Cancer Genomics and Diagnostics

CanDrivR-CS: A Cancer-Specific Machine Learning Framework for Distinguishing Recurrent and Rare Variants

CanDrivR-CS distinguishes recurrent and rare cancer variants using cancer-specific gradient boosting models, achieving high F1 scores in cross-validation.

Evaluation of ensilication technology for ambient DNA preservation

This paper evaluates ensilication technology for ambient DNA preservation, finding it maintains sequencing fidelity and reduces artifacts compared to frozen storage.

Genotype-to-phenotype mapping of somatic clonal mosaicism via single-cell co-capture of DNA mutations and mRNA transcripts

The paper develops scG2P to jointly capture DNA mutations and mRNA transcripts in single cells, revealing clonal architectures and phenotypes in aged esophageal tissue.

SNAP: Streamlined Nextflow Analysis Pipeline for Immunoprecipitation-Based Epigenomic Profiling of Circulating Chromatin

SNAP improves classification performance for cell-free chromatin data while maintaining data retention through optimized quality filters and modular design.

Somatic and germline mutational processes across the tree of life

This study identifies somatic and germline mutational signatures across 661 species, revealing greater diversity than in human samples, likely due to environmental or biological factors.

Multimodal AI for Single cfDNA Profiling and Cancer Screening

The paper develops a Transformer-based model called cfAI to profile and detect cancer-derived cfDNA by integrating methylation, fragmentomics, and histone modifications, achieving high sensitivity and specificity in multi-cancer detection.

BASCULE: Bayesian inference and clustering of mutational signatures leveraging biological priors

BASCULE uses a Bayesian framework to standardize and cluster mutational signatures across cancer types, identifying prognostic subtypes in over 8,000 samples.

Revealing the Drivers Underlying Distinct Evolutionary Trajectories in Lung Adenocarcinoma

This study identifies three distinct evolutionary trajectories in lung adenocarcinomas using whole-genome sequencing, revealing that some never-smoking tumors follow smoking-like paths driven by KRAS mutations.

Single-cell RNA-sequencing of cerebral spinal fluid identifies circulating tumour cells in children with brain cancer

This study uses single-cell RNA-sequencing of cerebrospinal fluid to identify circulating tumour cells in children with brain cancer, revealing their dynamics and potential as biomarkers.

Structural semantic evolutionary distance (SSED) unifies the selection of cancer driver genes across macroevolution and tumorigenesis.

This paper introduces SSED, a metric using ESM-3 to quantify evolutionary divergence, showing cancer driver genes are structurally constrained and that oncogenic mutations follow similar principles, predicting immune response.

Comprehensive benchmarking of methods for mutation calling in circulating tumor DNA

The paper benchmarks 9 somatic variant calling algorithms on cfDNA samples from cancer patients, identifying optimal methods based on ctDNA levels and sequencing depth.

PAN-CANCER ANALYSES IDENTIFY ONCOGENIC DRIVERS, EXPRESSION SIGNATURES, AND THERAPEUTIC VULNERABILITIES IN RHO GTPase PATHWAY GENES

This study identifies oncogenic drivers, expression signatures, and therapeutic vulnerabilities in RHO GTPase pathway genes across multiple cancer types using pan-cancer genomic and transcriptomic data.

Inferring cancer type-specific patterns of metastatic spread using Metient

Metient infers cancer type-specific metastatic spread patterns using gradient-based optimization, revealing new insights into metastasis mechanisms and treatment strategies.

Benchmarking of duplex sequencing approaches to reveal somatic mutation landscapes

This study benchmarks six duplex sequencing technologies to evaluate their performance in detecting somatic mutations across different genomic contexts and tissue types.

Improved circulating tumor DNA profiling by simultaneous extraction of DNA methylation and copy number information from Methylated DNA Sequencing data (MeD-seq)

The paper introduces MeD-seq, a method to profile circulating tumor DNA by analyzing DNA methylation and copy number changes, showing its effectiveness in detecting cancer-specific signals in plasma samples.

Widespread Epistasis between Cancer Driver Mutations and Allele-Specific Copy Number Variations

The paper identifies epistatic interactions between cancer driver mutations and allele-specific copy number variations in tumors, showing their impact on patient survival.

A pathway-informed mutual exclusivity framework to detect genetic interactions in pediatric cancer

The paper introduces PIGI, a pathway-informed framework to detect genetic interactions in pediatric cancer, identifying 35 new candidate gene pairs with potential synthetic lethal interactions.

The Impact of Variant Calling on Substitution Mutational Signature Inference

This study evaluates how variant calling strategies affect mutational signature inference, finding that consensus calling reduces artifacts while preserving true signals.

Pan-cancer multi-omics analysis identifies non-canonical oncogenic functions beyond DNA replication of the POLE2 subunit of DNA Polymerase epsilon

This study identifies POLE2 as a contributor to genome stability, a cancer biomarker, and therapeutic target through pan-cancer multi-omics analysis.

Accurate Variant Classification in Tumour-Only Genomic Data Using Interpretable Tabular Models

This paper presents a workflow using XGBoost, LightGBM, and TabNet to classify variants in tumour-only data, showing improved accuracy across ancestries and tumour types.

Identifying cooperative genes causing cancer progression with dynamic causal inference

This paper introduces a dynamic causal inference method to identify cooperative cancer driver genes by analyzing gene interactions across cancer progression, showing significant overlap with known drivers and uncovering novel ones.

Tissue architecture and immune niches govern ctDNA release in colorectal cancer

This study identifies tissue architecture and immune niches as key factors influencing ctDNA release in colorectal cancer by integrating spatial transcriptomics and in situ mutation detection.

The subclonal footprint of pervasive early dissemination in pancreatic cancer

This study analyzes ctDNA shedding in pancreatic cancer using paired tissue and plasma whole genomes, finding that subclones contribute significantly to early dissemination and ctDNA levels.

DTCFinder: a bench-to-bits toolkit for label-free, whole-spectrum analysis of disseminated and circulating tumor cells in liquid biopsies

DTCFinder detects and characterizes disseminated and circulating tumor cells in liquid biopsies using single-cell RNA sequencing and machine learning without relying on labels or markers.

A benchmark of DNA methylation deconvolution methods for tumoral fraction estimation using DecoNFlow

This study benchmarks DNA methylation deconvolution methods for tumoral fraction estimation in cfDNA, finding CelFiE most accurate and introducing DecoNFlow as a comprehensive pipeline.

ATHENA: A deep learning-based AI for functional prediction of genomic mutations and synergistic vulnerabilities in prostate cancer

ATHENA predicts functional genomic mutations and synergistic vulnerabilities in prostate cancer using deep learning, identifying driver events and therapeutic targets through integration with variant discovery pipelines.

Tracking ongoing chromosomal instability using single-cell whole-genome sequencing

This paper develops a computational framework using single-cell whole-genome sequencing to identify ongoing chromosomal instability, showing its utility in cancer models and clinical samples.

Single-cell phylodynamics reveal rapid late-stage colorectal cancer expansions

The paper uses single-cell whole-genome sequencing and Bayesian phylodynamic modeling to show colorectal tumors often originate decades before diagnosis and expand rapidly in the last two years.

Polyclonal origins of human premalignant colorectal lesions

This study shows that many premalignant colorectal polyps contain multiple genetic lineages, indicating polyclonal origins rather than single-cell expansion.

Theoretical estimates on the expected number of mutations for reconstructing clonal lineage trees

The paper estimates the number of mutations needed to reconstruct clonal lineage trees from large datasets, showing that modest mutation counts suffice for accurate reconstruction.