Gene expression and cancer classification

FrozenChicken: Promoting the meta-analysis of chicken microarray data

FrozenChicken provides frozen vectors for chicken microarray data to facilitate meta-analysis in vertebrate development studies.

sweetD: An R package using Hoeffding's D statistic to visualise the dependence between M and A for large numbers of gene expression samples

sweetD visualizes dependence between gene abundance and log fold difference in large expression datasets using Hoeffding's D statistic.

Hyperbolic geometry of gene expression

This paper shows that gene expression data can be better understood using hyperbolic geometry instead of Euclidean geometry, improving visualization and analysis.

BiSEK: a platform for a reliable differential expression analysis

BiSEK improves differential expression analysis by identifying affected pathways without relying on individual gene significance, showing results relevant to FUBP1 depletion and SARS-CoV-2 infection.

Invariant Differential Expression Analysis Reveals Mechanism of Cancer Resistance to Cell Cycle Inhibitors

The paper introduces iDEA, a method to improve differential expression analysis in cancer, revealing mechanisms of resistance to cell cycle inhibitors in retinoblastoma.

CuBlock: A cross-platform normalization method for gene-expression microarrays

CuBlock normalizes gene-expression microarray data across platforms, effectively separating biological groups, outperforming existing methods in this task.

rePROBE: Workflow for Revised Probe Assignment and Updated Probe-Set Annotation in Microarrays

The paper introduces rePROBE, an R package that updates probe assignments and annotations for microarrays using current genomic and transcriptomic data, improving accuracy for multiple species.

Per-sample standardization and asymmetric winsorization lead to accurate clustering of RNA-seq expression profiles

The paper introduces AWST, a data transformation method that improves clustering accuracy in RNA-seq data by reducing the need for gene selection.

GEOlimma: Differential Expression Analysis and Feature Selection Using Pre-Existing Microarray Data

GEOlimma improves differential expression analysis by integrating pre-existing microarray data, enhancing power and classification performance compared to Limma.

A regularized functional regression model enabling transcriptome-wide dosage-dependent association study of cancer drug response

The paper develops a regularized functional regression model to identify dosage-dependent gene expression associations with cancer drug response using in vitro data.

Adversarial generation of gene expression data

This study introduces gGAN, an adversarial simulator of gene expression data that outperforms existing simulators in realism and passes a Turing test for gene expression data.

On Transformative Adaptive Activation Functions in Neural Networks for Gene Expression Inference

The paper improves gene expression inference by replacing activation functions in neural networks, achieving lower mean absolute error.

Data-driven detection of subtype-specific and differentially expressed genes

This paper introduces OVE-FC and OVE-sFC tests to detect subtype-specific and differentially expressed genes, showing improved accuracy in deconvolution compared to existing methods.

COT: an efficient Python tool for detecting marker genes among many subtypes

COT detects marker genes across subtypes using cosine similarity in a one-sample test, applicable to multi-omics data.

HB-PLS: An algorithm for identifying biological process or pathway regulators by integrating Huber loss and Berhu penalty with Partial Least Square

HB-PLS identifies pathway regulators by integrating Huber loss and Berhu penalty with PLS, outperforming existing methods in efficiency and distinct regulator identification.

Multi-class Cancer Classification and Biomarker Identification using Deep Learning

This paper uses deep learning to classify 12 cancer types and identify biomarkers from RNA-SEQ data, achieving 94-95% accuracy and uncovering common cancer-related pathways and genes.

A new genome-wide method to identify genes with bimodal gene expression

The paper introduces a method to detect bimodal gene expression using Gaussian Mixture Models and identifies 554 bimodal genes across 25 tumor types, with 96 linked to prognosis differences.

BatchServer: a web server for batch effect evaluation, visualization and correction

BatchServer provides a web platform for evaluating, visualizing, and correcting batch effects in omics data using autoComBat, PVCA, and UMAP.

RSGSA: a Robust and Stable Gene Selection Algorithm

RSGSA is a supervised gene selection algorithm that improves stability and classification accuracy in gene expression datasets with phenotypes.

Latent Feature Representations for Human Gene Expression Data Improve Phenotypic Predictions

The paper constructs low-dimensional latent feature spaces of the human transcriptome using dimensionality reduction techniques and evaluates their ability to reconstruct data and improve phenotypic predictions.

A meta-learning approach for genomic survival analysis

This paper proposes a meta-learning framework for genomic survival analysis, showing it outperforms transfer learning with fewer samples and identifies important genes and pathways in cancer.

Identification of Differentially Expressed Gene Modules in Heterogeneous Diseases

DESMOND identifies differentially expressed gene modules in heterogeneous diseases by integrating gene expression and protein interaction data, improving reproducibility and biological relevance.

The shape of gene expression distributions matter: how incorporating distribution shape improves the interpretation of cancer transcriptomic data

This study shows that gene expression distributions in cancer data are often non-Normal, and using this information improves survival classification in cancer patients.

A hierarchical clustering and data fusion approach for disease subtype discovery

This paper introduces HC-fused, a hierarchical clustering and data fusion method for disease subtype discovery, showing superior performance on cancer data from TCGA.

Bias invariant RNA-seq metadata annotation

This paper introduces a deep-learning based domain adaptation algorithm for RNA-seq metadata annotation, achieving higher accuracy than existing methods for tissue, sample source, and patient sex prediction.

Reboot: a straightforward approach to identify genes and splicing isoforms associated with cancer patient prognosis

Reboot identifies genes and splicing isoforms linked to cancer patient prognosis using penalized Cox regression and bootstrap methods, validated on glioblastoma and pancreatic adenocarcinoma datasets.

dearseq: a variance component score test for RNA-Seq differential analysis that effectively controls the false discovery rate

dearseq is a differential expression analysis method that controls the false discovery rate without assuming data distribution, showing improved performance over existing methods in simulations and real data.

Robustifying genomic classifiers to batch effects via ensemble learning

This paper develops an ensemble learning approach to improve genomic classifiers by integrating predictions across batches rather than removing batch effects from merged data, showing better performance in independent validation.

Identifying Cross-Cancer Similar Patients via a Semi-Supervised Deep Clustering Approach

This paper identifies cross-cancer patients using a semi-supervised deep clustering approach, finding 20 patients that consistently co-cluster across different cancer types.

Maximizing the Reusability of Gene Expression Data by Predicting Missing Metadata

This study develops a framework to predict missing metadata in gene expression datasets to improve their reusability, outperforming standard metrics like F1-score.