Tumor radiogenomics in gliomas with Bayesian layered variable selection

We propose a statistical framework to analyze radiological magnetic resonance imaging (MRI) and genomic data to identify the underlying radiogenomic associations in lower grade gliomas (LGG). We devise a novel imaging phenotype by dividing the tumor region into concentric spherical layers that mimics the tumor evolution process. MRI data within each layer is represented by voxel-intensity-based probability density functions which capture the complete information about tumor heterogeneity. Under a Riemannian-geometric framework these densities are mapped to a vector of principal component scores which act as imaging phenotypes. Subsequently, we build Bayesian variable selection models for each layer with the imaging phenotypes as the response and the genomic markers as predictors. Our novel hierarchical prior formulation incorporates the interior-to-exterior structure of the layers, and the correlation between the genomic markers. We employ a computationally-efficient Expectation-Maximization-based strategy for estimation. Simulation studies demonstrate the superior performance of our approach compared to other approaches. With a focus on the cancer driver genes in LGG, we discuss some biologically relevant findings. Genes implicated with survival and oncogenesis are identified as being associated with the spherical layers, which could potentially serve as early-stage diagnostic markers for disease monitoring, prior to routine invasive approaches. We provide a R package that can be used to deploy our framework to identify radiogenomic associations.

Methylation-directed regulatory networks determine enhancing and silencing of mutation disease driver genes and explain inter-patient expression variation

BACKGROUND

Common diseases manifest differentially between patients, but the genetic origin of this variation remains unclear. To explore possible involvement of gene transcriptional-variation, we produce a DNA methylation-oriented, driver-gene-wide dataset of regulatory elements in human glioblastomas and study their effect on inter-patient gene expression variation.

RESULTS

In 175 of 177 analyzed gene regulatory domains, transcriptional enhancers and silencers are intermixed. Under experimental conditions, DNA methylation induces enhancers to alter their enhancing effects or convert into silencers, while silencers are affected inversely. High-resolution mapping of the association between DNA methylation and gene expression in intact genomes reveals methylation-related regulatory units (average size = 915.1 base-pairs). Upon increased methylation of these units, their target-genes either increased or decreased in expression. Gene-enhancing and silencing units constitute cis-regulatory networks of genes. Mathematical modeling of the networks highlights indicative methylation sites, which signified the effect of key regulatory units, and add up to make the overall transcriptional effect of the network. Methylation variation in these sites effectively describe inter-patient expression variation and, compared with DNA sequence-alterations, appears as a major contributor of gene-expression variation among glioblastoma patients.

CONCLUSIONS

We describe complex cis-regulatory networks, which determine gene expression by summing the effects of positive and negative transcriptional inputs. In these networks, DNA methylation induces both enhancing and silencing effects, depending on the context. The revealed mechanism sheds light on the regulatory role of DNA methylation, explains inter-individual gene-expression variation, and opens the way for monitoring the driving forces behind deferential courses of cancer and other diseases.

Mechanistic insights into the interactions between cancer drivers and the tumour immune microenvironment

BACKGROUND

The crosstalk between cancer and the tumour immune microenvironment (TIME) has attracted significant interest in the latest years because of its impact on cancer evolution and response to treatment. Despite this, cancer-specific tumour-TIME interactions and their mechanistic insights are still poorly understood.

METHODS

Here, we compute the significant interactions occurring between cancer-specific genetic drivers and five anti- and pro-tumour TIME features in 32 cancer types using Lasso regularised ordinal regression. Focusing on head and neck squamous cancer (HNSC), we rebuild the functional networks linking specific TIME driver alterations to the TIME state they associate with.

RESULTS

The 477 TIME drivers that we identify are multifunctional genes whose alterations are selected early in cancer evolution and recur across and within cancer types. Tumour suppressors and oncogenes have an opposite effect on the TIME and the overall anti-tumour TIME driver burden is predictive of response to immunotherapy. TIME driver alterations predict the immune profiles of HNSC molecular subtypes, and perturbations in keratinization, apoptosis and interferon signalling underpin specific driver-TIME interactions.

CONCLUSIONS

Overall, our study delivers a comprehensive resource of TIME drivers, gives mechanistic insights into their immune-regulatory role, and provides an additional framework for patient prioritisation to immunotherapy. The full list of TIME drivers and associated properties are available at http://www.network-cancer-genes.org .

The adaptive evolution of cancer driver genes

BACKGROUND

Cancer is a life-threatening disease in humans; yet, cancer genes are frequently reported to be under positive selection. This suggests an evolutionary-genetic paradox in which cancer evolves as a secondary product of selection in human beings. However, systematic investigation of the evolution of cancer driver genes is sparse.

RESULTS

Using comparative genomics analysis, population genetics analysis and computational molecular evolutionary analysis, the evolution of 568 cancer driver genes of 66 cancer types were evaluated at two levels, selection on the early evolution of humans (long timescale selection in the human lineage during primate evolution, i.e., millions of years), and recent selection in modern human populations (~ 100,000 years). Results showed that eight cancer genes covering 11 cancer types were under positive selection in the human lineage (long timescale selection). And 35 cancer genes covering 47 cancer types were under positive selection in modern human populations (recent selection). Moreover, SNPs associated with thyroid cancer in three thyroid cancer driver genes (CUX1, HERC2 and RGPD3) were under positive selection in East Asian and European populations, consistent with the high incidence of thyroid cancer in these populations.

CONCLUSIONS

These findings suggest that cancer can be evolved, in part, as a by-product of adaptive changes in humans. Different SNPs at the same locus can be under different selection pressures in different populations, and thus should be under consideration during precision medicine, especially for targeted medicine in specific populations.

Identifying cancer driver genes based on multi-view heterogeneous graph convolutional network and self-attention mechanism

BACKGROUND

Correctly identifying the driver genes that promote cell growth can significantly assist drug design, cancer diagnosis and treatment. The recent large-scale cancer genomics projects have revealed multi-omics data from thousands of cancer patients, which requires to design effective models to unlock the hidden knowledge within the valuable data and discover cancer drivers contributing to tumorigenesis.

RESULTS

In this work, we propose a graph convolution network-based method called MRNGCN that integrates multiple gene relationship networks to identify cancer driver genes. First, we constructed three gene relationship networks, including the gene-gene, gene-outlying gene and gene-miRNA networks. Then, genes learnt feature presentations from the three networks through three sharing-parameter heterogeneous graph convolution network (HGCN) models with the self-attention mechanism. After that, these gene features pass a convolution layer to generate fused features. Finally, we utilized the fused features and the original feature to optimize the model by minimizing the node and link prediction losses. Meanwhile, we combined the fused features, the original features and the three features learned from every network through a logistic regression model to predict cancer driver genes.

CONCLUSIONS

We applied the MRNGCN to predict pan-cancer and cancer type-specific driver genes. Experimental results show that our model performs well in terms of the area under the ROC curve (AUC) and the area under the precision-recall curve (AUPRC) compared to state-of-the-art methods. Ablation experimental results show that our model successfully improved the cancer driver identification by integrating multiple gene relationship networks.

DriverRWH: discovering cancer driver genes by random walk on a gene mutation hypergraph

Background

Recent advances in next-generation sequencing technologies have helped investigators generate massive amounts of cancer genomic data. A critical challenge in cancer genomics is identification of a few cancer driver genes whose mutations cause tumor growth. However, the majority of existing computational approaches underuse the co-occurrence mutation information of the individuals, which are deemed to be important in tumorigenesis and tumor progression, resulting in high rate of false positive.

Results

To make full use of co-mutation information, we present a random walk algorithm referred to as DriverRWH on a weighted gene mutation hypergraph model, using somatic mutation data and molecular interaction network data to prioritize candidate driver genes. Applied to tumor samples of different cancer types from The Cancer Genome Atlas, DriverRWH shows significantly better performance than state-of-art prioritization methods in terms of the area under the curve scores and the cumulative number of known driver genes recovered in top-ranked candidate genes. Besides, DriverRWH discovers several potential drivers, which are enriched in cancer-related pathways. DriverRWH recovers approximately 50% known driver genes in the top 30 ranked candidate genes for more than half of the cancer types. In addition, DriverRWH is also highly robust to perturbations in the mutation data and gene functional network data.

Conclusion

DriverRWH is effective among various cancer types in prioritizes cancer driver genes and provides considerable improvement over other tools with a better balance of precision and sensitivity. It can be a useful tool for detecting potential driver genes and facilitate targeted cancer therapies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04788-7.

gcMECM: graph clustering of mutual exclusivity of cancer mutations

Background

Next-generation sequencing platforms allow us to sequence millions of small fragments of DNA simultaneously, revolutionizing cancer research. Sequence analysis has revealed that cancer driver genes operate across multiple intricate pathways and networks with mutations often occurring in a mutually exclusive pattern. Currently, low-frequency mutations are understudied as cancer-relevant genes, especially in the context of networks.

Results

Here we describe a tool, gcMECM, that enables us to visualize the functionality of mutually exclusive genes in the subnetworks derived from mutation associations, gene–gene interactions, and graph clustering. These subnetworks have revealed crucial biological components in the canonical pathway, especially those mutated at low frequency. Examining the subnetwork, and not just the impact of a single gene, significantly increases the statistical power of clinical analysis and enables us to build models to better predict how and why cancer develops.

Conclusions

gcMECM uses a computationally efficient and scalable algorithm to identify subnetworks in a canonical pathway with mutually exclusive mutation patterns and distinct biological functions.

Comprehensive analysis of the cancer driver genes in breast cancer demonstrates their roles in cancer prognosis and tumor microenvironment

Background

Breast cancer is the most common malignancy in women. Cancer driver gene-mediated alterations in the tumor microenvironment are critical factors affecting the biological behavior of breast cancer. The purpose of this study was to identify the expression characteristics and prognostic value of cancer driver genes in breast cancer.

Methods

The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets are used as the training and test sets. Classified according to cancer and paracancerous tissues, we identified differentially expressed cancer driver genes. We further screened prognosis-associated genes, and candidate genes were submitted for the construction of a risk signature. Functional enrichment analysis and transcriptional regulatory networks were performed to search for possible mechanisms by which cancer driver genes affect breast cancer prognosis.

Results

We identified more than 200 differentially expressed driver genes and 27 prognosis-related genes. High-risk group patients had a lower survival rate compared to the low-risk group (P<0.05), and risk signature showed high specificity and sensitivity in predicting the patient prognosis (AUC 0.790). Multivariate regression analysis suggested that risk scores can independently predict patient prognosis. Further, we found differences in PD-1 expression, immune score, and stromal score among different risk groups.

Conclusion

Our study confirms the critical prognosis role of cancer driver genes in breast cancer. The cancer driver gene risk signature may provide a novel biomarker for clinical treatment strategy and survival prediction of breast cancer.

NOTCH1 mutations as prognostic marker in oral squamous cell carcinoma

Oral squamous cell carcinoma (OSCC) is the most common malignancy of the oral cavity with poor prognosis. The dysregulation of Notch signaling pathway has been implicated in the OSCC tumorigenesis. However, the clinical implication of NOTCH1 mutation status in OSCC remains unelucidated. We extracted the NOTCH1 gene mutations from a whole exome sequencing dataset of 168 frozen OSCC tumor specimens and validated these NOTCH1 gene mutations by Sanger sequencing. We also assessed these NOTCH1 gene mutations and its pathological significance in our OSCC tumor tissues using immunohistochemistry. Univariate and multivariate analyses were also used to determine whether the association between NOTCH1 mutation status and prognostic factors was independent of other parameters. In this study, we have identified 44 (26.19 %) NOTCH1 gene mutations from a whole-exome sequencing of 168 OSCC formalin-fixed, paraffin-embedded (FFPE) tissue specimen. These mutations distributed in different NOTCH1 function domains, including the EGF-like repeats region, negative regulatory region, and Ankyrin repeats region. The immunohistochemical staining analysis revealed that NOTCH1 expression was increased in oral cancer tissues. In addition, of the 43 OSCC tumors with NOTCH1 mutations, we observed that the majority were negative for NOTCH1 intracellular domain 1 (NICD1) staining (76.74 %), and 10 tumors were positive for NICD1 staining (23.26 %). In conclusion, our study suggested that NOTCH1 expression is associated with the progression of OSCC. We also demonstrated that presence of a mutated NOTCH1 gene will help prognostic stratification in OSCC when combined with other clinicopathologic parameters.

Germline and somatic mutation profile in Cancer patients revealed by a medium-sized pan-Cancer panel

Gene mutation detection and the resulted precision-medicine therapy is transforming clinical practice. Here, we report the use of a custom-developed, medium-sized, pan-cancer probe panel for the detection of somatic and germline mutations. We used a hybridization capture-based NGS assay for targeted deep sequencing of all exons and selected introns of 181 key cancer driver genes, covering both inherited risks and somatic mutations. We performed paired-variant calling on tumor samples and their matched normal samples. We processed clinical patient samples of formalin-fixed, paraffin embedded tumors (FFPE samples) and cell-free peripheral blood (cfDNA samples). We found germline mutations of inherited cancer risk at 9%; and discovered a novel germline mutation in BRCA1. Somatic mutation rate in driver genes is at 73.1%, much higher than previously reported. On recommending precision-medicine therapeutics, we achieved 91.6% for patients with FFPE samples.

Pan-cancer detection of driver genes at the single-patient resolution

BACKGROUND

Identifying the complete repertoire of genes that drive cancer in individual patients is crucial for precision oncology. Most established methods identify driver genes that are recurrently altered across patient cohorts. However, mapping these genes back to patients leaves a sizeable fraction with few or no drivers, hindering our understanding of cancer mechanisms and limiting the choice of therapeutic interventions.

RESULTS

We present sysSVM2, a machine learning software that integrates cancer genetic alterations with gene systems-level properties to predict drivers in individual patients. Using simulated pan-cancer data, we optimise sysSVM2 for application to any cancer type. We benchmark its performance on real cancer data and validate its applicability to a rare cancer type with few known driver genes. We show that drivers predicted by sysSVM2 have a low false-positive rate, are stable and disrupt well-known cancer-related pathways.

CONCLUSIONS

sysSVM2 can be used to identify driver alterations in patients lacking sufficient canonical drivers or belonging to rare cancer types for which assembling a large enough cohort is challenging, furthering the goals of precision oncology. As resources for the community, we provide the code to implement sysSVM2 and the pre-trained models in all TCGA cancer types ( https://github.com/ciccalab/sysSVM2 ).

Not all cancers are created equal: Tissue specificity in cancer genes and pathways

Tumors arise through waves of genetic alterations and clonal expansion that allow tumor cells to acquire cancer hallmarks, such as genome instability and immune evasion. Recent genomic analyses showed that the vast majority of cancer driver genes are mutated in a tissue-dependent manner, that is, are altered in some cancers but not others. Often the tumor type also affects the likelihood of therapy response. What is the origin of tissue specificity in cancer? Recent studies suggest that both cell-intrinsic and cell-extrinsic factors play a role. On one hand, cell type-specific wiring of the cell signaling network determines the outcome of cancer driver gene mutations. On the other hand, the tumor cells' exposure to tissue-specific microenvironments (e.g. immune cells) also contributes to shape the tissue specificity of driver genes and of therapy response. In the future, a more complete understanding of tissue specificity in cancer may inform methods to better predict and improve therapeutic outcomes.

MED12 exon 2 mutation is uncommon in intravenous leiomyomatosis: clinicopathologic features and molecular study

Intravenous leiomyomatosis (IVL) is a rare neoplasm that is characterized by smooth muscle cell proliferation within venous vessels. The aim of this study is to investigate the clinicopathological features, immunophenotypes, and MED12 gene mutations in IVL. Nine cases of IVL from the Affiliated Hospital of Qingdao University were collected, and the clinicopathological features were reviewed. The immunohistochemical expressions of p16, phosphatase and tensin homolog deleted on chromosome 10 (PTEN), alpha thalassemia/mental retardation syndrome X-linked (ATRX), retinoblastoma 1 (RB1), fumarate hydratase (FH), and p53, were evaluated. The mutation status of MED12 gene exon 2 was detected by Sanger sequencing. All the 9 patients were women ranging from 32 to 58 years, and uterine leiomyomas were identified in 5 patients. Immunohistochemical staining showed that all IVL and leiomyoma samples were positive for estrogen receptor and progesterone receptor, but negative for CD34. IVL displayed similar immunostaining patterns with their uterine counterparts with focal p16 immunostaining. FH, PTEN, ATRX, and RB1 were variably positive, and p53 and Ki-67 positive rates were less than 5% in all cases. Two novel genetic variations at MED12 exon 2, a synonymous mutation c.141C>T (p.Asn47=), and an in-frame deletion mutation c.133_147del15 (p.Phe45_Pro49del) were identified in two IVL cases. One missense mutation c.131G>A (p.Gly44Asp) was identified in one uterine leiomyoma. The remaining 11 tumor samples (7 IVL cases and 4 uterine leiomyomas) showed no mutations at MED12 exon 2. Our results showed two novel MED12 mutations in IVL. The MED12 mutations are different between IVL and uterine leiomyoma. These findings indicate that IVL is a unique entity and different from uterine leiomyoma.

Using Drosophila Models and Tools to Understand the Mechanisms of Novel Human Cancer Driver Gene Function

The formation, overgrowth and metastasis of tumors comprise a complex series of cellular and molecular events resulting from the combined effects of a variety of aberrant signaling pathways, mutations, and epigenetic alterations. Modeling this complexity in vivo requires multiple genes to be manipulated simultaneously, which is technically challenging. Here, we analyze how Drosophila research can further contribute to identifying pathways and elucidating mechanisms underlying novel cancer driver (risk) genes associated with tumor growth and metastasis in humans.

C3: Consensus Cancer Driver Gene Caller

Next-generation sequencing has allowed identification of millions of somatic mutations in human cancer cells. A key challenge in interpreting cancer genomes is to distinguish drivers of cancer development among available genetic mutations. To address this issue, we present the first web-based application, consensus cancer driver gene caller (C³), to identify the consensus driver genes using six different complementary strategies, i.e., frequency-based, machine learning-based, functional bias-based, clustering-based, statistics model-based, and network-based strategies. This application allows users to specify customized operations when calling driver genes, and provides solid statistical evaluations and interpretable visualizations on the integration results. C³ is implemented in Python and is freely available for public use at http://drivergene.rwebox.com/c3.

Differential analysis between somatic mutation and germline variation profiles reveals cancer-related genes

A major aim of cancer genomics is to pinpoint which somatically mutated genes are involved in tumor initiation and progression. We introduce a new framework for uncovering cancer genes, differential mutation analysis, which compares the mutational profiles of genes across cancer genomes with their natural germline variation across healthy individuals. We present DiffMut, a fast and simple approach for differential mutational analysis, and demonstrate that it is more effective in discovering cancer genes than considerably more sophisticated approaches. We conclude that germline variation across healthy human genomes provides a powerful means for characterizing somatic mutation frequency and identifying cancer driver genes. DiffMut is available at https://github.com/Singh-Lab/Differential-Mutation-Analysis.

Rational design of cancer gene panels with OncoPaD

BACKGROUND

Profiling the somatic mutations of genes which may inform about tumor evolution, prognostics and treatment is becoming a standard tool in clinical oncology. Commercially available cancer gene panels rely on manually gathered cancer-related genes, in a "one-size-fits-many" solution. The design of new panels requires laborious search of literature and cancer genomics resources, with their performance on cohorts of patients difficult to estimate.

RESULTS

We present OncoPaD, to our knowledge the first tool aimed at the rational design of cancer gene panels. OncoPaD estimates the cost-effectiveness of the designed panel on a cohort of tumors and provides reports on the importance of individual mutations for tumorigenesis or therapy. With a friendly interface and intuitive input, OncoPaD suggests researchers relevant sets of genes to be included in the panel, because prior knowledge or analyses indicate that their mutations either drive tumorigenesis or function as biomarkers of drug response. OncoPaD also provides reports on the importance of individual mutations for tumorigenesis or therapy that support the interpretation of the results obtained with the designed panel. We demonstrate in silico that OncoPaD designed panels are more cost-effective-i.e. detect a maximum fraction of tumors in the cohort by sequencing a minimum quantity of DNA-than available panels.

CONCLUSIONS

With its unique features, OncoPaD will help clinicians and researchers design tailored next-generating sequencing (NGS) panels to detect circulating tumor DNA or biopsy specimens, thereby facilitating early and accurate detection of tumors, genomics informed therapeutic decisions, patient follow-up and timely identification of resistance mechanisms to targeted agents. OncoPaD may be accessed through http://www.intogen.org/oncopad.