor Immune Estimation Resource (TIMER) databases. The prognostic value of CYP2J2 for KIRC was analyzed using GEPIA, UALCAN, OSkirc, and DriverDBv3 databases. We evaluated the expression levels of CYP2J2 transcript, protein, and promoter methylation at different clinical characteristics in KIRC through the UALCAN database. Simultaneously, CYP2J2 network-related functions were evaluated using the GeneMANIA interactive tool while the biological processes involved in CYP2J2 and its interactive genes were investigated through Metascape and FunRich. Then, we used TIMER to determine the correlation between CYP2J2 expression levels and immune infiltration levels in KIRC. In KIRC, the CYP2J2 gene, RNA, and protein were found to be overexpressed. However, the methylation level of CYP2J2 promoter in KIRC was lower than in normal tissues. Surprisingly, elevated expression levels of CYP2J2 exhibited better prognostic outcomes in KIRC. Evaluation of protein-protein interaction networks and biological processes revealed that CYP2J2 was principally involved in immune responses, apoptosis, and other metabolic processes. Moreover, we found that the expression levels of CYP2J2 were positively correlated with infiltration levels of B cells, CD8 + T cells, neutrophils, and dendritic cells in KIRC. Therefore, we speculated that the overexpression of CYP2J2 prolonged the survival outcome of KIRC patients, which may be related to the change of tumor immune microenvironment. Moreover, all these new understandings of CYP2J2 may provide important value for the early diagnosis and new targeted drug therapy of KIRC.

Gene Regulation Network of Prognostic Biomarker YAP1 in Human Cancers: An Integrated Bioinformatics Study

Background: Yes-associated protein 1 (YAP1) is the main downstream effector of the Hippo signaling pathway, which is involved in tumorigenesis. This study aimed to comprehensively understand the prognostic performances of YAP1 expression and its potential mechanism in pan-cancers by mining databases.

Methods: The YAP1 expression was evaluated by the Oncomine database and GEPIA tool. The clinical significance of YAP1 expression was analyzed by the UALCAN, GEPIA, and DriverDBv3 database. Then, the co-expressed genes with YAP1 were screened by the LinkedOmics, and annotated by the Metascape and DAVID database. Additionally, by the MitoMiner 4.0 v tool, the YAP1 co-expressed genes were screened to obtain the YAP1-associated mitochondrial genes that were further enriched by DAVID and analyzed by MCODE for the hub genes.

Results: YAP1 was differentially expressed in human cancers. Higher YAP1 expression was significantly associated with poorer overall survival and disease-free survival in adrenocortical carcinoma (ACC), brain Lower Grade Glioma (LGG), and pancreatic adenocarcinoma (PAAD). The LinkedOmics analysis revealed 923 co-expressed genes with YAP1 in adrenocortical carcinoma, LGG and PAAD. The 923 genes mainly participated in mitochondrial functions including mitochondrial gene expression and mitochondrial respiratory chain complex I assembly. Of the 923 genes, 112 mitochondrial genes were identified by MitoMiner 4.0 v and significantly enriched in oxidative phosphorylation. The MCODE analysis identified three hub genes including CHCHD1, IDH3G and NDUFAF5.

Conclusion: Our findings showed that the YAP1 overexpression could be a biomarker for poor prognosis in ACC, LGG and PAAD. Specifically, the YAP1 co-expression genes were mainly involved in the regulation of mitochondrial function especially in oxidative phosphorylation. Thus, our findings provided evidence of the carcinogenesis of YAP1 in human cancers and new insights into the mechanisms underlying the role of YAP1 in mitochondrial dysregulation.

Identifying Key Somatic Copy Number Alterations Driving Dysregulation of Cancer Hallmarks in Lower-Grade Glioma

Somatic copy-number alterations (SCNAs) are major contributors to cancer development that are pervasive and highly heterogeneous in human cancers. However, the driver roles of SCNAs in cancer are insufficiently characterized. We combined network propagation and linear regression models to design an integrative strategy to identify driver SCNAs and dissect the functional roles of SCNAs by integrating profiles of copy number and gene expression in lower-grade glioma (LGG). We applied our strategy to 511 LGG patients and identified 98 driver genes that dysregulated 29 cancer hallmark signatures, forming 143 active gene-hallmark pairs. We found that these active gene-hallmark pairs could stratify LGG patients into four subtypes with significantly different survival times. The two new subtypes with similar poorest prognoses were driven by two different gene sets (one including EGFR, CDKN2A, CDKN2B, INFA8, and INFA5, and the other including CDK4, AVIL, and DTX3), respectively. The SCNAs of the two gene sets could disorder the same cancer hallmark signature in a mutually exclusive manner (including E2F_TARGETS and G2M_CHECKPOINT). Compared with previous methods, our strategy could not only capture the known cancer genes and directly dissect the functional roles of their SCNAs in LGG, but also discover the functions of new driver genes in LGG, such as IFNA5, IFNA8, and DTX3. Additionally, our method can be applied to a variety of cancer types to explore the pathogenesis of driver SCNAs and improve the treatment and diagnosis of cancer.

Inhibiting of TACC3 Promotes Cell Proliferation, Cell Invasion and the EMT Pathway in Breast Cancer

Accumulating evidences indicate that transforming acidic coiled-coil 3 (TACC3) is a tumor-related gene, was highly expressed in a variety of human cancers, which is involved in cancer development. However, the potential role of TACC3 in breast cancer remains largely unknown. In the present study, we found that TACC3 was highly-expressed in breast cancer tissues, and its level was positively correlated with the clinical features of breast cancer patients. Specifically, TACC3 expression was significantly associated with the estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2) status, nodal status, the scarff-bloom-richardson (SBR) grade, nottingham prognostic index (NPI), age, subtypes, and triple-negative and basal-like status, suggesting that TACC3 may be a potential diagnostic indicator of breast cancer. Furthermore, functional studies have shown that inhibition of TACC3 can significantly promote the cell proliferation and viability of breast cancer cells. Moreover, TACC3 knockdown suppressed the expression of E-cadherin, but increased the expression of N-cadherin, Snail, ZEB1, and TWIST, which indicate that TACC3 may impact the migration of breast cancer cells in vitro. Taken together, these findings indicate that TACC3 may serve as a prognostic and therapeutic indicator of breast cancer.

Analysis of Nuclear Encoded Mitochondrial Gene Networks in Cervical Cancer

BACKGROUND

Cervical cancer (CC) is one of the most common female cancers in many developing and underdeveloped countries. High incidence, late presentation, and mortality suggested the need for molecular markers. Mitochondrial defects due to abnormal expression of nuclear-encoded mitochondrial genes (NEMG) have been reported during cancer progression. Nevertheless, the application of NEMG for the prognosis of CC is still elusive. Herein, we aimed to investigate the associations between NEMG and CC prognosis.

MATERIALS AND METHODS

The differentially expressed genes (DEG) in the TCGA-CESC dataset and NEMGs were retrieved from TACCO and Mitocarta2.0 databases, respectively. The impact of methylation on NEMG expression were predicted using DNMIVD and UALCAN tools. HCMDB tool was used to predict genes having metastatic potential. The prognostic models were constructed using DNMIVD, TACCO, GEPIA2, and SurvExpress. The functional enrichment analysis (FEA) was performed using clusterProfiler. The protein-protein interaction network (PPIN) was constructed to identify the hub genes (HG) using String and CytoHubba tools. Independent validation of the HG was performed using Oncomine and Human Protein Atlas databases. The druggable genes were predicted using DGIdb.

RESULTS

Among the 52 differentially expressed NEMG, 15 were regulated by DNA methylation. The expression level of 16, 10, and 7 has the potential for CC staging, prediction of metastasis, and prognosis. Moreover, 1 driver gene and 16 druggable genes were also identified. The FEA identified the enrichment of cancer-related pathways, including AMPK and carbon metabolism in cancer. The combined expression of 10 HG has been shown to affect patient survival.

CONCLUSION

Our findings suggest that the abnormal expression of NEMGs may play a critical role in CC development and progression. The genes identified in our study may serve as a prognostic indicator and therapeutic target in CC.
.

driveR: a novel method for prioritizing cancer driver genes using somatic genomics data

Background

Cancer develops due to “driver” alterations. Numerous approaches exist for predicting cancer drivers from cohort-scale genomics data. However, methods for personalized analysis of driver genes are underdeveloped. In this study, we developed a novel personalized/batch analysis approach for driver gene prioritization utilizing somatic genomics data, called driveR.

Results

Combining genomics information and prior biological knowledge, driveR accurately prioritizes cancer driver genes via a multi-task learning model. Testing on 28 different datasets, this study demonstrates that driveR performs adequately, achieving a median AUC of 0.684 (range 0.651–0.861) on the 28 batch analysis test datasets, and a median AUC of 0.773 (range 0–1) on the 5157 personalized analysis test samples. Moreover, it outperforms existing approaches, achieving a significantly higher median AUC than all of MutSigCV (Wilcoxon rank-sum test p < 0.001), DriverNet (p < 0.001), OncodriveFML (p < 0.001) and MutPanning (p < 0.001) on batch analysis test datasets, and a significantly higher median AUC than DawnRank (p < 0.001) and PRODIGY (p < 0.001) on personalized analysis datasets.

Conclusions

This study demonstrates that the proposed method is an accurate and easy-to-utilize approach for prioritizing driver genes in cancer genomes in personalized or batch analyses. driveR is available on CRAN: https://cran.r-project.org/package=driveR.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04203-7.

VWCE as a potential biomarker associated with immune infiltrates in breast cancer

PURPOSE

Von Willebrand Factor C and EGF Domains (VWCE) is an important gene that regulates cell adhesion, migration, and interaction. However, the correlation between VWCE expression and immune infiltrating in breast cancer remain unclear. In this study, we investigated the correlation between VWCE expression and immune infiltration levels in breast cancer.

METHODS

The expression of VWCE was analyzed by the tumor immune estimation resource (TIMER) and DriverDB databases. Furthermore, genes co-expressed with VWCE and gene ontology (GO) enrichment analysis were investigated by the STRING and Enrichr web servers. Also, we performed the single nucleotide variation (SNV), copy number variation (CNV), and pathway activity analysis through GSCALite. Subsequently, the relationship between VWCE expression and tumor immunity was analyzed by TIMER and TISIDB databases, and further verified the results using Quantitative Real-Time PCR (RT-PCR), Western blotting, and immunohistochemistry.

RESULTS

The results showed that the expression of VWCE mRNA in breast cancer tissue was significantly lower than that in normal tissues. We found that the expression level of VWCE was associated with subtypes, estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor-2 (HER2) status of breast cancer patients, but there was no significant difference in the expression of VWCE was found in age and nodal status. Further analyses indicated that VWCE was correlated with the activation or inhibition of multiple oncogenic pathways. Additionally, VWCE expression was negatively correlated with the expression of STAT1 (Th1 marker, r = - 0.12, p = 6e-05), but positively correlated with the expression of MS4A4A (r = 0.28, p = 0). These results suggested that the expression of VWCE was correlated with immune infiltration levels of Th1 and M2 macrophage in breast cancer.

CONCLUSIONS

In our study, VWCE expression was associated with a better prognosis and was immune infiltration in breast cancer. These findings demonstrate that VWCE is a potential prognostic biomarker and correlated with tumor immune cell infiltration, and maybe a promising therapeutic target in breast cancer.

Elevated Expression of PDZD11 Is Associated With Poor Prognosis and Immune Infiltrates in Hepatocellular Carcinoma

Epithelial cells are held together by tight and adherent junctions, which are destroyed by the activation of epithelial-to-mesenchymal transition (EMT). The PLEKHA7-PDZD11 complex has been reported to be important for epithelial cell adhesion and connecting tissues. However, there is no research regarding the expression and role of PDZD11 in liver hepatocellular carcinoma (LIHC) progression. Here, we analyzed PDZD11 mRNA expression and its clinical results in LIHC patient RNA sequencing data based on different open databases. Furthermore, we examined differences in PDZD11 expression in LIHC tissues and cell lines using western blotting and real-time qPCR. These results are the first to report that the mRNA and protein levels of PDZD11 are significantly overexpressed in LIHC. Moreover, high expression of PDZD11 was correlated with poor overall survival in patients with LIHC. Gene regulatory network analysis suggested that PDZD11 is mainly involved in copper ion homeostasis, proteasome, and oxidative phosphorylation pathways. Interestingly, we found that PDZD11 levels were positively correlated with the abundance of immune infiltrates. In particular, higher infiltration levels of CD4⁺ T cells and macrophage subsets significantly affected LIHC patient prognosis. Taken together, these results demonstrate that PDZD11 could be a potential diagnostic and prognostic biomarker in LIHC.

Systematic identification of clinically relevant miRNAs for potential miRNA-based therapy in lung adenocarcinoma

Lung adenocarcinoma (LUAD), the most common histological type of non-small cell lung cancer, is one of the most malignant and deadly diseases. Current treatments for advanced LUAD patients are far from ideal and require further improvements. Here, we utilized a systematic integrative analysis of LUAD microRNA sequencing (miRNA-seq) and RNA-seq data from The Cancer Genome Atlas (TCGA) to identify clinically relevant tumor suppressor miRNAs. Three miRNA candidates (miR-195-5p, miR-101-3p, and miR-338-5p) were identified based on their differential expressions, survival significance levels, correlations with targets, and an additive effect on survival among them. We further evaluated mimics of the three miRNAs to determine their therapeutic potential in inhibiting cancer progression. The results showed not only that each of the miRNA mimics alone but also the three miRNA mimics in combination were efficient at inhibiting tumor growth and progression with equal final concentrations, meaning that the three miRNA mimics in combination were more effective than the single miRNA mimics. Moreover, the combined miRNA mimics provided significant therapeutic effects in terms of reduced tumor volume and metastasis nodules in lung tumor animal models. Hence, our findings show the potential of using the three miRNAs in combination to treat LUAD patients with poor survival outcomes.

The microRNA-210-Stathmin1 Axis Decreases Cell Stiffness to Facilitate the Invasiveness of Colorectal Cancer Stem Cells

Simple Summary

Metastasis of tumor cells is the leading cause of death in cancer patients. Concurrent therapy with surgical removal of primary and metastatic lesions is the main approach for cancer therapy. Currently, therapeutic resistant properties of cancer stem cells (CSCs) are known to drive malignant cancer progression, including metastasis. Our study aimed to identify molecular tools dedicated to the detection and treatment of CSCs. We confirmed that microRNA-210-3p (miR-210) was upregulated in colorectal stem-like cancer cells, which targeted stathmin1 (STMN1), to decrease cell elasticity for increasing mobility. We envision that strategies for softening cellular elasticity will reduce the onset of CSC-orientated metastasis.

Abstract

Cell migration is critical for regional dissemination and distal metastasis of cancer cells, which remain the major causes of poor prognosis and death in patients with colorectal cancer (CRC). Although cytoskeletal dynamics and cellular deformability contribute to the migration of cancer cells and metastasis, the mechanisms governing the migratory ability of cancer stem cells (CSCs), a nongenetic source of tumor heterogeneity, are unclear. Here, we expanded colorectal CSCs (CRCSCs) as colonospheres and showed that CRCSCs exhibited higher cell motility in transwell migration assays and 3D invasion assays and greater deformability in particle tracking microrheology than did their parental CRC cells. Mechanistically, in CRCSCs, microRNA-210-3p (miR-210) targeted stathmin1 (STMN1), which is known for inducing microtubule destabilization, to decrease cell elasticity in order to facilitate cell motility without affecting the epithelial–mesenchymal transition (EMT) status. Clinically, the miR-210-STMN1 axis was activated in CRC patients with liver metastasis and correlated with a worse clinical outcome. This study elucidates a miRNA-oriented mechanism regulating the deformability of CRCSCs beyond the EMT process.

Artificial intelligence in cancer research: learning at different levels of data granularity

From genome-scale experimental studies to imaging data, behavioral footprints, and longitudinal healthcare records, the convergence of big data in cancer research and the advances in Artificial Intelligence (AI) is paving the way to develop a systems view of cancer. Nevertheless, this biomedical area is largely characterized by the co-existence of big data and small data resources, highlighting the need for a deeper investigation about the crosstalk between different levels of data granularity, including varied sample sizes, labels, data types, and other data descriptors. This review introduces the current challenges, limitations, and solutions of AI in the heterogeneous landscape of data granularity in cancer research. Such a variety of cancer molecular and clinical data calls for advancing the interoperability among AI approaches, with particular emphasis on the synergy between discriminative and generative models that we discuss in this work with several examples of techniques and applications.

High Mobility Group Box 1 Promotes Lung Cancer Cell Migration and Motility via Regulation of Dynamin-Related Protein 1

High mobility group box 1 (HMGB1) has been demonstrated to promote the migration and invasion of non-small cell lung cancer (NSCLC). However, the mechanism of action of HMGB1 in regulating tumor mobility remains unclear. Therefore, we aimed to investigate whether HMGB1 affects mitochondria distribution and regulates dynamin-related protein 1 (DRP1)-mediated lamellipodia/filopodia formation to promote NSCLC migration. The regulation of mitochondrial membrane tension, dynamics, polarization, fission process, and cytoskeletal rearrangements in lung cancer cells by HMGB1 was analyzed using confocal microscopy. The HMGB1-mediated regulation of DRP1 phosphorylation and colocalization was determined using immunostaining and co-immunoprecipitation assays. The tumorigenic potential of HMGB1 was assessed in vivo and further confirmed using NSCLC patient samples. Our results showed that HMGB1 increased the polarity and mobility of cells (mainly by regulating the cytoskeletal system actin and microtubule dynamics and distribution), promoted the formation of lamellipodia/filopodia, and enhanced the expression and phosphorylation of DRP1 in both the nucleus and cytoplasm. In addition, HMGB1 and DRP1 expressions were positively correlated and exhibited poor prognosis and survival in patients with lung cancer. Collectively, HMGB1 plays a key role in the formation of lamellipodia and filopodia by regulating cytoskeleton dynamics and DRP1 expression to promote lung cancer migration.

Algorithm-Based Meta-Analysis Reveals the Mechanistic Interaction of the Tumor Suppressor LIMD1 With Non-Small-Cell Lung Carcinoma

Non-small-cell lung carcinoma (NSCLC) is the major type of lung cancer, which is among the leading causes of cancer-related deaths worldwide. LIMD1 was previously identified as a tumor suppressor in lung cancer, but their detailed interaction in this setting remains unclear. In this study, we have carried out multiple genome-wide bioinformatic analyses for a comprehensive understanding of LIMD1 in NSCLC, using various online algorithm platforms that have been built for mega databases derived from both clinical and cell line samples. Our results indicate that LIMD1 expression level is significantly downregulated at both mRNA and protein levels in both lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC), with a considerable contribution from its promoter methylation rather than its gene mutations. The Limd1 gene undergoes mutation only at a low rate in NSCLC (0.712%). We have further identified LIMD1-associated molecular signatures in NSCLC, including its natural antisense long non-coding RNA LIMD1-AS1 and a pool of membrane trafficking regulators. We have also identified a subgroup of tumor-infiltrating lymphocytes, especially neutrophils, whose tumor infiltration levels significantly correlate with LIMD1 level in both LUAD and LUSC. However, a significant correlation of LIMD1 with a subset of immune regulatory molecules, such as IL6R and TAP1, was only found in LUAD. Regarding the clinical outcomes, LIMD1 expression level only significantly correlates with the survival of LUAD (p<0.01) but not with that of LUSC (p>0.1) patients. These findings indicate that LIMD1 plays a survival role in LUAD patients at least by acting as an immune regulatory protein. To further understand the mechanisms underlying the tumor-suppressing function of LIMD1 in NSCLC, we show that LIMD1 downregulation remarkably correlates with the deregulation of multiple pathways that play decisive roles in the oncogenesis of NSCLC, especially those mediated by EGFR, KRAS, PIK3CA, Keap1, and p63, in both LUAD and LUSC, and those mediated by p53 and CDKN2A only in LUAD. This study has disclosed that LIMD1 can serve as a survival prognostic marker for LUAD patients and provides mechanistic insights into the interaction of LIMD1 with NSCLC, which provide valuable information for clinical applications.

Statistical and bioinformatic analysis of hemimethylation patterns in non-small cell lung cancer

Background

DNA methylation is an epigenetic event involving the addition of a methyl-group to a cytosine-guanine base pair (i.e., CpG site). It is associated with different cancers. Our research focuses on studying non-small cell lung cancer hemimethylation, which refers to methylation occurring on only one of the two DNA strands. Many studies often assume that methylation occurs on both DNA strands at a CpG site. However, recent publications show the existence of hemimethylation and its significant impact. Therefore, it is important to identify cancer hemimethylation patterns.

Methods

In this paper, we use the Wilcoxon signed rank test to identify hemimethylated CpG sites based on publicly available non-small cell lung cancer methylation sequencing data. We then identify two types of hemimethylated CpG clusters, regular and polarity clusters, and genes with large numbers of hemimethylated sites. Highly hemimethylated genes are then studied for their biological interactions using available bioinformatics tools.

Results

In this paper, we have conducted the first-ever investigation of hemimethylation in lung cancer. Our results show that hemimethylation does exist in lung cells either as singletons or clusters. Most clusters contain only two or three CpG sites. Polarity clusters are much shorter than regular clusters and appear less frequently. The majority of clusters found in tumor samples have no overlap with clusters found in normal samples, and vice versa. Several genes that are known to be associated with cancer are hemimethylated differently between the cancerous and normal samples. Furthermore, highly hemimethylated genes exhibit many different interactions with other genes that may be associated with cancer. Hemimethylation has diverse patterns and frequencies that are comparable between normal and tumorous cells. Therefore, hemimethylation may be related to both normal and tumor cell development.

Conclusions

Our research has identified CpG clusters and genes that are hemimethylated in normal and lung tumor samples. Due to the potential impact of hemimethylation on gene expression and cell function, these clusters and genes may be important to advance our understanding of the development and progression of non-small cell lung cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-07990-7.

Role of the NLRP3 Inflammasome: Insights Into Cancer Hallmarks

In response to a variety of stresses, mammalian cells activate the inflammasome for targeted caspase-dependent pyroptosis. The research community has recently begun to deduce that the activation of inflammasome is instigated by several known oncogenic stresses and metabolic perturbations; nevertheless, the role of inflammasomes in the context of cancer biology is less understood. In manipulating the expression of inflammasome, researchers have found that NLRP3 serves as a deterministic player in conducting tumor fate decisions. Understanding the mechanistic underpinning of pro-tumorigenic and anti-tumorigenic pathways might elucidate novel therapeutic onco-targets, thereby providing new opportunities to manipulate inflammasome in augmenting the anti-tumorigenic activity to prevent tumor expansion and achieve metastatic control. Accordingly, this review aims to decode the complexity of NLRP3, whereby summarizing and clustering findings into cancer hallmarks and tissue contexts may expedite consensus and underscore the potential of the inflammasome in drug translation.

Identification of theranostic factors for patients developing metastasis after surgery for early-stage lung adenocarcinoma

Rationale: Lung adenocarcinoma (LUAD) is an aggressive disease with high propensity of metastasis. Among patients with early-stage disease, more than 30% of them may relapse or develop metastasis. There is an unmet medical need to stratify patients with early-stage LUAD according to their risk of relapse/metastasis to guide preventive or therapeutic approaches. In this study, we identified 4 genes that can serve both therapeutic and diagnostic (theranostic) purposes.

Methods: Three independent datasets (GEO, TCGA, and KMPlotter) were used to evaluate gene expression profile of patients with LUAD by unbiased screening approach. Upon significant genes uncovered, functional enrichment analysis was carried out. The predictive power of their expression on patient prognosis were evaluated. Once confirmed their theranostic roles by integrated bioinformatics, we further conducted in vitro and in vivo validation.

Results: We found that four genes (ADAM9, MTHFD2, RRM2, and SLC2A1) were associated with poor patient outcomes with an increased hazard ratio in LUAD. Knockdown of them, both separately and simultaneously, suppressed lung cancer cell proliferation and migration ability in vitro and prolonged survival time in metastatic tumor mouse models. Moreover, these four biomarkers were found to be overexpressed in tumor tissues from LUAD patients, and the total immunohistochemical staining scores correlated with poor prognosis.

Conclusions: These results suggest that these four identified genes could be theranostic biomarkers for stratifying high-risk patients who develop relapse/metastasis in early-stage LUAD. Developing therapeutic approaches for the four biomarkers may benefit early-stage LUAD patients after surgery.

Comprehensive analysis of glutathione peroxidase-1 (GPX1) expression and prognostic value in three different types of renal cell carcinoma

Background

Glutathione peroxidase-1 (GPX1) is generally expressed in tissues with high oxygen tension such as the kidneys and lungs, and its main function is to degrade reactive oxygen species (ROS) and protect cells from oxidative stress. Studies have shown that GPX1 is upregulated in many tumor tissues and is closely related to tumor progression and metastasis. This study aimed to explore the possibility of GPX1 as a biomarker for kidney chromophobe cell carcinoma (KICH), kidney renal papillary cell carcinoma (KIRP), and kidney renal clear cell carcinoma (KIRC).

Methods

The Oncomine and GEPIA databases were used to analyze the GPX1 expression differences between tumor and normal tissues, and the UALCAN, GEPIA and DriverDBv3 databases were used to perform the survival analyses. The GeneMANIA interactive tool was then used to find the GPX1-related protein-protein interaction (PPI). Following this, the LinkedOmics database was used for the enrichment analysis of GPX1, and the Timer database was used to estimate the abundance of immune infiltration. Finally, quantitative polymerase chain reaction (qPCR) was performed on patient specimens collected in the clinic to confirm the database findings.

Results

In our study, we found that the expression of GPX1 in three types of renal cell carcinoma (RCC) were upregulated, and the high expression of GXP1 was related to the poor prognosis of patients with KICH and KIRC. On the contrary, KIRP patients with a high expression of GPX1 had a better prognosis. In addition, GPX1 was related to the abundance of immune cell infiltration. The results of the qPCR analysis confirmed that the expression of GPX1 in RCC was increased compared with the control group (P<0.05).

Conclusions

Our results indicate that the expression of GPX1 is related to the prognosis of three types of RCC. As such, GPX1 expression could be a reliable diagnostic and prognostic biomarker for RCC and, more importantly, may provide a new direction for therapeutic strategies.

OncoVar: an integrated database and analysis platform for oncogenic driver variants in cancers

The prevalence of neutral mutations in cancer cell population impedes the distinguishing of cancer-causing driver mutations from passenger mutations. To systematically prioritize the oncogenic ability of somatic mutations and cancer genes, we constructed a useful platform, OncoVar (https://oncovar.org/), which employed published bioinformatics algorithms and incorporated known driver events to identify driver mutations and driver genes. We identified 20 162 cancer driver mutations, 814 driver genes and 2360 pathogenic pathways with high-confidence by reanalyzing 10 769 exomes from 33 cancer types in The Cancer Genome Atlas (TCGA) and 1942 genomes from 18 cancer types in International Cancer Genome Consortium (ICGC). OncoVar provides four points of view, 'Mutation', 'Gene', 'Pathway' and 'Cancer', to help researchers to visualize the relationships between cancers and driver variants. Importantly, identification of actionable driver alterations provides promising druggable targets and repurposing opportunities of combinational therapies. OncoVar provides a user-friendly interface for browsing, searching and downloading somatic driver mutations, driver genes and pathogenic pathways in various cancer types. This platform will facilitate the identification of cancer drivers across individual cancer cohorts and helps to rank mutations or genes for better decision-making among clinical oncologists, cancer researchers and the broad scientific community interested in cancer precision medicine.

Genomic and transcriptomic landscape of conjunctival melanoma

Conjunctival melanoma (CJM) is a rare but potentially lethal and highly-recurrent cancer of the eye. Similar to cutaneous melanoma (CM), it originates from melanocytes. Unlike CM, however, CJM is relatively poorly characterized from a genomic point of view. To fill this knowledge gap and gain insight into the genomic nature of CJM, we performed whole-exome (WES) or whole-genome sequencing (WGS) of tumor-normal tissue pairs in 14 affected individuals, as well as RNA sequencing in a subset of 11 tumor tissues. Our results show that, similarly to CM, CJM is also characterized by a very high mutation load, composed of approximately 500 somatic mutations in exonic regions. This, as well as the presence of a UV light-induced mutational signature, are clear signs of the role of sunlight in CJM tumorigenesis. In addition, the genomic classification of CM proposed by TCGA seems to be well-applicable to CJM, with the presence of four typical subclasses defined on the basis of the most frequently mutated genes: BRAF, NF1, RAS, and triple wild-type. In line with these results, transcriptomic analyses revealed similarities with CM as well, namely the presence of a transcriptomic subtype enriched for immune genes and a subtype enriched for genes associated with keratins and epithelial functions. Finally, in seven tumors we detected somatic mutations in ACSS3, a possible new candidate oncogene. Transfected conjunctival melanoma cells overexpressing mutant ACSS3 showed higher proliferative activity, supporting the direct involvement of this gene in the tumorigenesis of CJM. Altogether, our results provide the first unbiased and complete genomic and transcriptomic classification of CJM.

Conjunctival melanoma is an extremely rare form of cancer of the eye that arises from melanocytes–the cells producing the protective pigment melanin–in the outmost layer of the eye: the conjunctiva. This tissue, similarly to the skin, can also be exposed to UV light radiation from the sun. We investigated the genetic background of this rare form of cancer in samples from fourteen patients, by global DNA and RNA sequencing. Our results showed that conjunctival melanoma is genetically very similar to cutaneous melanoma. More precisely, in tumor DNA we detected signs of damage caused by UV light, as well as mutations in the genes BRAF, NF1 and NRAS/HRAS, previously described to be involved in cutaneous melanoma. Analysis of tumor gene expression also revealed similarities between these two types of cancer, some of which could be used as prognostic factors or as indicators of a patients’ response to therapy. In addition, we identified frequent somatic mutations in ACSS3, a gene not yet associated with either conjunctival or cutaneous melanoma, which represents a potential key player in oncogenesis of conjunctival melanoma.

Identification and validation of a miRNA-based prognostic signature for cervical cancer through an integrated bioinformatics approach

Cervical cancer is the fourth most common cancer in women worldwide. Increasing evidence has shown that miRNAs are related to the progression of cervical cancer. However, the mechanisms that affect the prognosis of cancer are still largely unknown. In the present study, we sought to identify miRNAs associated with poor prognosis of patient with cervical cancer, as well as the possible mechanisms regulated by them. The miRNA expression profiles and relevant clinical information of patients with cervical cancer were obtained from The Cancer Genome Atlas (TCGA). The selection of prognostic miRNAs was carried out through an integrated bioinformatics approach. The most effective miRNAs with synergistic and additive effects were selected for validation through in vitro experiments. Three miRNAs (miR-216b-5p, miR-585-5p, and miR-7641) were identified as exhibiting good performance in predicting poor prognosis through additive effects analysis. The functional enrichment analysis suggested that not only pathways traditionally involved in cancer but also immune system pathways might be important in regulating the outcome of the disease. Our findings demonstrated that a synergistic combination of three miRNAs may be associated, through their regulation of specific pathways, with very poor survival rates for patients with cervical cancer.

Multi-Approach Bioinformatics Analysis of Curated Omics Data Provides a Gene Expression Panorama for Multiple Cancer Types

Studies describing the expression patterns and biomarkers for the tumoral process increase in number every year. The availability of new datasets, although essential, also creates a confusing landscape where common or critical mechanisms are obscured amidst the divergent and heterogeneous nature of such results. In this work, we manually curated the Gene Expression Omnibus using rigorous filtering criteria to select the most homogeneous and highest quality microarray and RNA-seq datasets from multiple types of cancer. By applying systems biology approaches, combined with machine learning analysis, we investigated possible frequently deregulated molecular mechanisms underlying the tumoral process. Our multi-approach analysis of 99 curated datasets, composed of 5,406 samples, revealed 47 differentially expressed genes in all analyzed cancer types, which were all in agreement with the validation using TCGA data. Results suggest that the tumoral process is more related to the overexpression of core deregulated machinery than the underexpression of a given gene set. Additionally, we identified gene expression similarities between different cancer types not described before and performed an overall survival analysis using 20 cancer types. Finally, we were able to suggest a core regulatory mechanism that could be frequently deregulated.

Integration of Online Omics-Data Resources for Cancer Research

The manifestations of cancerous phenotypes necessitate alterations at different levels of information-flow from genome to proteome. The molecular alterations at different information processing levels serve as the basis for the cancer phenotype to emerge. To understand the underlying mechanisms that drive the acquisition of cancer hallmarks it is required to interrogate cancer cells using multiple levels of information flow represented by different omics - such as genomics, epigenomics, transcriptomics, and proteomics. The advantage of multi-omics data integration comes with a trade-off in the form of an added layer of complexity originating from inherently diverse types of omics-datasets that may pose a challenge to integrate the omics-data in a biologically meaningful manner. The plethora of cancer-specific online omics-data resources, if able to be integrated efficiently and systematically, may facilitate the generation of new biological insights for cancer research. In this review, we provide a comprehensive overview of the online single- and multi-omics resources that are dedicated to cancer. We catalog various online omics-data resources such as The Cancer Genome Atlas (TCGA) along with various TCGA-associated data portals and tools for multi-omics analysis and visualization, the International Cancer Genome Consortium (ICGC), Catalogue of Somatic Mutations in Cancer (COSMIC), The Pathology Atlas, Gene Expression Omnibus (GEO), and PRoteomics IDEntifications (PRIDE). By comparing the strengths and limitations of the respective online resources, we aim to highlight the current biological and technological challenges and possible strategies to overcome these challenges. We outline the available schemes for the integration of the multi-omics dimensions for stratifying cancer patients and biomarker prediction based on the integrated molecular-signatures of cancer. Finally, we propose the multi-omics driven systems-biology approaches to realize the potential of precision onco-medicine as the future of cancer research. We believe this systematic review will encourage scientists and clinicians worldwide to utilize the online resources to explore and integrate the available omics datasets that may provide a window of opportunity to generate new biological insights and contribute to the advancement of the field of cancer research.

Systems Biology and Experimental Model Systems of Cancer

Over the past decade, we have witnessed an increasing number of large-scale studies that have provided multi-omics data by high-throughput sequencing approaches. This has particularly helped with identifying key (epi)genetic alterations in cancers. Importantly, aberrations that lead to the activation of signaling networks through the disruption of normal cellular homeostasis is seen both in cancer cells and also in the neighboring tumor microenvironment. Cancer systems biology approaches have enabled the efficient integration of experimental data with computational algorithms and the implementation of actionable targeted therapies, as the exceptions, for the treatment of cancer. Comprehensive multi-omics data obtained through the sequencing of tumor samples and experimental model systems will be important in implementing novel cancer systems biology approaches and increasing their efficacy for tailoring novel personalized treatment modalities in cancer. In this review, we discuss emerging cancer systems biology approaches based on multi-omics data derived from bulk and single-cell genomics studies in addition to existing experimental model systems that play a critical role in understanding (epi)genetic heterogeneity and therapy resistance in cancer.

GMIEC: a shiny application for the identification of gene-targeted drugs for precision medicine

Abstract

Background

Precision medicine is a medical approach that takes into account individual genetic variability and often requires Next Generation Sequencing data in order to predict new treatments. Here we present GMIEC, Genomic Modules Identification et Characterization for genomics medicine, an application that is able to identify specific drugs at the level of single patient integrating multi-omics data such as RNA-sequencing, copy-number variation, methylation, Chromatin Immuno-Precipitation and Exome/Whole Genome sequencing. It is also possible to include clinical data related to each patient. GMIEC has been developed as a web-based R-Shiny platform and gives as output a table easy to use and explore.

Results

We present GMIEC, a Shiny application for genomics medicine. The tool allows the users the integration of two or more multiple omics datasets (e.g. gene-expression, copy-number), at sample level, to identify groups of genes that share common genomic and corresponding drugs. We demonstrate the characteristics of our application by using it to analyze a prostate cancer data set.

Conclusions

GMIEC provides a simple interface for genomics medicine. GMIEC was develop with Shiny to provide an application that does not require advanced programming skills. GMIEC consists of three sub-application for the analysis (GMIEC-AN), the visualization (GMIEC-VIS) and the exploration of results (GMIEC-RES). GMIEC is an open source software and is available at https://github.com/guidmt/GMIEC-shiny

microRNA-26a Directly Targeting MMP14 and MMP16 Inhibits the Cancer Cell Proliferation, Migration and Invasion in Cutaneous Squamous Cell Carcinoma

Purpose

To investigate the specific effect and underlying mechanism of microRNA-26a-5p (miR-26a) in cutaneous squamous cell carcinoma (CSCC).

Methods

miR-26a and MMP14/16 mRNA expression were detected by qRT-PCR analysis. Functional experiments were used to detect the role of miR-26a on CSCC progression. Western blot was used for protein detection. Luciferase assay was used to detect miR-26a directly targeting MMP14 and MMP16. Xenograft nude mice model was used to determine the effect of miR-26a on tumorigenesis.

Results

miR-26a was decreased in CSCC tissues and cells. Forced miR-26a suppressed the progression of SCL-1 and A431 cells. Furthermore, miR-26a directly targeted MMP14 and MMP16 to inhibit their expression. Forced expression of MMP14 and MMP16 removed the miR-26a’s inhibitory effect on CSCC development. The in vivo tumor growth assay showed that miR-26a suppressed CSCC tumorigenesis by targeting MMP14 and MMP16.

Conclusion

Our study suggested miR-26a inhibits cancer cell proliferation, migration and invasion in CSCC by targeting MMP14 and MMP16. 

Identification and Analysis of p53-Regulated Enhancers in Hepatic Carcinoma

Enhancers can act as cis-regulatory elements to control transcriptional regulation by recruiting transcription factors (TFs) in a distance and orientation-independent manner. However, it is still unclear how p53 participates in the enhancer network as TF in hepatic carcinoma under the condition of DNA damage. A total of 14,286 active enhancers were identified through the integration of stable and unstable enhancer RNAs (eRNAs) captured by CAGE and GRO-seq, respectively. Furthermore, 218 p53-bound enhancers (Enh_p53) were identified by analyzing p53 ChIP-seq in HepG2 cells after DNA damage. The results showed that the enhancer expression and histone markers of enhancers (H3K4me1, H3K4me2, H3K4me3, H3K9ac, and H3K27ac) revealed significantly higher level on Enh_p53 than Enh_no−p53 which suggested that p53 participated in regulating enhancer activity and chromatin structure. By analyzing 124 TFs ChIP-seq from ENCODE, 93 TFs were found significantly enriched on Enh_p53 such as GATA4, YY1, and CTCF, indicating p53 may co-regulate enhancers with TFs participation. Moreover, significantly differentially expressed 438 miRNAs and 1,264 mRNAs were identified by analyzing small RNA-seq and RNA-seq, and 26 Enh_p53-miRNAs and 145 Enh_p53-mRNA interactions were identified by the integration of 3D genome data and genomic distance. The functional enrichment analysis showed that these miRNA targets and mRNAs were significantly involved in tumor biological processes and signaling pathways such as DNA replication, p53 signaling pathway, hepatitis B, focal adhesion, etc. The above results indicated that p53 participated in regulating enhancer network in hepatic carcinoma and Enh_p53 exhibited significantly different characteristics with Enh_no−p53.

Integrative Network Fusion: A Multi-Omics Approach in Molecular Profiling

Recent technological advances and international efforts, such as The Cancer Genome Atlas (TCGA), have made available several pan-cancer datasets encompassing multiple omics layers with detailed clinical information in large collection of samples. The need has thus arisen for the development of computational methods aimed at improving cancer subtyping and biomarker identification from multi-modal data. Here we apply the Integrative Network Fusion (INF) pipeline, which combines multiple omics layers exploiting Similarity Network Fusion (SNF) within a machine learning predictive framework. INF includes a feature ranking scheme (rSNF) on SNF-integrated features, used by a classifier over juxtaposed multi-omics features (juXT). In particular, we show instances of INF implementing Random Forest (RF) and linear Support Vector Machine (LSVM) as the classifier, and two baseline RF and LSVM models are also trained on juXT. A compact RF model, called rSNFi, trained on the intersection of top-ranked biomarkers from the two approaches juXT and rSNF is finally derived. All the classifiers are run in a 10x5-fold cross-validation schema to warrant reproducibility, following the guidelines for an unbiased Data Analysis Plan by the US FDA-led initiatives MAQC/SEQC. INF is demonstrated on four classification tasks on three multi-modal TCGA oncogenomics datasets. Gene expression, protein expression and copy number variants are used to predict estrogen receptor status (BRCA-ER, N = 381) and breast invasive carcinoma subtypes (BRCA-subtypes, N = 305), while gene expression, miRNA expression and methylation data is used as predictor layers for acute myeloid leukemia and renal clear cell carcinoma survival (AML-OS, N = 157; KIRC-OS, N = 181). In test, INF achieved similar Matthews Correlation Coefficient (MCC) values and 97% to 83% smaller feature sizes (FS), compared with juXT for BRCA-ER (MCC: 0.83 vs. 0.80; FS: 56 vs. 1801) and BRCA-subtypes (0.84 vs. 0.80; 302 vs. 1801), improving KIRC-OS performance (0.38 vs. 0.31; 111 vs. 2319). INF predictions are generally more accurate in test than one-dimensional omics models, with smaller signatures too, where transcriptomics consistently play the leading role. Overall, the INF framework effectively integrates multiple data levels in oncogenomics classification tasks, improving over the performance of single layers alone and naive juxtaposition, and provides compact signature sizes.

Integrated Multi-Omics Analyses in Oncology: A Review of Machine Learning Methods and Tools

In recent years, high-throughput sequencing technologies provide unprecedented opportunity to depict cancer samples at multiple molecular levels. The integration and analysis of these multi-omics datasets is a crucial and critical step to gain actionable knowledge in a precision medicine framework. This paper explores recent data-driven methodologies that have been developed and applied to respond major challenges of stratified medicine in oncology, including patients' phenotyping, biomarker discovery, and drug repurposing. We systematically retrieved peer-reviewed journals published from 2014 to 2019, select and thoroughly describe the tools presenting the most promising innovations regarding the integration of heterogeneous data, the machine learning methodologies that successfully tackled the complexity of multi-omics data, and the frameworks to deliver actionable results for clinical practice. The review is organized according to the applied methods: Deep learning, Network-based methods, Clustering, Features Extraction, and Transformation, Factorization. We provide an overview of the tools available in each methodological group and underline the relationship among the different categories. Our analysis revealed how multi-omics datasets could be exploited to drive precision oncology, but also current limitations in the development of multi-omics data integration.

Dissecting the Functional Mechanisms of Somatic Copy-Number Alterations Based on Dysregulated ceRNA Networks across Cancers

Somatic copy-number alterations (SCNAs) drive tumor growth and evolution. However, the functional roles of SCNAs across the genome are still poorly understood. We provide an integrative strategy to characterize the functional roles of driver SCNAs in cancers based on dysregulated competing endogenous RNA (ceRNA) networks. We identified 44 driver SCNAs in lower-grade glioma (LGG). The dysregulated patterns losing all correlation relationships dominated dysregulated ceRNA networks. Homozygous deletion of six genes in 9p21.3 characterized an LGG subtype with poor prognosis and contributed to the dysfunction of cancer-associated pathways in a complementary way. The pan-cancer analysis showed that different cancer types harbored different driver SCNAs through dysregulating the crosstalk with common ceRNAs. The same SCNAs destroyed their ceRNA networks through different miRNA-mediated ceRNA regulations in different cancers. Additionally, some SCNAs performed different functional mechanisms in different cancers, which added another layer of complexity to cancer heterogeneity. Compared with previous methods, our strategy could directly dissect functional roles of SCNAs from the view of ceRNA networks, which not only complemented the functions of protein-coding genes but also provided a new avenue to characterize the functions of noncoding RNAs. Also, our strategy could be applied to more types of cancers to identify pathogenic mechanism driven by the SCNAs.

Targeting lipid droplet lysophosphatidylcholine for cisplatin chemotherapy

This study aims to explore lipidic mechanism towards low‐density lipoprotein receptor (LDLR)‐mediated platinum chemotherapy resistance. By using the lipid profiling technology, LDLR knockdown was found to increase lysosomal lipids and decrease membranous lipid levels in EOC cells. LDLR knockdown also down‐regulated ether‐linked phosphatidylethanolamine (PE‐O, lysosomes or peroxisomes) and up‐regulated lysophosphatidylcholine [LPC, lipid droplet (LD)]. This implies that the manner of using Lands cycle (conversion of lysophospholipids) for LDs might affect cisplatin sensitivity. The bioinformatics analyses illustrated that LDLR‐related lipid entry into LD, rather than an endogenous lipid resource (eg Kennedy pathway), controls the EOC prognosis of platinum chemotherapy patients. Moreover, LDLR knockdown increased the number of platinum‐DNA adducts and reduced the LD platinum amount. By using a manufactured LPC‐liposome‐cisplatin (LLC) drug, the number of platinum‐DNA adducts increased significantly in LLC‐treated insensitive cells. Moreover, the cisplatin content in LDs increased upon LLC treatment. Furthermore, lipid profiles of 22 carcinoma cells with differential cisplatin sensitivity (9 sensitive vs 13 insensitive) were acquired. These profiles revealed low storage lipid levels in insensitive cells. This result recommends that LD lipidome might be a common pathway in multiple cancers for platinum sensitivity in EOC. Finally, LLC suppressed both cisplatin‐insensitive human carcinoma cell training and testing sets. Thus, LDLR‐platinum insensitivity can be due to a defective Lands cycle that hinders LPC production in LDs. Using lipidome assessment with the newly formulated LLC can be a promising cancer chemotherapy method.

Analyzing the prognostic value of DKK1 expression in human cancers based on bioinformatics

Background

The Dickkopf1 (DKK1) gene encodes a protein that belongs to the Dickkopf family. The protein can inhibit the Wnt signaling pathway which plays a key role in the carcinogenesis and progression of various types of cancers. Based on this, we hypothesized that the differential expression of DKK1 may figure significantly in cancers by regulating Wnt signaling pathway transduction. In this study, we conducted bioinformatics analysis to evaluate the prognostic and therapeutic value of DKK1 expression level in human cancers.

Methods

The expression level was analyzed by using the Oncomine database and Gene Expression Profiling Interactive Analysis tool. The analysis of prognosis was conducted by using the UALCAN, Gene Expression Profiling Interactive Analysis (GEPIA), and DriverDBv3 databases. We also investigated using DKK1 promoter methylation to define cancer types through the UALCAN database. Meanwhile, the related functional networks of DKK1 were analyzed by using the GeneMANIA interactive tool and Cytoscape software. Furthermore, Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway analysis was conducted using the Metascape online website, and we used the cBioPotartal database to explored DKK1 expression, aberrant information, and the co-expression genes in the subgroups of lung cancer. Finally, we performed the overall survival (OS) meta-analysis of the DKK1 expression in lung squamous cell carcinoma (LUSC) via the Lung Cancer Explorer (LCE).

Results

DKK1 was differentially expressed in different types of human cancers. DKK1 was overexpressed in human cancers including head and neck squamous cell carcinoma (HNSC), LUSC, and pancreatic adenocarcinoma (PAAD). Overexpression of DKK1 indicated adverse OS in bladder urothelial carcinoma (BLCA), HNSC, and PADD, but no difference in OS was found between the LUSC and healthy groups. The high expression of DKK1 was also associated with shorter disease-free survival (DFS) in HNSC, LUSC, and PAAD. Gene regulation network analysis indicated that DKK1 was mainly involved in Wnt signaling pathways and several other signaling pathways.

Conclusions

Our findings showed that DKK1 is significantly expressed in various cancers and could be a biomarker for targeted therapy and a predictor for prognosis of these specific cancers. The bioinformatics analysis revealed a significant overexpression of DKK1 in HNSC, LUSC, and PAAD, with DKK1 overexpression being associated with adverse outcome in these patients, but how DKK1 expression levels relate to hematological malignancies and prognosis is still unclear. These new insights into the function of DKK1 may provide a basis for new targeted drug therapy and an avenue for further investigation into the mechanisms underlying carcinogenesis of DKK1 in different cancer types.

Expression and prognostic potential of GPX1 in human cancers based on data mining

BACKGROUND

Glutathione peroxidase-1 (GPX1) is a member of the GPX family, which considered an enzyme that interacts with oxidative stress. GPX1 differential expression is closely correlated with carcinogenesis and disease progression. In this study, we used bioinformatics analysis to investigate GPX1 expression level and explore the prognostic information in different human cancers.

METHODS

Expression was analyzed via the Oncomine database and Gene Expression Profiling Interactive Analysis tool, and potential prognostic analysis was evaluated using the UALCAN, GEPIA, and DriverDBv3 databases. Then, the UALCAN database was used to find the promoter methylation of GPX1 in defied cancer types. While GPX1 related functional networks were found within the GeneMANIA interactive tool and Cytoscape software. Moreover, Metascape online website was used to analyze Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathways.

RESULTS

We found that GPX1 was commonly overexpressed in most human cancers. High expression of GPX1 could lead to poor outcomes in Brain Lower Grade Glioma, while GPX1 over expression was correlated with better prognosis in Kidney renal papillary cell carcinoma (KIPP). High GPX1 expression was marginally associated with poor prognosis in acute myeloid leukemia (AML). Gene regulation network suggested that GPX1 mainly involved in pathways including the glutathione metabolism, ferroptosis, TP53 regulates metabolic genes, reactive oxygen species (ROS) metabolic process, and several other signaling pathways.

CONCLUSIONS

Our findings revealed that GPX1 showed significant expression differences among cancers and served as a prognostic biomarker for defined cancer types. The data mining effectively revealed useful information about GPX1 expression, prognostic values, and potential functional networks in cancers, thus providing researchers with an available way to further explore the mechanism underlying carcinogenesis of genes of interest in different cancers.

Identification of Prognostic Dosage-Sensitive Genes in Colorectal Cancer Based on Multi-Omics

Several studies have already identified the prognostic markers in colorectal cancer (CRC) based on somatic copy number alteration (SCNA). However, very little information is available regarding their value as a prognostic marker. Gene dosage effect is one important mechanism of copy number and dosage-sensitive genes are more likely to behave like driver genes. In this work, we propose a new pipeline to identify the dosage-sensitive prognostic genes in CRC. The RNAseq data, the somatic copy number of CRC from TCGA were assayed to screen out the SCNAs. Wilcoxon rank-sum test was used to identify the differentially expressed genes in alteration samples with |SCNA| > 0.3. Cox-regression was used to find the candidate prognostic genes. An iterative algorithm was built to identify the stable prognostic genes. Finally, the Pearson correlation coefficient was calculated between gene expression and SCNA as the dosage effect score. The cell line data from CCLE was used to test the consistency of the dosage effect. The differential co-expression network was built to discover their function in CRC. A total of six amplified genes (NDUFB4, WDR5B, IQCB1, KPNA1, GTF2E1, and SEC22A) were found to be associated with poor prognosis. They demonstrate a stable prognostic classification in more than 50% threshold of SCNA. The average dosage effect score was 0.5918 ± 0.066, 0.5978 ± 0.082 in TCGA and CCLE, respectively. They also show great stability in different data sets. In the differential co-expression network, these six genes have the top degree and are connected to the driver and tumor suppressor genes. Function enrichment analysis revealed that gene NDUFB4 and GTF2E1 affect cancer-related functions such as transmembrane transport and transformation factors. In conclusion, the pipeline for identifying the prognostic dosage-sensitive genes in CRC was proved to be stable and reliable.