Multi-omic data helps improve prediction of personalised tumor suppressors and oncogenes

The progression of tumorigenesis starts with a few mutational and structural driver events in the cell. Various cohort-based computational tools exist to identify driver genes but require a large number of samples to produce reliable results. Many studies use different methods to identify driver mutations/genes from mutations that have no impact on tumour progression; however, a small fraction of patients show no mutational events in any known driver genes. Current unsupervised methods map somatic and expression data onto a network to identify the perturbation in the network. Our method is the first machine learning model to classify genes as tumour suppressor gene (TSG), oncogene (OG) or neutral, thus assigning the functional impact of the gene in the patient. In this study, we develop a multi-omic approach, PIVOT (Personalised Identification of driVer OGs and TSGs), to train on experimentally or computationally validated mutational and structural driver events. Given the lack of any gold standards for the identification of personalised driver genes, we label the data using four strategies and, based on classification metrics, show gene-based labelling strategies perform best. We build different models using SNV, RNA, and multi-omic features to be used based on the data available. Our models trained on multi-omic data improved predictions compared to mutation and expression data, achieving an accuracy >0.99 for BRCA, LUAD and COAD datasets. We show network and expression-based features contribute the most to PIVOT. Our predictions on BRCA, COAD and LUAD cancer types reveal commonly altered genes such as TP53, and PIK3CA, which are predicted drivers for multiple cancer types. Along with known driver genes, our models also identify new driver genes such as PRKCA, SOX9 and PSMD4. Our multi-omic model labels both CNV and mutations with a more considerable contribution by CNV alterations. While predicting labels for genes mutated in multiple samples, we also label rare driver events occurring in as few as one sample. We also identify genes with dual roles within the same cancer type. Overall, PIVOT labels personalised driver genes as TSGs and OGs and also identifies rare driver genes. PIVOT is available at https://github.com/RamanLab/PIVOT.

m6A-TSHub: unveiling the context-specific m6A methylation and m6A-affecting mutations in 23 human tissues

As the most pervasive epigenetic marker present on mRNA and lncRNA, N6-methyladenosine (m6A) RNA methylation has been shown to participate in essential biological processes. Recent studies revealed the distinct patterns of m6A methylome across human tissues, and a major challenge remains in elucidating the tissue-specific presence and circuitry of m6A methylation. We present here a comprehensive online platform m6A-TSHub for unveiling the context-specific m6A methylation and genetic mutations that potentially regulate m6A epigenetic mark. m6A-TSHub consists of four core components, including (1) m6A-TSDB: a comprehensive database of 184,554 functionally annotated m6A sites derived from 23 human tissues and 499,369 m6A sites from 25 tumor conditions, respectively; (2) m6A-TSFinder: a web server for high-accuracy prediction of m6A methylation sites within a specific tissue from RNA sequences, which was constructed using multi-instance deep neural networks with gated attention; (3) m6A-TSVar: a web server for assessing the impact of genetic variants on tissue-specific m6A RNA modification; and (4) m6A-CAVar: a database of 587,983 TCGA cancer mutations (derived from 27 cancer types) that were predicted to affect m6A modifications in the primary tissue of cancers. The database should make a useful resource for studying the m6A methylome and genetic factor of epitranscriptome disturbance in a specific tissue (or cancer type). m6A-TSHub is accessible at: www.xjtlu.edu.cn/biologicalsciences/m6ats.

SIRT7 is a prognostic biomarker in kidney renal clear cell carcinoma that is correlated with immune cell infiltration

Background: SIRT7 has been shown to be expressed in many cancer types, including kidney renal clear cell carcinoma (KIRC), but its functional role in this oncogenic context remains to be firmly defined. This study was designed to explore correlations between SIRT7 and KIRC characteristics using the TCGA database. Methods: Relationships between SIRT7 expression and KIRC patient clinicopathological characteristics were assessed through Kruskal-Wallis tests, Wilcoxon signed-rank tests, and logistic regression analyses. Area under the ROC curve (AUC) values were used to assess the prognostic value of SIRT7 as a means of classifying KIRC patients. The functional role of SIRT7 in this cancer type was assessed through GO/KEGG enrichment analyses and immune cell infiltration analyses. Results: In KIRC patients, higher levels of SIRT7 expression were associated with Race, M stage, T stage (all P < 0.05). SIRT7 offered significant diagnostic value in ROC curve analyses (AUC = 0.912), and elevated SIRT7 levels were linked to worse patient overall survival (OS; P < 0.001). The expression of SIRT7 was independently related with KIRC patient OS (HR: 1.827; 95%CI: 1.346-2.481; P<0.001). In GO/KEGG analyses, SIRT7 was found to be associated with ubiquitin-mediated proteolysis and nucleotide excision repair. Higher SIRT7 expression was related to the enhanced infiltration of certain immune cells.Conclusions: Increased SIRT7 expression was associated with a worse KIRC patient prognosis, and immune infiltrates, suggesting it may offer value as a prognostic biomarker for this cancer type.

Prevalence and Spectrum of Second Primary Malignancies among People Living with HIV in the French Dat’AIDS Cohort

Background: We aimed to describe the prevalence and spectrum of second primary cancer (SPC) in HIV-positive cancer survivors. Methods: A multicenter retrospective study was performed using longitudinal data from the French Dat’AIDS cohort. Subjects who had developed at least two primary cancers were selected. The spectrum of SPCs was stratified by the first primary cancer type and by sex. Results: Among the 44,642 patients in the Dat’AIDS cohort, 4855 were diagnosed with cancer between 1 December 1983 and 31 December 2015, of whom 444 (9.1%) developed at least two primary cancers. The most common SPCs in men were non-Hodgkin lymphoma (NHL) (22.8%), skin carcinoma (10%) and Kaposi sarcoma (KS) (8.4%), and in women the most common SPCs were breast cancer (16%), skin carcinoma (9.3%) and NHL (8%). The pattern of SPCs differed according to first primary cancer and by sex: in men, NHL was the most common SPC after primary KS and KS was the most common SPC after primary NHL; while in women, breast cancer was the most common SPC after primary NHL and primary breast cancer. Conclusion: The frequency and pattern of subsequent cancers among HIV-positive cancer survivors differed according to the first primary cancer type and sex.

Noncoding RNAs and Deep Learning Neural Network Discriminate Multi-Cancer Types

Detecting cancers at early stages can dramatically reduce mortality rates. Therefore, practical cancer screening at the population level is needed. To develop a comprehensive detection system to classify multiple cancer types. We integrated an artificial intelligence deep learning neural network and noncoding RNA biomarkers selected from massive data. Our system can accurately detect cancer vs. healthy objects with 96.3% of AUC of ROC (Area Under Curve of a Receiver Operating Characteristic curve), and it surprisingly reaches 78.77% of AUC when validated by real-world raw data from a completely independent data set. Even validating with raw exosome data from blood, our system can reach 72% of AUC. Moreover, our system significantly outperforms conventional machine learning models, such as random forest. Intriguingly, with no more than six biomarkers, our approach can easily discriminate any individual cancer type vs. normal with 99% to 100% AUC. Furthermore, a comprehensive marker panel can simultaneously multi-classify common cancers with a stable 82.15% accuracy rate for heterogeneous cancerous tissues and conditions.: This detection system provides a promising practical framework for automatic cancer screening at population level. Key points: (1) We developed a practical cancer screening system, which is simple, accurate, affordable, and easy to operate. (2) Our system binarily classify cancers vs. normal with >96% AUC. (3) In total, 26 individual cancer types can be easily detected by our system with 99 to 100% AUC. (4) The system can detect multiple cancer types simultaneously with >82% accuracy.

Multi-cancer classification; an analysis of neural network complexity

Background: Cancer identification is generally framed as binary classification, normally discrimination of a control group from a single cancer group. However, such models lack any cancer-specific information, as they are only trained on one cancer type. The models fail to account for competing cancer risks. For example, an ostensibly healthy individual may have any number of different cancer types, and a tumor may originate from one of several primary sites. Pan-cancer evaluation requires a model trained on multiple cancer types, and controls, simultaneously, so that a physician can be directed to the correct area of the body for further testing. Methods: We introduce novel neural network models to address multi-cancer classification problems across several data types commonly applied in cancer prediction, including circulating miRNA expression, protein, and mRNA. In particular, we present an analysis of neural network depth and complexity, and investigate how this relates to classification performance. Comparisons of our models with state-of-the-art neural networks from the literature are also presented. Results: Our analysis evidences that shallow, feed-forward neural net architectures offer greater performance when compared to more complex deep feed-forward, Convolutional Neural Network (CNN), and Graph CNN (GCNN) architectures considered in the literature. Conclusion: The results show that multiple cancers and controls can be classified accurately using the proposed models, across a range of expression technologies in cancer prediction. Impact: This study addresses the important problem of pan-cancer classification, which is often overlooked in the literature. The promising results highlight the urgency for further research.

Bioinformatics Analysis of the Potentially Functional circRNA-miRNA-mRNA Network in Breast Cancer

Objective: Breast cancer (BC) is a heterogeneous type of cancer that occurs as a result of distinct molecular alterations in breast tissue. Although there are many new developments in treatment and targeted therapy for BC in recent years, this cancer type is still the most common one among women with high morbidity and mortality. Therefore, new research is still needed for biomarker detection. Methods: GSE101124 and GSE182471 datasets were obtained from Gene Expression Omnibus (GEO) database to evaluate differentially expressed circular RNAs (circRNAs). The Cancer Genome Atlas (TCGA) and Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) databases were used to identify the significantly dysregulated microRNAs (miRNAs) and genes considering the Prediction Analysis of Microarray (PAM50) classification. The circRNA-miRNA-gene relationship was investigated using the Cancer Specific CircRNA (v2.0) (CSCD), miRDB, miRWalk and miRTarBase databases. The circRNA-miRNA-mRNA regulatory network was constructed using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) annotation. The protein-protein interaction network was constructed by the STRING 2021 database and visualized by the Cytoscape tool (v3.9.0). Then, raw miRNA data and genes were filtered using some selection criteria according to a specific expression level in PAM50 subgroups. A bottleneck method was utilized to obtain highly interacted hub genes using cytoHubba Cytoscape plugin. The overall survival (OS) and disease-free survival (DFS) analysis were performed for these hub genes, which are detected within the miRNA and circRNA axis in our study. Results: We identified three circRNAs, three miRNAs, and eighteen candidate target genes that may play an important role in BC. In addition, it has been determined that these molecules can be useful in the classification of BC, especially in determining the basal-like breast cancer (BLBC) subtype. Conclusions: We conclude that hsa_circ_0000515/ miR-486-5p/ SDC1 axis may be an important biomarker candidate in distinguishing patients in the BLBC group, especially according to the PAM50 classification of BC.

A Pan-Cancer Study of KMT2 Family as Therapeutic Targets in Cancer

Objective. Exome sequencing studies have shown that the histone-lysine N-methyltransferase 2 (KMT2) gene is one of the most commonly mutated genes in a range of human malignancies and is linked to some of the most common and deadly solid tumors. However, the connection between this gene family’s function and tumor type, immunological subtype, and molecular subtype dependency is still unknown. Methods. We examine the expression patterns of the histone-lysine N-methyltransferase 2 (KMT2) gene, as well as their relationship to patient survival. We also used a pan-cancer analysis to link their function to immunological subtypes, the tumor microenvironment, and treatment sensitivity. Results. Using the TCGA pan-cancer data, researchers looked at and examined KMT2 expression patterns and their links to patient survival and the tumor microenvironment in 33 cancer types. The expression of the KMT2 family changes significantly across and within cancer types, indicating significant inter- and intracancer heterogeneity. Patients’ overall survival was often linked to the expression of KMT2 family members. However, the direction of the link differed depending on the KMT2 isoform and cancer type studied. Notably, in all cancer types examined, nearly all KMT2 family members were substantially linked with overall survival in patients with renal clear cell carcinoma (KIRC). Furthermore, all KMT2 genes have a strong relationship with immune infiltrate subtypes, as well as varying degrees of stromal cell infiltration and tumor cell stemness. Finally, we discovered that higher expression of KMT2s, particularly KMT2F and KMT2G, was linked to greater chemotherapeutic sensitivity in several cell lines. Conclusions. The necessity to investigate each KMT2 member as a distinct entity inside each particular cancer type is highlighted by our comprehensive investigation of KMT2 gene expression and its relationship with immune infiltrates, tumor microenvironment, and cancer patient outcomes. Our research also confirmed the identification of KMT2 as a potential therapeutic target in cancer, but further laboratory testing is required.

Variable roles of interleukin-17F in different cancers

Background Interleukin (IL)-17 family is a group of six cytokines that plays a central role in inflammatory processes and participates in cancer progression. Interleukin-17A has been shown to have mainly a protumorigenic role, but the other members of the IL-17 family, including IL-17F, have received less attention. Methods We applied systematic review guidelines to study the role of IL-17F, protein and mRNA expression, polymorphisms, and functions, in cancer. We carried out a systematic search in PubMed, Ovid Medline, Scopus, and Cochrane libraries, yielding 79 articles that met the inclusion criteria. Results The findings indicated that IL-17F has both anti- and protumorigenic roles, which depend on cancer type and the molecular form and location of IL-17F. As an example, the presence of IL-17F protein in tumor tissue and patient serum has a protective role in oral and pancreatic cancers, whereas it is protumorigenic in prostate and bladder cancers. These effects are proposed to be based on multiple mechanisms, such as inhibition of angiogenesis, vasculogenic mimicry and cancer cell proliferation, migration and invasion, and aggravating the inflammatory process. No solid evidence emerged for the correlation between IL-17F polymorphisms and cancer incidence or patients’ prognosis. Conclusion IL-17F is a multifaceted cytokine. There is a clear demand for more well-designed studies of IL-17F to elucidate its molecular mechanisms in different types of cancer. The studies presented in this article examined a variety of different designs, study populations and primary/secondary outcomes, which unfortunately reduces the value of direct interstudy comparisons.

Sapienic Acid Metabolism Influences Membrane Plasticity and Protein Signaling in Breast Cancer Cell Lines

The importance of sapienic acid (6c-16:1), a monounsaturated fatty acid of the n-10 family formed from palmitic acid by delta-6 desaturase, and of its metabolism to 8c-18:1 and sebaleic acid (5c,8c-18:2) has been recently assessed in cancer. Data are lacking on the association between signaling cascades and exposure to sapienic acid comparing cell lines of the same cancer type. We used 50 μM sapienic acid supplementation, a non-toxic concentration, to cultivate MCF-7 and 2 triple-negative breast cancer cells (TNBC), MDA-MB-231 and BT-20. We followed up for three hours regarding membrane fatty acid remodeling by fatty acid-based membrane lipidome analysis and expression/phosphorylation of EGFR (epithelial growth factor receptor), mTOR (mammalian target of rapamycin) and AKT (protein kinase B) by Western blotting as an oncogenic signaling cascade. Results evidenced consistent differences among the three cell lines in the metabolism of n-10 fatty acids and signaling. Here, a new scenario is proposed for the role of sapienic acid: one based on changes in membrane composition and properties, and the other based on changes in expression/activation of growth factors and signaling cascades. This knowledge can indicate additional players and synergies in breast cancer cell metabolism, inspiring translational applications of tailored membrane lipid strategies to assist pharmacological interventions.