Gene expression and cancer classification

This paper introduces an R package for multiple kernel learning, showing improved classification by combining multiple data sources in ovarian cancer.

The paper estimates subject- and cell-type-specific gene expression from tissue using an empirical Bayes method, enabling downstream analyses like eQTL identification.

The paper evaluates cell lines as models of primary tumors across 22 types using transcriptomic data, finding cell-cycle pathways are consistently upregulated in cell lines.

This study predicts genomic alterations from gene expression in multiple cancers, showing gene aggregation improves prediction accuracy in noisy single-cell data.

The paper introduces a nonparametric method for RNA-seq differential expression analysis that accounts for inferential uncertainty, showing improved false discovery rate control compared to existing methods.

BALLI improves differential expression analysis in RNA-seq by addressing small sample bias and providing better power and precision than existing methods.

This paper introduces an anomaly detection approach to classify cancer using only healthy tissue transcriptomes, showing improved accuracy with more normal samples.

The paper evaluates single-subject methods for differential gene expression analysis and finds ensemble methods perform best in identifying DEGs without replicates.

MetaGxData provides standardized cancer transcriptomic and clinical data for cross-tumour analyses, identifying common prognostic genes and creating a multi-cancer gene signature.

pcaExplorer provides an interactive R/Bioconductor tool for RNA-seq PCA analysis, enabling reproducible exploration and visualization of principal components.

TuBA identifies clinically relevant tumor transcriptional profiles in breast cancer by detecting biclusters with significant gene set agreement and clinical implications across multiple datasets.

ABioTrans is an R-based web tool for transcriptomics analysis that provides statistical methods and visualization for RNA-Seq data from GEO.

The paper introduces a new normalization method, RUV-III, for Nanostring nCounter gene expression data and evaluates its effectiveness across multiple datasets.

NormExpression evaluates normalization methods using AUCVC and mSCC metrics, showing consistency across bulk and scRNA-seq data, and provides an R package for method comparison.

This paper improves neuropsychiatric condition classification by using gene ontology terms as features, enhancing performance, stability, and interpretability of classifiers.

The paper defines common gene co-expression principles across cancers to improve molecular stratification by identifying consistent expression modules and their associations with mutation states.

This paper introduces JMAP, a model averaging method for genetic risk prediction that integrates pathway information, showing improved accuracy over existing methods in simulations and real cancer datasets.

The paper investigates how DNA repair gene expression, adjusted by the PCNA metagene, predicts overall survival in 18 cancers using TCGA data, finding significant predictive power in 61% of cancers.

The paper introduces a stabilized ICA method that improves reproducibility and interpretability of transcriptomic signals across cancer datasets, and applies it to colorectal cancer.

Topconfects ranks differentially expressed genes by confident log fold change effect sizes, improving biological interpretation over p-value-based rankings.

NOJAH is an interactive tool for genome-wide cluster analysis and heatmap construction, providing replicable results through integrated workflows and documentation.

The paper develops a method to detect differential signals in high-throughput data from mixed cell-type samples and shows its superior performance through simulations and real datasets.

The paper proposes eight RNA-seq data transformations to improve analysis accuracy, showing that limma+rv and limma+l2 perform best in simulations and real data.

This paper introduces I-Boost, a method that improves survival time prediction by integrating multiple genomics data types with clinical variables, showing better accuracy than existing methods.

The paper introduces new features in the TCGAbiolinks R/Bioconductor package to improve cancer data analysis, including better differential expression analysis, tumor purity estimation, and integration of normal samples from GTEx.

netDx is a supervised patient classification framework that integrates diverse data types and provides interpretable results, showing high performance in cancer survival prediction.

The paper develops a statistical method to model transcriptional heterogeneity in breast tumors, identifying CBX2 as an oncogene overexpressed in invasive breast carcinoma and linked to poorer survival.

SUBSTRA is a Bayesian method for patient stratification that balances predictive performance and interpretability by using regularized biclustering and iterative feature weighting.

IRIS-DGE is a server-based tool for differential gene expression analysis that integrates multiple R-based methods and provides interactive visualizations for RNA-seq data.

ACE identifies correlations between clinical and genomic data in TCGA using an evolutionary algorithm for efficient analysis.