dimensionality reduction

Random Matrix Theory-guided sparse PCA for single-cell RNA-seq data

This paper improves PCA for single-cell RNA-seq by using Random Matrix Theory to guide sparse PCA, enhancing reconstruction and classification performance.

Improved spatial memory in a modular network mimicking the prefrontal-thalamo-hippocampal triangular circuit

The paper constructs a neural network model of the prefrontal-thalamo-hippocampal circuit to study how modular division of labor supports spatial memory tasks.

FishFeats: streamlined quantification of multimodal labeling at the single-cell level in 3D tissues

FishFeats provides a streamlined pipeline for quantifying multimodal labeling in 3D tissues, integrating image analysis steps into a user-friendly tool for non-computational users.

PantheonOS: An Evolvable Multi-Agent Framework for Automatic Genomics Discovery

PantheonOS introduces a multi-agent framework for automated genomics analysis, achieving super-human performance in batch correction and gene panel selection across diverse biological tasks.

NeuroMDAVIS: Visualization of single-cell multi-omics data under deep learning framework

NeuroMDAVIS is a multi-omics visualization model that integrates multiple data modalities to generate low-dimensional embeddings preserving local and global structures, outperforming existing methods.

Closed-loop error damping in human BCI using pre-error motor cortex activity

This paper introduces error modulation in motor BCIs using pre-error motor cortex activity to improve cursor control performance during online BCI tasks.

ARCH3D: A foundation model for global genome architecture

This paper introduces ARCH3D, a foundation model for global genome architecture that captures spatial structure and interchromosomal interactions using a novel masked locus modeling task.

Partially Shared Multi-Modal Embedding Learns Holistic Representation of Cell State

APOLLO learns shared and modality-specific information from multi-modal single-cell data to improve understanding of cell state and regulatory mechanisms.

Characterizing MINFLUX imaging performance with DNA origami

The paper characterizes MINFLUX imaging performance over extended durations using DNA origami and corrects residual drift, achieving sub-2 nm precision without loss of site precision.

ABaCo: Addressing Heterogeneity Challenges in Metagenomic Data Integration with Adversarial Generative Models

ABaCo integrates metagenomic data from multiple studies, corrects technical heterogeneity, and outperforms existing methods while preserving biological signals.

The Intrinsic Manifold of Spontaneous Activity Constrains Cortical Responses to Naturalistic Stimuli

This paper shows that the low-dimensional manifold of spontaneous brain activity can represent responses to naturalistic stimuli, using a deep variational autoencoder trained on resting-state fMRI data.

Pixel2Gene enables histology-guided reconstruction and prediction of spatial gene expression

Pixel2Gene improves spatial gene expression reconstruction and prediction by integrating histology images with spatial transcriptomics data.

Skip-Zeros Variational Inference in the Million-Cell Era of Single-Cell Transcriptomics

UNISON is a scalable framework for single-cell transcriptomics that uses skip-zeros variational inference to efficiently analyze large, sparse datasets while preserving interpretability.

Synergistic and Redundant Information Dynamics Exhibit Dissociable Alterations Across Major Psychiatric Disorders

The paper analyzes information dynamics in brain networks of psychiatric disorders, finding distinct patterns of synergy and redundancy reduction or increase in schizophrenia, autism, and ADHD.

Conditioned Graph Reconstruction of Brain Functional Network Connectivity Reveals Interpretable Latent Axes of Sex and Fluid Intelligence

The paper introduces a conditional graph variational autoencoder to model brain functional connectivity, reconstructing network patterns conditioned on sex and fluid intelligence while preserving interpretable latent features.

BEEP Learning: Multi-View Image Decomposition for Massively Multiplexed Biological Fluorescence Microscopy

This paper introduces BEEP learning, a machine learning framework for multi-view fluorescence unmixing that improves discrimination of fluorophores in multiplexed imaging.

Insect Size Matters: Using Image and Dimensions Together Improves Image Classification

This paper improves insect classification by integrating body size information into deep learning models, showing it enhances accuracy, especially in low-data and taxonomic misclassification scenarios.

Transformers Outperform ConvNets for Root Segmentation: A Systematic Comparison Across Nine Datasets

The paper compares Transformer and ConvNet models for root segmentation across nine datasets, finding Transformers significantly outperform ConvNets in accuracy and root-diameter agreement.

Common neural representation between visual perception and imagery in eidetikers

The paper investigates neural substrates shared between eidetic imagery and perception, finding that eidetikers show stronger shared representations in higher-level visual cortex.

An automated image analysis pipeline for wide-field optical redox imaging of patient-derived cancer organoids

The paper develops an automated image analysis pipeline for wide-field optical redox imaging of patient-derived cancer organoids, achieving high accuracy and significantly faster processing than manual methods.

Construction of distinct k-mer color sets via set fingerprinting

The paper introduces a Monte Carlo algorithm to efficiently construct and compress distinct k-mer color sets in microbial genome indexing, reducing memory usage and improving performance.

A NOVEL DEEP LEARNING MODEL, RDBCYCYLEGAN-CBAM FOR LOW-DOSE CT IMAGE DENOISING

This paper introduces RDBCycleGAN-CBAM, a deep learning model for low-dose CT image denoising, achieving higher PSNR and SSIM than existing methods.

Deep Neural Networks to Register and Annotate Cells in Moving and Deforming Nervous Systems

This paper introduces deep neural networks for aligning and annotating cells in moving nervous systems, achieving high accuracy in registration and cell type identification.

High-Resolution 3D Histology of the Murine Kidney Using Synchrotron X-Ray Micro-CT

This paper introduces a synchrotron-based 3D histology method for murine kidneys, achieving high-resolution imaging and quantitative morphometric analysis.

MRI-Based Structural Development of the Human Newborn Hypothalamus

This study examines how prenatal factors affect hypothalamus volume in newborns using MRI data, finding associations with gestational age, postmenstrual age, sex, and maternal smoking.

Improving SCVI for low-count cells through self-supervised augmentation

The paper improves SCVI by addressing low-UMI cell representation issues through self-supervised augmentation, preserving cell identity and variation at lower UMI depths.

Stimulus-Driven Thermodynamic Shifts and Geometric Reorganization in Mouse Primary Visual Cortex

The paper shows that stimulus-driven changes in neuronal connectivity and thermodynamic states lead to more efficient coding in mouse V1 by reorganizing neural activity onto low-dimensional manifolds.

Cross-subject Mapping of Neural Activity with Restricted Boltzmann Machines

The paper develops a method to map neural activity across subjects using Restricted Boltzmann Machines, achieving high accuracy in decoding target features with a Fisher divergence-based training technique.

A hyperparameter benchmark of VAE-based methods for scRNA-seq batch integration

This study benchmarks VAE-based methods for scRNA-seq batch integration, finding dataset-dependent performance trade-offs and recommending hyperparameter settings for optimal results.

tensorOmics: Data integration for longitudinal omics data using tensor factorisation

tensorOmics integrates longitudinal multi-omics data using tensor factorisation to capture coordinated molecular responses across time and omic layers.