It is demonstrated that pairing a genomic language model with widely available clinical imaging can serve as a hypothesis-free discovery tool for gene--imaging associations invisible to conventional approaches.
Abstract
The function of many genes is still unknown, and conventional driver-discovery methods, which rely on how frequently a gene is mutated, cannot assess genes that are only rarely affected. Here we pair Evo~2-based genome analysis with routine clinical imaging to identify gene--phenotype associations at genome-wide scale. For every somatic mutation across three TCGA cohorts (cRCC=clear cell renal cell carcinoma, HCC=hepatocellular carcinoma, and BC=breast cancer; $n = 340$ total), Evo~2 predicts a severity score, with no task-specific training. Per-gene severity summaries are then correlated with radiomic features extracted from paired tumor segmentations, controlling for total mutation burden. In TCGA-cRCC ($n = 162$), this sweep recovers established renal-cancer drivers and identifies 46 additional genes reaching false discovery rate (FDR) significance absent from curated cancer-gene panels, several of which are Mendelian ciliopathy and cytoskeletal-disease genes. These results demonstrate that pairing a genomic language model with widely available clinical imaging can serve as a hypothesis-free discovery tool for gene--imaging associations invisible to conventional approaches.
BackgroundGastric cancer (GC) remains a major cause of cancer-related mortality, and biomarkers for early detection are needed.MethodsStomach and blood expression quantitative trait loci were integrated with two GC genome-wide association studies using Mendelian randomization (MR), Bayesian colocalization, and summary-data-based MR/heterogeneity in dependent instruments (SMR/HEIDI) testing. Bulk and single-cell transcriptomic analyses characterized candidate expression and lesion-associated patterns. CDC16 protein expression was evaluated by immunohistochemistry in 53 paired GC and non-neoplastic tissues, followed by paired and exploratory receiver operating characteristic analyses.ResultsMR prioritized PILRB, CDC16, and GABPB1-AS1; SMR/HEIDI provided complementary support, while colocalization for CDC16 and GABPB1-AS1 was suggestive and model-dependent. Bulk-tissue CDC16 abundance was higher in GC, but the modest TCGA-STAD tumor-normal difference (log2FC = 0.210, FDR = 0.019) was attenuated after proliferation adjustment (log2FC = -0.002, FDR = 0.987), indicating close coupling with proliferative activity. Single-cell analysis localized CDC16 predominantly to epithelial populations, and the proportion of CDC16-detectable epithelial cells increased across lesion categories (ρ = 0.735; permutation P = 0.031). CDC16 H-scores were higher in GC than in paired non-neoplastic tissues (161.15 ± 45.11 vs 102.15 ± 54.50; P < 0.001), with higher cancer-tissue scores in 41 of 53 cases. Exploratory AUC was 0.794 (95% CI, 0.704-0.874; sensitivity, 66.0%; specificity, 79.2%).ConclusionsConvergent genetic, transcriptomic, and protein-level evidence prioritizes CDC16 as a GC-associated candidate tissue marker whose expression is closely linked to proliferative activity. Prospective validation in independent cohorts, including appropriate disease controls and blood-based evaluation, is warranted.
Fan Xue, Dao-Cheng Wu, Xu Zhang· International Journal of Bio...· 0 citations
TP53 and FOXA1 mutations, observed in primary and metastatic castration-resistant prostate cancer, were associated with aggressive disease phenotypes and potential therapeutic resistance, and these findings highlight TP53 and FOXA1 as promising biomarkers and potential therapeutic targets in prostate cancer.
Background/Objectives: Breast cancer (BC) is the most frequently diagnosed malignancy and a leading cause of cancer-related mortality in women worldwide. This disease is highly heterogeneous and dynamic, and chromosomal instability (CIN) plays a key role in the acquisition of these traits by generating genetic diversity that promotes tumor adaptation and influences therapeutic response and prognosis. Although several methods have been developed to quantify CIN, they are not readily applicable to human tumors and are limited in resolution, hindering a comprehensive understanding of intratumoral heterogeneity. In this study, we aimed to quantify CIN levels and clonal heterogeneity (CH) in HER2-positive (HER2+) and triple-negative (TNBC) breast cancer using single-cell RNA sequencing (scRNA-seq) data. Methods: We analyzed publicly available scRNA-seq data from HER2+ and TNBC tumors and non-malignant controls. CIN was scored at single-cell resolution using transcriptomic signatures, clonal heterogeneity was estimated from single-cell diversity metrics, and copy number alterations were inferred computationally. Differential expression and functional enrichment analyses were performed between cells with very low and extreme CIN levels, with key comparisons confirmed at the patient level. Results: Our analyses revealed pronounced intra- and intertumoral heterogeneity, with higher CIN levels in TNBC than in HER2+ and control samples. Genes differentially expressed in cells with extreme CIN values were mainly involved in cell division and related processes, and included candidate biomarkers not previously reported in this context. Our findings suggest a positive but statistically non-significant trend was observed between CIN and CH. Conclusions: Single-cell approaches such as scRNA-seq provide a powerful framework to elucidate CIN-related mechanisms and to identify potential biomarkers of BC aggressiveness and prognosis, supporting their further application in the study of intratumoral heterogeneity.
María Paula Meléndez-Flórez, N. Rangel, Milena Rondón-Lagos et al.· Biomedicines· 0 citations
The interplay between somatic mutations and copy number alterations influences tumor evolution and prognosis. These alterations are often treated independently, overlooking gene mutant dosage (GMD)—a key property of their interaction. Here we develop a computational framework that infers mutation copy number and multiplicity from targeted sequencing panels without requiring matched normal samples. We derive GMD for over 500,000 mutations across 60,000 pan-cancer samples. By stratifying more than 20,000 patients according to GMD across multiple genes, we identify 46 tumor-type-specific biomarkers predictive of survival, 13 of which were undetectable using binary mutant/wild-type models, 26 were associated with metastatic spread and 20 predicted metastatic tropism. Our method reveals GMD patterns as independent predictors of disease prognosis, metastatic potential and site-specific dissemination across diverse tumor types. This augmented insight into genomic drivers enhances our understanding of cancer progression and metastasis and holds the potential to substantially enhance biomarker discovery. The authors present INCOMMON, an open-source Bayesian inference tool that determines the multiplicity and copy number of driver mutations from tumor sequencing datasets.
N. Calonaci, E. Krasniqi, D. Čolić et al.· Nature Genetics· 0 citations
Triple-negative breast cancer (TNBC) is characterized by aggressive behaviour, high tumor heterogeneity, and an increased likelihood of recurrence and early metastasis. These factors hinder successful treatment. Genetic diagnosis enables personalized clinical recommendations and treatment options. The objective of this study was to validate whole-exome sequencing (WES) and variant prioritization in cancer susceptibility genes (CSG) associated with hereditary cancer (HC) predisposition in TNBC patients (n = 24). We present the development of a reproducible bioinformatic pipeline and its technical validation in a validation cohort (n = 25). This cohort comprised individuals with diverse primary tumors who had a previously confirmed molecular diagnosis of a hereditary cancer syndrome, serving as gold-standard cases to assess the pipeline’s analytical accuracy. We consolidated a comprehensive panel of cancer genes and determined all variants in the TNBC discovery cohort (12.5% of patients), identifying three pathogenic germline variants (gPV) in ATM, RAD51D, and BRCA1. These genes are involved in the molecular pathway of DNA repair by homologous recombination (HRD). Our results demonstrate that the developed bioinformatic pipeline provides reliable genetic diagnosis of cancer predisposition syndromes from exome data, applicable not only to TNBC patients but also to individuals with any cancer suspected of having a hereditary component.
D. Alzate, A. Y. Sánchez, Y. Pacheco et al.· PLoS ONE· 0 citations
Cancer of unknown primary (CUP) is a heterogeneous group of rare metastatic malignancies that lack an identifiable primary site. Molecular profiling has emerged as a tool to improve diagnostic accuracy, identify actionable alterations, and guide therapy. Relevant literature was identified through searches of PubMed, Scopus and Web of Science. We focused on tissue-of-origin (TOO) prediction, prognostic biomarkers, molecularly targeted therapy, and immunotherapy, synthesizing findings of clinical trials, meta-analyses, and cohort studies. Molecular analyses ranging from single‑gene assays to integrated multi‑omics (genomics, transcriptomics, proteomics and epigenomics) surveys enhance TOO determination, detect actionable DNA alterations and therapy selection. TOO-directed therapy has shown mixed survival benefit; however, integrating molecular data with clinical and pathological evaluation via multidisciplinary tumor boards seem to improve outcomes. Comprehensive genomic profiling enables targeted therapy for patients with actionable alterations such as MSI-high, TMB-high, BRAFV600E, NTRK, or RET fusions. Immunotherapy has shown promising activity, especially in biomarker-selected subgroups, although the predictive criteria remain heterogeneous. Molecular profiling is increasingly central in CUP management, supporting both diagnostic clarification and individualized treatment. Evidence indicates that molecularly guided therapy, particularly when integrated with clinicopathological data, can improve survival and expand treatment options beyond empirical chemotherapy. Future research should aim to optimize biomarker-driven strategies, standardize predictive assays, and improve access to advanced molecular diagnostics.
K. Tałasiewicz, Michał Mikula, Aleksandra Kapała· Critical reviews in oncology...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
