dimensionality reduction

scPRINT-2: Towards the next-generation of cell foundation models and benchmarks

scPRINT-2 is a single-cell foundation model pre-trained on 350 million cells across 16 organisms, achieving state-of-the-art performance in expression denoising, cell embedding, and cell type prediction.

Bridging Glucose Metabolism and Intrinsic Functional Organization of the Human Cortex

This study links intrinsic functional brain organization to energy expenditure patterns, showing that functional connectivity dimensions explain significant variance in metabolic maps.

Latent-centric Isotropic Resolution Enhancement for Expansion Microscopy Imaging via Neural Compression and Self-supervised Learning

This paper introduces a self-supervised framework for isotropic super-resolution and compression in expansion microscopy, enabling efficient storage and analysis of whole-organ datasets.

SCALPEL: A pipeline for processing large-scale spatial transcriptomics data

SCALPEL processes large-scale spatial transcriptomics data with improved 3D segmentation, doublet detection, and spatial alignment, showing better performance than previous methods.

HiCInterpolate: 4D Spatiotemporal Interpolation of Hi-C Data for Genome Architecture Analysis.

HiCInterpolate interpolates Hi-C contact matrices between timestamps to generate high-resolution 3D genome structures, preserving key chromatin features.

Multimodal Connectivity-based Cortical Segmentation with Graph Neural Networks

This paper evaluates Graph Neural Networks for brain cortex segmentation using structural MRI and diffusion MRI data, finding GAT models perform competitively with non-graph methods.

In Vivo Surface Reconstruction of Three Signal-defined Intracortical Layers Using 5T 3D FLAIR MR Imaging

This paper develops BrainMLSR, an automated framework to reconstruct three intracortical signal-defined layers from 5T 3D FLAIR MRI, showing high agreement with manual annotations and revealing regional morphometric differences in epilepsy.

Identifying unmeasured heterogeneity in microbiome data via quantile thresholding (QuanT)

This paper introduces QuanT, a non-parametric method to identify unmeasured heterogeneity in microbiome data, improving downstream analysis accuracy.

Reaction-enabled, highly sensitive super-resolution imaging

This paper introduces RIED, a super-resolution imaging method using electrochemiluminescence, achieving high sensitivity and resolution for intracellular structures.

Extracting large-scale neural activity with Suite2p

Suite2p processes large-scale neural recordings, outperforming existing methods in motion correction and cell detection.

A lightweight, ultrafast and general embedding framework for large-scale spatial omics data

This paper introduces SpaLP, a framework for spatial omics data that outperforms existing methods in efficiency and accuracy across multiple tasks.

Neuronavigation-free and MRI-free Localization of Deep Brain Targets

This paper introduces an MRI-free and neuronavigation-free method for localizing deep brain targets using scalp measurements, achieving a mean localization error of 3.82 mm.

ChromBERT-tools: A versatile toolkit for context-specific embedding of transcription regulators across different cell types

ChromBERT-tools generates context-specific embeddings of transcription regulators for different cell types to enhance regulatory modeling and interpretation.

Robust quality assessment of cryo-EM maps, tomograms and micrographs by statistics-based local resolution estimation

The paper introduces a statistics-based method for local resolution estimation in cryo-EM maps, tomograms, and micrographs to improve quality assessment and reconstruction.

Phenotypic Signatures of CTNNB1 Syndrome: Longitudinal Neuropsychological Outcomes and Three-Dimensional Facial Morphology

This study provides longitudinal cognitive and facial morphology data for CTNNB1 syndrome, showing stable global functioning and significant facial shape associations with behavioral issues.

Joint Modeling of Cellular Heterogeneity and Condition Effects with scPCA in Single-Cell RNA-Seq

This paper introduces scPCA, a dimensionality reduction framework that integrates cellular heterogeneity and condition effects in scRNA-seq data, revealing transcriptional changes across conditions and cell types.

Age-related changes in proprioception are of limited size, outcome-dependent and task-dependent.

This study assesses age-related changes in upper limb proprioception across multiple tasks and finds that effects are variable, task-dependent, and often require large samples to detect.

2D HSQC-derived "dark forest" image with enhanced local resolution via first derivative processing-logarithmic cosine transformation (FDP-LCT): Demonstration on per-O-ethylated kappa- and iota-carrageenans

This paper introduces an algorithm to enhance the resolution of 2D HSQC NMR spectra for better structural analysis of polysaccharides.

spRefine Denoises and Imputes Spatial Transcriptomics with a Reference-Free Framework Powered by Genomic Language Model

spRefine denoises and imputes spatial transcriptomics data using a reference-free deep learning framework, improving data integration and revealing new aging-related biological insights.

Symmetric Fusion of fMRI and EEG for Spectrally Resolved Functional Neuroimaging

FSINC improves EEG-fMRI source imaging by integrating frequency-resolved EEG with fMRI via a data-driven neurovascular coupling model, achieving better spatial and temporal accuracy in both simulated and real data.

Domain Adaptation Improves Cross-site Classification of First-episode Schizophrenia from Multimodal Neuroimaging Data

This study evaluates methods to improve cross-site classification of first-episode schizophrenia using multimodal neuroimaging data, finding that domain adaptation strategies partially mitigate site-related performance degradation.

AMULETY: A Python package to embed adaptive immune receptor sequences

AMULETY generates language model embeddings for adaptive immune receptor sequences using pre-trained models, enabling users to leverage different models for downstream applications.

Geometric Fragility in Alzheimer's Disease: Probing the Loss of Hippocampal Hierarchical Abstraction via Contrastive Point Cloud Modeling

This study uses contrastive point cloud modeling to show that Alzheimer's disease involves a loss of hippocampal hierarchical abstraction, revealing a linear geometric degradation correlated with cognitive decline.

Integrating Functional Transdiagnostic Dimensions of Psychopathology with Cortical Organization

The paper identifies a latent variable linking resting-state functional connectivity and psychopathology in children and adolescents, revealing a neurotypical-neurodivergent continuum.

Optimality with room to vary: stiff and sloppy modes in a sensory population

The paper analyzes retinal ganglion cell activity to show how sensory population codes balance normative optimality with variability through stiff and sloppy modes.

An Integrated Single-Cell and Epigenomic Resource for Comparative Analysis of the Basal Ganglia

The paper presents an integrated single-cell and epigenomic resource for exploring basal ganglia datasets across species, enabling multi-omics and comparative analysis.

MultiGEOmics: Graph-Based Integration of Multi-Omics via Biological Information Flows

MultiGEOmics integrates multi-omics data using graph-based methods to model cross-omics regulatory signals and handle missing data, showing strong performance across cancer and Alzheimer's datasets.

Seed fossil record of Solanaceae revisited

Morphometric and micro-CT analysis of 110 fossil and 354 extant Solanaceae seeds identifies 10 specimen-backed fossil species, including four new genera and a fossilized elaiosome, spanning Eocene to Holocene deposits.

The progression of sex differences in brain networks across the lifespan

The paper analyzes sex differences in brain networks across the lifespan using a new AI-based method, finding minimal differences in early childhood, pronounced in adulthood, and diverging in later life.

Dimensionality Reduction with Strong Global Structure Preservation

This paper introduces two objectives to improve dimensionality reduction by preserving both global and local structure, achieving better performance on biomedical datasets.