dimensionality reduction

scPyviewer: a Python-native interactive viewer from AnnData single-cell data

scPyviewer provides a Python-native interactive viewer for AnnData single-cell data, matching R Shiny tools' features and outperforming them in speed and memory efficiency.

The Leiden ex vivo human growth plate model in severe tall stature: a proof-of-concept study

This study demonstrates the feasibility of establishing an ex vivo human growth plate model using tissue from percutaneous epiphysiodesis procedures to study bone growth and pediatric growth disorders.

Reproducibility of Diffusion, Shape, and Connectivity Metrics Across Scanners: Implications for Multi-Site Tractography

The paper evaluates the reproducibility of diffusion, shape, and connectivity metrics across scanners in multi-site tractography studies and finds shape metrics show the highest reproducibility.

CoTRA: a comprehensive R/Shiny framework for transparent bulk and single-cell RNA-seq analysis

CoTRA provides an R/Shiny framework for bulk and single-cell RNA-seq analysis, offering integrated workflows and reproducibility across 49 criteria.

Two dominant axes structure high-dimensional object representations in the human ventral temporal cortex

The paper identifies two dominant axes organizing high-dimensional object representations in the human ventral temporal cortex using fMRI and factor analysis.

Naturalistic narrative listening reveals a core brain network for detecting moral information

The paper identifies a core brain network for detecting moral content using fMRI data from naturalistic narrative listening.

Overcoming artificial structures in resolution-enhanced Hi-C data by signal decomposition and multi-scale attention

This paper introduces SHARP, a method to enhance Hi-C data resolution while avoiding artificial structures by decomposing signals and using multi-scale attention.

SPARKLE: evidence-constrained correction of local RNA leakage in high-resolution spatial transcriptomics

SPARKLE corrects local RNA leakage in spatial transcriptomics by using spatial evidence within each sample, improving cell-type localization and communication inference.

A shared functional organisation underlies vascular disease remodelling

The paper identifies a reproducible functional organisation in vascular remodelling across diseases, showing coordinated rebalancing of biological functions rather than isolated pathway changes.

Anatomy-aware, label-informed approach improves image registration for challenging datasets

This paper introduces a label-informed image registration method that improves registration accuracy for challenging datasets, particularly in developmental studies.

Atlas-scale single-cell analysis beyond in-memory paradigm with scAtlasPy

scAtlasPy enables efficient analysis of large single-cell atlases by using disk-resident computing, achieving high speed and low memory usage compared to existing tools.

CORSICA: Reproducible Suppression of Cochlear Implant Artifacts in EEG evoked by Continuous Speech

CORSICA suppresses cochlear implant artifacts in speech-evoked EEG data using cross-correlation with speech stimuli, outperforming alternative methods with minimal manual intervention.

A framework for analyzing C. elegans neural activity using multi-dimensional hyperbolic embedding

The paper analyzes C. elegans neural activity in response to bacterial stimuli, showing that hyperbolic embedding better describes neural dynamics than Euclidean space.

Fast timescale reparameterization of stable neural manifolds

This paper shows that motor cortex maintains a dynamic library of state eigenvectors that generalize over reach directions, with stable dynamics that drift and jump during repetitive tasks.

multiTEMPTED: Joint Dimensionality Reduction of Longitudinal Multi-omic Data with Modality-Specific Temporal Dynamics

multiTEMPTED jointly reduces dimensionality of longitudinal multi-omic data, capturing modality-specific temporal dynamics and outperforms existing methods in speed and accuracy.

Multi-Stage Singular Value Decomposition for Ultrafast Ultrasound Imaging of Microbubbles

This paper introduces MS-SVD, a multi-stage SVD framework for ultrafast CEUS imaging that improves microbubble detection and vascular contrast by separating nonlinear signals from tissue clutter.

DRUMS - A Flexible New Deep Learning Tool for Precise Cortical Surface Alignment across Individuals

DRUMS improves cortical surface alignment by using deep learning with multimodal supervision, outperforming existing methods in accuracy and reproducibility.

platpy: A spatial-first framework for multi-layer spatial transcriptomic analysis

platpy provides a spatial-first framework for analyzing spatial transcriptomic data, improving efficiency and reducing errors in multi-layer spatial analysis.

On doubting image quality assessment metrics for microscopy virtual staining

This study evaluates image quality assessment metrics for virtual staining in microscopy, finding they poorly translate to biological contexts and suggest application-specific model evaluation is needed.

Global tree encoding of atlas-scale single-cell genomics

The paper introduces MILK, a framework for organizing single-cell genomic data into tree representations, enabling scalable analysis and preserving information across large datasets.

Systematic evaluation of structural connectome thresholding in whole-brain network modelling

This study evaluates how thresholding structural connectomes affects whole-brain network models, finding that optimal sparsity depends on resolution and that tailored thresholding improves model performance.

Modelling interpretable patient-level representationsfrom structured and simple multimodal data

The paper introduces FACTMx, a variational framework that jointly models structured and simple patient-level data to learn interpretable representations, showing improved performance in simulations and applications to cancer and cardiovascular cohorts.

Decision Confidence Neuron in Echo State Network for Continual Evaluation of EEG Motor Imagery Classification Quality

The paper introduces a decision confidence neuron for ESNs to continuously monitor classification confidence in EEG motor imagery tasks, showing it correlates with decision accuracy and discriminates correct from error decisions.

Field-of-view confounding shapes genetic discovery from self-supervised cardiac-imaging phenotypes

This paper shows that self-supervised cardiac-imaging phenotypes can encode field-of-view confounding, which affects genetic discovery and requires correction before association testing.

spatialMET: an open and scalable framework for spatial metabolomics analysis

spatialMET provides a scalable, open-source framework for end-to-end spatial metabolomics analysis, identifying spatial domains and differential metabolites in mouse cancer data.

Weighted sliced inverse regression for scalable supervised dimensionality reduction of spatial transcriptomics data

This paper introduces wSIR, a scalable supervised dimensionality reduction method for spatial transcriptomics data, which retains predictive power of spatial coordinates and is demonstrated on breast cancer and embryonic development datasets.

HiC-LEGO: Biologically Guided High-Resolution 3D Genome Reconstruction Preserves Chromatin Organization at Kilobase Resolution

HiC-LEGO reconstructs high-resolution 3D genomes from Hi-C data, preserving chromatin organization and regulatory interactions at kilobase resolution.

MIND the gap: methodological considerations and guidance for structural MRI similarity network analysis with MIND

The paper provides methodological guidance for using MIND, a structural MRI similarity metric, and demonstrates how input properties affect its accuracy using simulated and observational data.

Generating cell-level protein-expression labels on HE from serial-section immunohistochemistry, with ground-truth-free quality control

The paper introduces CellDF, a method to generate cell-level protein-expression labels from HE images using serial-section IHC data, with quality control metrics that do not require ground truth.

scRep: A Latent-Space Self-Distilled Foundation Model for Single-Cell Representation Learning

scRep is a foundation model for single-cell representation learning that uses self-distillation in latent space, achieving strong performance without task-specific fine-tuning.