matrix factorization

Devising reliable and accurate epigenetic clocks: choosing the optimal computational solution

This paper evaluates a method using principal components for age and cancer prediction, finding it less accurate than CpG-based models despite improved reliability.

Systemic Metabolic Depletion of Intestine Microbiome Undermines Melanoma Immunotherapy Effectiveness

This study identifies that systemic metabolic depletion of the gut microbiome reduces immunotherapy effectiveness in melanoma by altering cobalamin and amino acid metabolism.

Dimensionality reduction methods for extracting functional networks from large-scale CRISPR screens

This study evaluates dimensionality reduction methods to normalize CRISPR screen data and improve functional gene network extraction from the DepMap.

Single-cell multi-omics map of human foetal blood in Down's Syndrome

This study integrates single-cell multi-omics data from human foetal blood to show that trisomy restructures enhancer-gene maps, leading to dysregulated erythroid differentiation and haematological abnormalities in Down's Syndrome.

Delineating mouse β-cell identity during lifetime and in diabetes with a single cell atlas

This paper integrates multiple scRNA-seq datasets to create a mouse islet atlas, revealing new insights into β-cell states and differences across diabetes models.

Examining dynamics of three-dimensional genome organization with multi-task matrix factorization

This paper introduces TGIF, a multi-task NMF method for analyzing 3D genome dynamics across Hi-C datasets, identifying structural changes linked to regulatory signals and gene expression.

Mapping the Macrostructure and Microstructure of the in vivo Human Hippocampus using Diffusion MRI

This paper uses diffusion MRI and computational methods to map macro- and microstructural features of the human hippocampus, revealing distinct subfield characteristics and potential fiber pathway orientations.

Tensor decomposition and principal component analysis-based unsupervised feature extraction outperforms state-of-the-art methods when applied to histone modification profiles

The paper introduces a method combining tensor decomposition and PCA to identify histone modifications from genomic data, outperforming existing methods.

GeneVector: Identification of transcriptional programs using dense vector representations defined by mutual information.

GeneVector uses mutual information-based dimensionality reduction to identify transcriptional programs and classify cell types across single-cell RNA-seq datasets.

Alterations in grey matter structure linked to frequency-specific cortico-subcortical connectivity in schizophrenia via multimodal data fusion

Malaria resistance-related biological adaptation and complex evolutionary footprints of Tai-Kadai people inferred from 796 genomes

This study analyzes 796 Tai-Kadai genomes to uncover genetic adaptations and evolutionary history, revealing Malaria resistance and population structure in TK-speaking groups.

LSMMD-MA: Scaling multimodal data integration for single-cell genomics data analysis

LSMMD-MA scales multimodal data integration for single-cell genomics to handle up to a million cells per modality, improving on existing methods by two orders of magnitude.

SparseRNAFolD: Sparse RNA pseudoknot-free Folding including Dangles

SparseRNAFolD improves RNA secondary structure prediction by reducing time and space complexity while including dangles contributions, outperforming LinearFold in memory use and speed for sequences up to 1000 bases.

Complexity of spatiotemporal plantar pressure patterns during everyday behaviours

This study analyzes spatiotemporal plantar pressure patterns during various daily activities, revealing spatial complexity and identifying key pressure components.

Dictionary learning for integrative, multimodal, and scalable single-cell analysis

Dictionary learning enables assumption-free cross-modality single-cell mapping via multi-omic bridge datasets and scales single-modality integration to millions of cells in under an hour via atomic sketching.

GARSA: An integrative pipeline for genome wide association studies and polygenic risk score inference in admixed human populations

GARSA is a pipeline for GWAS and PRS analysis in admixed populations, incorporating bias correction and multiple analytical methods.

Antithrombin, protein C and protein S: Genome and transcriptome wide association studies identify 7 novel loci regulating plasma levels

Genomic evidence of genuine wild versus admixed olive populations evolving in the same natural environments in western Mediterranean Basin

The paper investigates genetic structure in wild and cultivated olive populations in the western Mediterranean Basin, finding more admixed than genuine wild olive trees.

A multi-omics atlas of the human retina at single-cell resolution

This paper presents a comprehensive single-cell multi-omics atlas of the human retina, integrating RNA-seq and ATAC-seq data to identify cell-type-specific cis-element-gene relationships and transcription factor regulons.

Missense variants in PLA2G6 contribute to a spectrum of clinical syndromes and provide pharmacogenomic correlates

The paper identifies deleterious PLA2G6 variants in individuals with severe mental illness and suggests pharmacogenomic implications for antipsychotic drug binding.

The temporal transcriptomic signature of cartilage formation

This study examines the temporal transcriptomic changes during in vitro chondrogenesis using RNA sequencing and identifies key transcription factors involved in cartilage formation.

Structural and functional classification of G-quadruplex families within the human genome

The paper classifies G-quadruplex sequences into families using computational methods and annotates them based on sequence, structure, and functional properties.

Graph Regularized Probabilistic MatrixFactorization for Drug-Drug Interactions Prediction

This paper introduces a graph-regularized probabilistic matrix factorization method for predicting drug-drug interactions and shows it outperforms existing techniques on the DrugBank dataset.

Network-based multi-omics integration reveals metabolic at-risk profile within treated HIV-infection

The paper identifies three metabolic at-risk profiles in HIV-treated patients using multi-omics integration, highlighting dysregulated metabolism and microbiome interactions.

μ-PBWT: Enabling the Storage and Use of UK Biobank Data on a Commodity Laptop

The paper introduces μ-PBWT, a memory-efficient method for haplotype matching, reducing memory usage by up to 25 times and enabling storage of large datasets on a commodity laptop.

A Spectral Method for Assessing and Combining Multiple Data Visualizations

The paper introduces a spectral method to evaluate and combine multiple data visualizations, improving the quality of the consensus visualization by leveraging eigenscores.

Characterization of Integrase and Excisionase Activity in Cell-free Protein Expression System Using a Modeling and Analysis Pipeline

The paper presents a pipeline for modeling and parameter identification in biological systems, applied to characterize integrase and excisionase activity in cell-free systems.

Probabilistic embedding, clustering, and alignment for integrating spatial transcriptomics data with PRECAST

PRECAST integrates spatial transcriptomics data from multiple sources, improving cell and domain detection through probabilistic embedding, clustering, and alignment.

Trans-Acting Genotypes Drive mRNA Expression Affecting Metabolic And Thermal Tolerance Traits

The paper identifies trans-acting eQTLs driving mRNA expression linked to physiological traits in killifish, showing most eQTLs are trans-acting rather than cis-acting.

Integrative Multiomics to Dissect the Lung Transcriptional Landscape of Pulmonary Arterial Hypertension

The paper identifies a gene co-expression module, the pink module, associated with PAH severity, genetic risk, vascular cell types, and therapeutic targets through multiomics integration.