Gene expression and cancer classification

The impact of different sources of heterogeneity on loss of accuracy from genomic prediction models

This study investigates how different sources of heterogeneity affect the accuracy of genomic prediction models across studies, finding that some sources have limited impact while others, like identical generative models, significantly improve consistency.

Independent component analysis provides clinically relevant insights into the biology of melanoma patients

The paper uses independent component analysis to analyze melanoma transcriptomic data, distinguishing biological signals from technical biases and identifying clinically relevant patient characteristics.

A prize-collecting Steiner tree application for signature selection to stratify diffuse large B-cell lymphoma subtypes

The paper applies a prize-collecting Steiner tree method to select a gene signature for classifying diffuse large B-cell lymphoma subtypes, outperforming existing methods.

Deep Learning Based Tumor Type Classification Using GeneExpression Data

This paper classifies 33 tumor types using deep learning on RNA-Seq data, achieving 95.59% accuracy, higher than GA/KNN, and validates tumor-specific biomarkers through heat-map analysis.

DCARS: Differential correlation across ranked samples

DCARS identifies gene pairs with significant changes in correlation across survival stages, outperforming existing methods in cancer datasets.

Using supervised learning methods for gene selection in RNA-Seq case-control studies

This study compares supervised learning methods to differential expression analysis for gene selection in RNA-Seq case-control studies, showing improved survival-associated gene identification in cancer datasets.

Diagnostic value of blood gene expression-based classifiers as exemplified for acute myeloid leukemia

The paper develops robust and accurate classifiers for AML diagnosis using blood gene expression data and machine learning across 12,029 samples from 105 studies.

Introduce a New Approach to Detect Genes Associated to Oral Squamous Cell Carcinoma

This paper introduces a new method to detect genes associated with oral squamous cell carcinoma by analyzing both mean and variance differences between cases and controls, identifying 456 differentially variable and 2,375 differentially expressed gene probes.

Differential gene expression analysis tools exhibit substandard performance for long non-coding RNA-sequencing data

The paper evaluates DE tools for RNA-seq data, finding poor performance for lncRNAs and highlighting the need for more samples to achieve acceptable sensitivity.

Impulse model-based differential expression analysis of time course sequencing data

ImpulseDE2 improves differential expression analysis for time course sequencing data by modeling gene expression trajectories with an impulse model, outperforming existing methods in distinguishing transient and monotonic changes.

Differential expression analysis of log-ratio transformed counts: benchmarking methods for RNA-Seq data

The paper benchmarks ALDEx2, a log-ratio transformation method, against conventional RNA-Seq differential expression tools, showing it has high precision and comparable performance on real data.

ACDtool: a web-server extending the original Audic-Claverie statistical test to the comparison of large data sets of counts.

ACDtool extends the Audic-Claverie test to compare large count datasets across multiple omics fields, offering three analysis types and a web interface.

Detecting Differential Variable microRNAs via Model-Based Clustering

The paper introduces a model-based clustering method to detect differentially variable microRNAs, outperforming the probe-wise F test and identifying 7 novel microRNAs in hepatocellular carcinoma.

GEESE: Metabolically driven latent space learning for gene expression data

GEESE uses a metabolic model to improve latent space learning for gene expression data, capturing biologically relevant features and enhancing generalization.

IDEA: a web server for Interactive Differential Expression Analysis with R Packages

IDEA provides a web-based interface for differential expression analysis using R packages, enabling non-programmers to perform analysis without coding.

Adjusting for systematic technical biases in risk assessment of gene signatures in transcriptomic cancer cohorts

The paper addresses technical biases in gene signature risk assessment in cancer cohorts by proposing a method to correct for systematic variability, ensuring accurate survival associations.

Biological classification with RNA-Seq data: Can alternative splicing enhance machine learning classifier?

This study evaluates if isoform-level RNA-Seq data improves supervised machine learning performance in biological classification tasks compared to gene-level data.

Stratified time-course gene preselection shows a pre-diagnostic transcriptomic signal for metastasis in blood cells: a proof of concept from the NOWAC study

The paper shows that stratified time-course gene preselection improves prediction of metastasis in blood cells using NOWAC data.

runibic: a Bioconductor package for parallel row-based biclustering of gene expression data

The paper introduces a Bioconductor package called runibic that provides a parallel version of the UniBic biclustering algorithm for gene expression data, improving speed and memory efficiency.

Blazing Signature Filter: a library for fast pairwise similarity comparisons

The paper introduces BSF, a fast pairwise similarity algorithm that filters unproductive comparisons to enable efficient data mining.

A data mining paradigm for identifying key factors in biological processes using gene expression data

This paper introduces a data mining paradigm to identify key factors in biological processes using gene expression data, demonstrating its effectiveness in epidermal development and cold-induced thermogenesis.

DeepProfile: Deep learning of cancer molecular profiles for precision medicine

DeepProfile trains a variational autoencoder on thousands of unlabeled gene expression samples to learn low-dimensional representations that improve prediction of acute myeloid leukemia drug responses and cancer phenotypes.

polyCluster: Defining Communities of Reconciled Cancer Subtypes with Biological and Prognostic Significance

polyCluster reconciles cancer subtypes from different clustering methods to identify prognostically significant subtype communities in breast and uveal melanoma datasets.

Nonparametric analysis of contributions to variance in genomics and epigenomics data

MAVRIC is a nonparametric method that quantifies variance explained by covariates and performs differential analysis on various genomics and epigenomics data types.

CCA based multi-view feature selection for multi-omics data integration

This study introduces a CCA-based multi-view feature selection method for integrating multi-omics data, showing improved survival prediction for kidney renal clear cell carcinoma compared to single-view and existing fusion approaches.

Personalized Regression Enables Sample-Specific Pan-Cancer Analysis

This paper introduces a personalized regression method to analyze pan-cancer gene expression data, revealing sample-specific patterns that population-level methods miss.

A deep learning approach for uncovering lung cancer immunome patterns

This paper uses deep Boltzmann machines to model immune cell gene expression patterns in lung adenocarcinoma, linking hidden nodes to clinical characteristics and immune cell infiltration.

Differential Proportionality - A Normalization-Free Approach To Differential Gene Expression

This paper introduces a normalization-free method for detecting differential gene expression by analyzing changes in gene ratio proportionality between sample groups.

Continuous Biomarker Assessment by Exhaustive Survival Analysis

The paper introduces survivALL, an R package for continuous biomarker assessment through survival analysis, enabling reliable evaluation of gene expression cut-points across datasets.

Bi-clustering interpretation and prediction of correlation between gene expression and protein abundance

The paper uses bi-clustering to identify gene clusters with consistent expression-protein abundance correlations and builds a prediction model to forecast protein abundance from mRNA data.