Gene expression and cancer classification

reComBat removes batch effects from large-scale, multi-source omics data while preserving biological variation, outperforming existing methods for public gene expression data of Pseudomonas aeruginosa.

MFmap predicts cancer subtypes and similarity between cell lines and tumours using a semi-supervised generative model with high accuracy.

This paper introduces a restricted maximum-likelihood method to estimate latent variance components in gene expression data, improving efficiency and accuracy over gradient-based approaches.

This paper introduces OptICA, a method for selecting optimal dimensionality in ICA of transcriptomic data, showing it outperforms existing methods across multiple datasets.

The paper uses principal component analysis to identify differentially expressed genes in cells exposed to a collagen-glycosaminoglycan mesh compared to controls, finding enriched biological terms and better performance than linear regression.

The paper introduces DFC, a method using adaptive LASSO logistic regression to identify a small subset of genes that distinguish cell populations, addressing issues with P-value inflation in single-cell transcriptomics.

The paper applies sparse k-means clustering to human brain development data to identify age-related clusters and reveal lifespan-stage changes in the visual cortex.

The paper evaluates the impact of two-step batch correction on gene expression analysis and proposes strategies to mitigate its effects on statistical inference.

This study uses a variational autoencoder to encode tumor transcriptomes and combines it with XGBoost to predict chemotherapy response across five cancer types, achieving better performance than PCA or raw data in four types.

SGI identifies clinical subgroups in omics datasets using hierarchical clustering and association testing, demonstrated on a metabolomics dataset from a type 2 diabetes study.

This paper demonstrates that using stratified loss functions in deep learning improves survival prediction in multi-dataset settings by accounting for varying baseline hazards.

The paper applies matrix factorization methods to reduce high-dimensional multi-omic data in systems pharmacology, identifying predictive gene signatures for drug response.

This study compares five methods for analyzing gene expression in multi-sampled prostate tumors and benchmarks linear mixed models against generalized linear models for differential expression analysis.

This study combines ensemble systems biology feature selection with a bimodal deep neural network to improve breast cancer prognosis prediction, achieving high AUC and validated survival differentiation.

BlackSheep identifies molecules aberrantly enriched in small sample cohorts using non-parametric differential extreme value analysis.

The paper analyzes 2,692 gene expression datasets to show tissue heterogeneity affects 5-15% of samples, advocating for quality control methods to address this issue.

The paper introduces K2Taxonomer, an unsupervised algorithm for identifying molecular subgroups in bulk and single-cell transcriptomics data, and applies it to breast cancer TILs to find a shared transcriptional program associated with better prognosis.

The paper introduces Target-Decoy MineR to assess the biological relevance of variables in noisy datasets by calculating false-positive rates and biological relevance probabilities.

The paper develops a deep learning framework to integrate multi-omics data for cancer prognosis, showing improved risk prediction across 15 cancers and identifying prognostic markers in breast cancer.

noisyR reduces technical noise in sequencing data to enhance biological signal detection and improve downstream analysis consistency.

The paper evaluates the transcriptional fidelity of cancer models using a machine learning tool called CancerCellNet, finding that genetically engineered mice and tumoroids generally show higher fidelity than other model types.

The paper benchmarks RNA-seq association analyses for continuous exposures using DESeq2, edgeR, limma, and linear regression, finding linear regression methods are faster and better at controlling false detections.

This paper introduces PAN, a method to create sample-specific gene networks using annotations, improving breast cancer relapse prediction compared to existing methods.

The paper introduces a statistical test for detecting heterogeneous effects in omics data and shows its ability to identify disease-relevant genes missed by traditional methods.

This paper introduces GAIN-GTEx, a gene expression imputation method that outperforms existing approaches on RNA-Seq data from cancer types and diverse tissues.

The paper uses machine learning and differential expression analysis to identify secretory pathway genes associated with cancer characteristics, finding KIF20A and KIF23 as key genes across multiple cancer types.

The paper introduces an R package for divergence analysis of omics data, demonstrating its use with Cancer Genome Atlas data.

This study predicts tissue-specific gene expression from whole blood transcriptome using GTEx data, finding that blood expression can predict expression in many tissues and improve disease prediction.

This paper evaluates and adapts the CONSTANd normalization method for use with count data from sequencing, showing its efficiency and ability to remove bias across different experimental conditions.

This study compares sparse biclustering algorithms on simulated and real datasets, finding Bayesian algorithms with sparsity constraints perform well on simulated data but are slow, while post-processed non-negative matrix factorization algorithms show strong performance on real datasets.