Predicting the recurrence of noncoding regulatory mutations in cancer

Glioblastoma is a fast and aggressively growing tumor in the brain and spinal cord. Mutation of amino acid residues in targets proteins, which are involved in glioblastoma, alters the structure and function and may lead to disease. In this study, we collected a set of 9386 disease-causing (drivers) mutations based on the recurrence in patient samples and experimentally annotated as pathogenic and 8728 as neutral (passenger) mutations. We observed that Arg is highly preferred at the mutant sites of drivers, whereas Met and Ile showed preferences in passengers. Inspecting neighboring residues at the mutant sites revealed that the motifs YP, CP and GRH, are preferred in drivers, whereas SI, IQ and TVI are dominant in neutral. In addition, we have computed other sequence-based features such as conservation scores, Position Specific Scoring Matrices (PSSM) and physicochemical properties, and developed a machine learning-based method, GBMDriver (GlioBlastoma Multiforme Drivers), for distinguishing between driver and passenger mutations. Our method showed an accuracy and AUC of 73.59% and 0.82, respectively, on 10-fold cross-validation and 81.99% and 0.87 in a blind set of 1809 mutants. The tool is available at https://web.iitm.ac.in/bioinfo2/GBMDriver/index.html. We envisage that the present method is helpful to prioritize driver mutations in glioblastoma and assist in identifying therapeutic targets.

BackgroundCancer sequencing projects are now measuring somatic mutations in large numbers of cancer genomes. A key challenge in interpreting these data is to distinguish driver mutations, mutations important for cancer development, from passenger mutations that have accumulated in somatic cells but without functional consequences. A common approach to identify genes harboring driver mutations is a single gene test that identifies individual genes that are recurrently mutated in a significant number of cancer genomes. However, the power of this test is reduced by: (1) the necessity of estimating the background mutation rate (BMR) for each gene; (2) the mutational heterogeneity in most cancers meaning that groups of genes (e.g. pathways), rather than single genes, are the primary target of mutations.ResultsWe investigate the problem of discovering driver pathways, groups of genes containing driver mutations, directly from cancer mutation data and without prior knowledge of pathways or other interactions between genes. We introduce two generative models of somatic mutations in cancer and study the algorithmic complexity of discovering driver pathways in both models. We show that a single gene test for driver genes is highly sensitive to the estimate of the BMR. In contrast, we show that an algorithmic approach that maximizes a straightforward measure of the mutational properties of a driver pathway successfully discovers these groups of genes without an estimate of the BMR. Moreover, this approach is also successful in the case when the observed frequencies of passenger and driver mutations are indistinguishable, a situation where single gene tests fail.ConclusionsAccurate estimation of the BMR is a challenging task. Thus, methods that do not require an estimate of the BMR, such as the ones we provide here, can give increased power for the discovery of driver genes.

Cancer sequencing projects are now measuring somatic mutations in large numbers of cancer genomes. A key challenge in interpreting these data is to distinguish driver mutations, mutations important for cancer development, from passenger mutations that have accumulated in somatic cells but without functional consequences. A common approach to identify genes harboring driver mutations is a single gene test that identifies individual genes that are mutated in a significant number of cancer genomes. However, the power of this test is reduced by the mutational heterogeneity in most cancer genomes and by the necessity of estimating the background mutation rate (BMR). We investigate the problem of discovering driver pathways, groups of genes containing driver mutations, directly from cancer mutation data and without prior knowledge of pathways or other interactions between genes. We introduce two generative models of somatic mutations in cancer and study the algorithmic complexity of discovering driver pathways in both models. We show that a single gene test for driver genes is highly sensitive to the estimate of the BMR. In contrast, we show that an algorithmic approach that maximizes a straightforward measure of the mutational properties of a driver pathway successfully discovers these groups of genes without an estimate of the BMR. Moreover, this approach is also successful in the case when the observed frequencies of passenger and driver mutations are indistinguishable, a situation where single gene tests fail.KeywordsMarkov Chain Monte CarloSomatic MutationGreedy AlgorithmDriver MutationDriver GeneThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

ObjectiveTo explore potential functional biomarkers in diabetes mellitus (DM) by utilizing gene pathway cross-talk.MethodsFirstly, potential disrupted pathways that were enriched by differentially expressed genes (DEGs) were identified based on biological pathways downloaded from the Ingenuity Pathways Analysis (IPA) database. In addition, we quantified the pathway crosstalk for each pair of pathways based on Discriminating Score (DS). Random forest (RF) classification was then employed to find the top 10 pairs of pathways with a high area under the curve (AUC) value between DM samples versus normal samples based on 10-fold cross-validation. Finally, a Monte Carlo Cross-Validation was applied to demonstrate the identified pairs of pathways by a mutual information analysis.ResultsA total of 247 DEGs in normal and disease samples were identified. Based on the F-test, 50 disrupted pathways were obtained with false discovery rate (FDR) < 0.01. Simultaneously, after calculating the DS, the top 10 pairs of pathways were selected based on a higher AUC value as measured by RF classification. From the Monte Carlo Cross-Validation, we considered the top 10 pairs of pathways with higher AUC values ranked for all 50 bootstraps as the most frequently detected ones.ConclusionThe pairs of pathways identified in our study might be key regulators in DM.

Development of machine learning solutions for prediction of functional and clinical significance of cancer driver genes and mutations are paramount in modern biomedical research and have gained a significant momentum in a recent decade. In this work, we integrate different machine learning approaches, including tree based methods, random forest and gradient boosted tree (GBT) classifiers along with deep convolutional neural networks (CNN) for prediction of cancer driver mutations in the genomic datasets. The feasibility of CNN in using raw nucleotide sequences for classification of cancer driver mutations was initially explored by employing label encoding, one hot encoding, and embedding to preprocess the DNA information. These classifiers were benchmarked against their tree-based alternatives in order to evaluate the performance on a relative scale. We then integrated DNA-based scores generated by CNN with various categories of conservational, evolutionary and functional features into a generalized random forest classifier. The results of this study have demonstrated that CNN can learn high level features from genomic information that are complementary to the ensemble-based predictors often employed for classification of cancer mutations. By combining deep learning-generated score with only two main ensemble-based functional features, we can achieve a superior performance of various machine learning classifiers. Our findings have also suggested that synergy of nucleotide-based deep learning scores and integrated metrics derived from protein sequence conservation scores can allow for robust classification of cancer driver mutations with a limited number of highly informative features. Machine learning predictions are leveraged in molecular simulations, protein stability, and network-based analysis of cancer mutations in the protein kinase genes to obtain insights about molecular signatures of driver mutations and enhance the interpretability of cancer-specific classification models.

I come from a family in which there have been no scientists or doctors. I was interested, however, in biology at school and started my scientific career by training in medicine at Oxford University and Guys Hospital, London. Practising as a doctor reinforced my curiosity about the biological processes underlying human disease. As a consequence, I pursued a clinical vocation in histopathology, a discipline that couples exposure to the sights and smells of the autopsy room with a daily journey into the often beautiful, sometimes ugly world of healthy and diseased human tissues under the microscope. After an introduction to general histopathology in Nick Wright's department at the Hammersmith Hospital, London, I completed my postgraduate medical training in neuropathology with Peter Lantos at the Maudsley Hospital, London.Peering at the nuclei of cancer cells under the microscope, for me it was a matter of fascination that hidden within them were the key events converting normal cells into cancer cells, and frustration because they were out of reach. Many of the tissue samples examined by pathologists are from cancers. The clonal theory of cancer development and the general role of DNA mutations in generating cancer cell clones had been established by 1986 when I was working as a junior pathologist. Indeed, the first mutated cancer gene, HRAS , had recently been identified through application of the, then new, technologies of recombinant DNA technology. Peering at the nuclei of cancer cells under the microscope, for me it was a matter of fascination that hidden within them were the key events converting normal cells into cancer cells, and frustration because they were out of reach. So, I took 3 years break from medicine to study for a PhD, learning the methods and thinking of molecular oncology in Colin Cooper's laboratory at the Institute …

Anastomotic leakage (AL) is one of the commonest and most serious complications after rectal cancer surgery. The previous analyses on predictors for AL included small-scale patients, and their prediction models performed unsatisfactorily. Clinical data of 5,220 patients who underwent anterior resection for rectal cancer were scrutinized to create a prediction model via random forest classifier. Additionally, data of 836 patients served as the test dataset. Patients diagnosed with AL within 6 months' follow-up were recorded. A total of 20 candidate factors were included. Receiver operating characteristic (ROC) curve was conducted to determine the clinical efficacy of our model, and compare the predictive performance of different models. The incidence of AL was 6.2% (326/5,220). A multivariate logistic regression analysis and the random forest classifier indicated that sex, distance of tumor from the anal verge, bowel stenosis or obstruction, preoperative hemoglobin, surgeon volume, diabetes, neoadjuvant chemoradiotherapy, and surgical approach were significantly associated with AL. After propensity score matching, the temporary stoma was not identified as a protective factor for AL (P=0.58). Contrastingly, the first year of performing laparoscopic surgery was a predictor (P=0.009). We created a predictive random forest classifier based on the above predictors that demonstrated satisfactory prediction efficacy. The area under the curve (AUC) showed that the random forest had higher efficiency (AUC =0.87) than the nomogram (AUC =0.724). Our findings suggest that eight factors may affect the incidence of AL. Our random forest classifier is an innovative and practical model to effectively predict AL, and could provide rational advice on whether to perform a temporary stoma, which might reduce the rate of stoma and avoid the ensuing complications.

Motivation: The number of missense mutations being identified in cancer genomes has greatly increased as a consequence of technological advances and the reduced cost of whole-genome/whole-exome sequencing methods. However, a high proportion of the amino acid substitutions detected in cancer genomes have little or no effect on tumour progression (passenger mutations). Therefore, accurate automated methods capable of discriminating between driver (cancer-promoting) and passenger mutations are becoming increasingly important. In our previous work, we developed the Functional Analysis through Hidden Markov Models (FATHMM) software and, using a model weighted for inherited disease mutations, observed improved performances over alternative computational prediction algorithms. Here, we describe an adaptation of our original algorithm that incorporates a cancer-specific model to potentiate the functional analysis of driver mutations.Results: The performance of our algorithm was evaluated using two separate benchmarks. In our analysis, we observed improved performances when distinguishing between driver mutations and other germ line variants (both disease-causing and putatively neutral mutations). In addition, when discriminating between somatic driver and passenger mutations, we observed performances comparable with the leading computational prediction algorithms: SPF-Cancer and TransFIC.Availability and implementation: A web-based implementation of our cancer-specific model, including a downloadable stand-alone package, is available at http://fathmm.biocompute.org.uk.Contact:fathmm@biocompute.org.ukSupplementary information:Supplementary data are available at Bioinformatics online.

Type-II topoisomerases resolve topological stress in DNA through controlled double-strand breaks. While TOP2A is a chemotherapy target in proliferating cells, the ubiquitously expressed TOP2B is a potential off-target. Here we explore roles of TOP2B in mutagenesis by generating DNA-binding maps of TOP2B, CTCF, and RAD21 in human cancer samples and analyzing these maps for driver mutations and mutational processes in 6500 whole cancer genomes. TOP2B-CTCF-RAD21 and TOP2B-RAD21 sites are enriched in somatic mutations and structural variants (SVs), especially at evolutionary conserved sites displaying high transcription and long-range chromatin interactions. TOP2B binding underlies SVs and hotspot mutations in cancer-driving genes such as TP53, MYC, FOXA1, and VHL, and many cis-regulatory elements. We show that the TOP2B-bound mutational hotspot at RMRP drives tumor initiation and growth in vivo. These data highlight TOP2B as a protector of the genome from topological challenges whose aberrant activity promotes driver and passenger mutations in cancer genomes. Citation Format: Jüri Reimand, 1 Christian A. Lee, 1 Robin H. Oh, 2 Zoe P. Klein, 1 Nina Adler, 1 Sana Akhtar Alvi, 3 Ellen Langille, 2 Elisa Pasini, 4 Kevin C. Cheng, 1 Diala Abd-Rabbo, 1 Huayun Hou, 3 Ricky Tsai, 2 Mamatha Bhat, 1 Daniel Schramek, 2 Michael D. Wilson, 3 Liis Uusküla-Reimand3. Topoisomerase IIb binding underlies frequently mutated elements in cancer genomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 2780.

Non-coding mutations can create splice sites, however the true extent of how such somatic non-coding mutations affect RNA splicing are largely unexplored. Here we use the MiSplice pipeline to analyze 783 cancer cases with WGS data and 9494 cases with WES data, discovering 562 non-coding mutations that lead to splicing alterations. Notably, most of these mutations create new exons. Introns associated with new exon creation are significantly larger than the genome-wide average intron size. We find that some mutation-induced splicing alterations are located in genes important in tumorigenesis (ATRX, BCOR, CDKN2B, MAP3K1, MAP3K4, MDM2, SMAD4, STK11, TP53 etc.), often leading to truncated proteins and affecting gene expression. The pattern emerging from these exon-creating mutations suggests that splice sites created by non-coding mutations interact with pre-existing potential splice sites that originally lacked a suitable splicing pair to induce new exon formation. Our study suggests the importance of investigating biological and clinical consequences of noncoding splice-inducing mutations that were previously neglected by conventional annotation pipelines. MiSplice will be useful for automatically annotating the splicing impact of coding and non-coding mutations in future large-scale analyses.

AimsWe aim to develop a pragmatic screening tool for heart failure at the general population level.Methods and resultsThis study was conducted within the Hamburg‐City‐Health‐Study, an ongoing, prospective, observational study enrolling randomly selected inhabitants of the city of Hamburg aged 45–75 years. Heart failure was diagnosed per current guidelines. Using only digital electrocardiograms (ECGs), a convolutional neural network (CNN) was built to discriminate participants with and without heart failure. As comparisons, known risk variables for heart failure were fitted into a logistic regression model and a random forest classifier. Of the 5299 individuals included into this study, 318 individuals (6.0%) had heart failure. Using only the digital ECGs instead of several risk variables as an input, the CNN provided a comparable predictive accuracy for heart failure versus the logistic regression model and the random forest classifier [area under the curve (AUC) of 0.75, a sensitivity of 0.67 and a specificity of 0.69 for the CNN; AUC 0.77, a sensitivity of 0.63 and a specificity of 0.76 for the logistic regression; AUC 0.79, a sensitivity of 0.67 and a specificity of 0.72 for the random forest classifier].ConclusionsUsing a CNN build on digital ECGs only and requiring no additional input, we derived a screening tool for heart failure in the general population. This could be perfectly embedded into clinical routine of general practitioners, as it builds on an already established diagnostic tool and does not require additional, time‐consuming input. This could help to alleviate the underdiagnosis of heart failure.

Most of somatic mutations in cancer occur outside of gene coding regions. These mutations may disrupt the gene regulation by affecting protein-DNA interaction. A study of these disruptions is important in understanding tumorigenesis. However, current computational tools process DNA sequence variants individually, when predicting the effect on protein-DNA binding. Thus, it is a daunting task to identify functional regulatory disturbances among thousands of mutations in a patient. Previously, we have reported and validated a pipeline for identifying functional non-coding somatic mutations in cancer patient cohorts, by integrating diverse information such as gene expression, spatial distribution of the mutations, and a biophysical model for estimating protein binding affinity. Here, we present a new user-friendly Python package BayesPI-BAR2 based on the proposed pipeline for integrative whole-genome sequence analysis. This may be the first prediction package that considers information from both multiple mutations and multiple patients. It is evaluated in follicular lymphoma and skin cancer patients, by focusing on sequence variants in gene promoter regions. BayesPI-BAR2 is a useful tool for predicting functional non-coding mutations in whole genome sequencing data: it allows identification of novel transcription factors (TFs) whose binding is altered by non-coding mutations in cancer. BayesPI-BAR2 program can analyze multiple datasets of genome-wide mutations at once and generate concise, easily interpretable reports for potentially affected gene regulatory sites. The package is freely available at http://folk.uio.no/junbaiw/BayesPI-BAR2/.

Integrating machine learning (ML) models into healthcare systems is a rapidly evolving field with the potential to revolutionize care delivery. This study aimed to classify fertility rates and identify significant predictors using ML models among reproductive women in Ethiopia. This study utilized eight ML models in 5864 reproductive-age women using Ethiopian Demographic Health Survey (EDHS), 2019 data. Phyton programming language was used to develop these models. Predictors of fertility rate were determined using the feature important techniques. The performance of models was evaluated using accuracy, area under the curve (AUC), precision, recall, F1-score, specificity, and sensitivity. The mean age of participants was 32.7 (± 5.6) years. The random forest classifier (accuracy = 0.901 and AUC = 0.961) followed by a one-dimensional convolutional neural network (accuracy = 0.899 and AUC = 0.958), logistic regression (accuracy = 0.874 and AUC = 0.937), and gradient boost classifier (accuracy = 0.851 and AUC 0.927) were the top performing ML models. Family size, age, occupation, and education with an average importance score of 0.198, 0.151, 0.118, and 0.081, respectively were the top significant predictors of the fertility rate. The best ML models to classify and predict fertility rates were random forest, one-dimensional convolutional neural network, logistic regression, and gradient boost classifier. The findings on important factors of fertility rate can inform targeted public health, programs that address disparities related to family size, occupation, education, and other socioeconomic factors.

Background: Liver cancer accounts for more than 700,000 deaths each year making it the third leading cause of cancer-related deaths worldwide. Late diagnosis of the disease is the reason behind most deaths. Driver mutations are genetic alterations in tumor cells, which are responsible for the development of liver cancer; therefore, the identification of genetic biomarkers is necessary for the prediction and early diagnosis of liver cancer. Objectives: The main objective of this study is to identify pathogenic alleles that may act as potential biomarkers for the prediction of liver cancer. It also identifies the role of novel genes in liver cancer that are not known to cause the disease. Methods: The mutation data of non-coding variants were downloaded from the catalogue of somatic mutations in cancer (COSMIC) databases. Different bioinformatics tools were, then, used to retrieve mutations in liver cancer. The genetic alterations in hepatocellular carcinoma (HCC) were analyzed. Results: The present study successfully identified pathogenic alleles (consistent mutations) along with a set of novel genes that might be involved in the development of liver cancer. It identified non-coding mutations near human genes and transcription factor binding sites of HepG2 cells. This study also identified mutations near the genes that are involved in the Ras/MAFK signaling pathway of the Hepatitis B virus. Conclusions: The pathogenic alleles identified in this study may provide targeted therapy for the treatment of liver cancer. The identification of novel genes may help to understand the progression of liver cancer at the molecular level. The identified driver mutations may act as potential biomarkers and therapeutic targets for early prediction and treatment of liver cancer.

Tuberculosis (TB) diagnostics urgently require non-sputum biomarkers to address the limitations of conventional methods in distinguishing active TB (ATB) from latent infection (LTBI), healthy controls (HCs), and TB-mimicking diseases (ODs, other diseases). Transcriptomic data from GSE83456 and GSE152532 were combined to form the selection dataset. Marker genes were identified from differentially expressed autophagy-related genes using a Random Forest classifier. The optimal gene signature was selected based on optimal performance through a linear Support Vector Machine (SVM) classifier with cross-validation. The signature was subsequently evaluated in six independent evaluation datasets and validated using whole blood samples collected from 70 participants. We identified a novel four-gene autophagy-related signature (CASP1, FAS, TRIM5, C5) in the selection dataset. This signature demonstrated robust diagnostic accuracy across multiple evaluation datasets: Area Under the Curve (AUC) 0.83-0.98 for ATB vs. LTBI and 0.82-0.94 for ATB vs. HCs. Crucially, it maintained high specificity (AUC 0.89-0.90) against ODs. RT-qPCR validation in whole blood samples confirmed elevated expression in ATB, while an SVM model achieved promising differentiation (AUC 0.86 for ATB vs. LTBI and AUC 0.99 for ATB vs. HCs). Our findings yielded a four-gene signature in whole blood that is robustly diagnostic for ATB, validated across multiple evaluation datasets and clinical samples. The autophagy-driven specificity and PCR-compatible design of this signature offer a blood-based, cost-effective strategy to enhance TB detection, addressing WHO-aligned diagnostic needs.