Gene expression and cancer classification

This paper introduces a statistical nonparametric method to identify consistently important features across samples and demonstrates its application to RNA sequencing data.

The paper normalizes gene expression data using multiple housekeeping genes to identify consistent genetic correlations in breast cancer, finding that multiple gene normalization improves correlation consistency compared to single gene normalization.

The paper introduces an ensemble classifier using XGBoost to robustly classify tumor samples into six immune subtypes based on gene-level features.

The paper constructs a colorectal cancer diagnosis model using expression profiles and a BPNN algorithm, achieving high accuracy and AUC values.

The paper uses independent component analysis on bulk tissue microarrays to derive gene expression signatures that reveal biologically relevant transcriptional modules in Drosophila brain tissues.

AdaTiSS provides a data-adaptive method to quantify tissue specificity scores for genes, improving comparability across tissues and genes.

This study evaluates 11 feature selection methods across 27 biomedical datasets to determine their performance based on data complexity and provides recommendations for method selection.

This paper introduces a corrected estimator for mutual information that accounts for measurement error, showing improved accuracy in simulated and real gene expression data.

The paper investigates how technical noise affects bulk RNA-seq differential gene expression analysis and proposes noise-adjustment methods to improve consistency across tools.

PCAGO is a web tool that uses PCA to analyze RNA-Seq data, offering normalization, filtering, visualization, and parameter selection to improve clustering and interpretation.

This paper develops a deep learning model integrating microarray and clinical data to predict NSCLC prognosis with high accuracy.

The paper evaluates multiple compression methods and dimensionalities on gene expression data, finding no single best approach but showing benefits from ensemble methods across different dimensions.

DNetPRO identifies low-dimensional signatures from high-throughput data by combining discriminant analysis with network processing, outperforming existing methods with fewer variables.

The paper benchmarks machine learning models for genetic data analysis using FRESA.CAD Binary Classification Benchmarking, showing quantitative and qualitative advantages over existing tools.

The paper develops a web tool to quantify and visualize gene signature expression in blood cell types using RNA data.

This study compares gene expression profiles of leukocytes from different isolation methods, finding that RBC lysis reduces globin mRNA and improves transcriptome quality.

This paper introduces a multimodal framework to detect gene-gene interactions from large expression data, showing improved performance over existing methods in simulations and biological applications.

The paper develops a Bayesian method to identify differentially expressed transcripts from RNA-seq data, showing improved performance for lowly expressed transcripts in breast cancer analysis.

The paper introduces superSeq, a method to determine optimal sequencing depth for RNA-Seq studies, finding most studies are undersequenced and saturation depends on statistical methods.

This paper explores using balances to engineer features for classifying health biomarkers and benchmarks transformation methods across different biomarker types.

The paper uses a self-organizing map to analyze differential expression data from MMP-9 inhibitor-treated T-cells, identifying novel gene targets regulated by MMP-9.

This paper introduces a new distance metric, dr, that satisfies the triangular inequality for gene expression profiles and shows it provides more robust clustering than the absolute correlation distance.

The paper evaluates representation learning techniques for transcriptomics-based phenotype prediction using public data and finds that l2-regularized regression on centered log-ratio transformed abundances performs best.

The paper introduces iDEG, a method to identify differentially expressed genes in a single subject without replicates, showing improved accuracy over existing methods in simulations.

The paper introduces absSimSeq, a tool for simulating RNA-Seq data that accounts for compositional data properties, and shows that compositional normalization improves differential analysis performance.

The paper introduces a semi-parametric Bayesian method for differential expression analysis in RNA-seq data, showing improved performance over existing methods.

This study introduces DGS, a method combining denoising autoencoders and transfer learning to predict prostate cancer metastasis using both labelled and unlabelled gene expression data, achieving higher accuracy than existing signatures.

The paper introduces drPCA, a supervised dimensionality reduction method that improves differentiation in case/control studies by using differential vectors, and shows it outperforms PCA and PLS-DA in cancer gene expression datasets.

MS.liverK is an R package that implements tools for computing molecular subtypes and functional signatures in liver cancer using transcriptome data.

The paper introduces a framework to rank genomic features by differential effect size, showing that pseudocounts optimize the balance between fold-change and difference sorting.