Cancer Genomics and Diagnostics

The paper develops a mutation simulator to assess genome robustness under COSMIC mutational signatures, finding higher resilience in human and bat genomes to APOBEC-related mutations.

This study analyzes the evolutionary patterns and genomic instability in Wilms tumors, identifying subtype-specific evolutionary sequences and proposing CNA diversity as a prognostic biomarker.

CaMutQC is an R package that performs quality control and filtering of cancer somatic mutations, improving accuracy and comparability across studies.

This study uses linked-read sequencing to phase somatic mutations in multiple myeloma, extending haplotype blocks and revealing phasing patterns in key mutated genes.

This study identifies pre-analytic factors affecting NGS success in FFPE samples for lung and ovarian cancer, highlighting DNA quality, sample type, and fixation time as key predictors.

This paper establishes a diverse collection of preclinical models for pediatric solid tumors and demonstrates their utility in precision medicine trials and drug testing.

The paper develops CliPP to estimate subclonal mutation load (sML) and shows it predicts survival and immunotherapy response in cancers with low to moderate TMB.

StarSignDNA is an algorithm for mutational signature analysis in cancer sequencing data, improving accuracy and biological validity by addressing overfitting and underfitting.

This study evaluates tree balance statistics for detecting growth rate heterogeneity in lineage tracing trees of cancer populations, finding J1 performs well while Sackin index is less effective.

The paper identifies previously overlooked activating promoter mutations in cancer using a computational workflow applied to the PCAWG dataset, validating several candidates through luciferase assays.

The paper develops a web application that uses machine learning to match genetic alterations with drug candidates for cancer patients, identifying optimal clusters of drugs based on interaction scores.

This study compares cancer cell lines to native cancers using CNV profiles and ML to identify cell lines that best represent specific cancer types.

This study shows that DNA mismatch repair-deficient mouse fibroblasts can tolerate high mutation burdens in vitro without significant growth effects, revealing mutation-driven population changes and negative selection.

The paper introduces Jumble, an algorithm that improves the sensitivity of detecting copy number alterations in low ctDNA samples, achieving detection limits of 1-2% for amplifications and 4-8% for deletions.

The paper identifies co-occurring mutation pairs in cancer genomes that are associated with metastasis and provides a tool for predicting metastatic potential.

The paper introduces a divide-and-conquer method for large-scale evolutionary analysis of single-cell DNA data, demonstrating its effectiveness through experiments on clustering and dimensionality reduction methods.

DIALECT improves the identification of mutual exclusivity and co-occurrence between driver mutations by distinguishing them from passenger mutations in cancer genomics data.

The paper shows that human fibroblasts undergo negative selection against harmful mutations, maintaining growth despite high mutation loads.

This paper introduces an optimal transport-based domain adaptation method to correct technical biases in cfDNA sequencing data, improving cancer detection and enabling cohort integration.

The paper identifies two lung cancer subtypes defined by APOBEC mutagenesis levels and shows their association with different genetic mutations, aging, and immune responses.

The paper uses machine learning to identify cancer driver mutations, finding that methods incorporating protein structure or functional genomic data outperform those using only evolutionary data, and validates their predictions using real-world patient cohorts.

The paper investigates how PTEN and ARID1A haploinsufficiency in colonic epithelium contributes to oncogenic transformation by altering clone dynamics and creating a pro-tumorigenic microenvironment.

The paper analyzes cancer mutations in protein structures to identify novel driver genes by distinguishing mutational patterns in tumour suppressors and oncogenes.

DeepCNA identifies cancer-specific copy number aberrations across 13 cancer types using explainable deep learning for diagnosis and biomarker discovery.

This paper develops a method to detect somatic mutations in tumor samples without matched normal samples using long-read sequencing and demonstrates its application on cancer cell lines.

This study uses machine learning to identify 5-hydroxymethylation biomarkers in cell-free DNA that can detect occult colorectal cancer up to 36 months before diagnosis.

The paper introduces POI, a tool that integrates genomic and transcriptomic data to suggest personalized therapies by matching patients to treatments using PreMedKB and predictive algorithms.

This paper introduces a method to enrich short double-stranded cfDNA from plasma for genome-wide DNA footprinting, identifying disease-specific biomarkers in cancer and sepsis.

This study evaluates how different random seeds affect subclonal reconstruction algorithms in cancer genomics, finding significant variability in subclone estimates across seeds.

BayVarC is a Bayesian variant caller for liquid biopsy that detects low-frequency variants with high sensitivity and low false positive rate.