dimensionality reduction

NicheTrans: Spatial-aware Cross-omics Translation

NicheTrans translates single-omics data into spatial multi-omics insights by integrating spatial and multi-modal information, outperforming existing methods in disease and tissue analysis.

Coding agents author interpretable single-cell embedding models from the literature

This paper introduces coding agents that generate interpretable single-cell embeddings from literature-based gene programs without training data or databases.

Hemodynamic Signals Reshape Biological Inference in Widefield Calcium Imaging

This study shows hemodynamic correction significantly alters biological interpretations of cortical dynamics in calcium imaging by reducing global connectivity and changing network structure.

Assessing tensor decomposition quality of immune profiling data from a dictionary learning perspective

The paper introduces metrics for assessing tensor decomposition quality in immune profiling data using a dictionary learning framework.

Dendritic Wave Recurrent Neural Networks

This paper introduces the Dendritic Wave Recurrent Neural Network (DW-RNN), which improves upon wave recurrent neural networks by adding dendritic input integration, achieving better performance and stability on sequential tasks.

Lossless compression of k-mer matrices enabling random row access

The paper introduces kmcomp, a lossless compression method for k-mer matrices that enables fast random access, achieving significant storage savings with minimal query overhead.

Modeling patient tissues at molecular resolution with Eva

Eva is a foundation model for tissue imaging data that learns multi-scale spatial representations and performs various tasks like cross-modal inference and patient stratification.

ROOTQUANT: AUTOMATED ROOT TRAIT QUANTIFICATION FROMMINIRHIZOTRON IMAGES USING DEEP LEARNING

RootQuant uses deep learning to automatically quantify root traits from minirhizotron images, achieving high accuracy across species without pixel-level annotations.

Night-to-night sleep EEG variability over one year

This study examines night-to-night variability in sleep patterns using EEG data over one year, finding that macrostructural patterns are relatively stable while temporal alignment varies significantly.

FEATMAP: Targeted Correction of Acquisition Signatures Harmonizes Medical Foundation Model Embeddings and Enables Robust Task Generalization

FEATMAP corrects acquisition signatures in medical foundation model embeddings to improve cross-condition similarity and task generalization without retraining.

Disentangling Cephalopod Chromatophores Motor Units with Computer Vision

The paper uses computer vision and dimensionality reduction to analyze chromatophore motor control in cephalopods, revealing the organization of motor units and their neural control.

ScaleSurfer: multi-scale anatomical segmentation and parcellation of the human brain

ScaleSurfer rapidly estimates whole-brain morphometric features using a multi-scale vision transformer, achieving speeds up to five orders of magnitude faster than traditional methods.

Extracting host-specific developmental signatures from longitudinal microbiome data

The paper introduces PARAFAC2 to model subject-specific temporal patterns in longitudinal microbiome data, outperforming CP in capturing individual variations and enabling replicable pattern discovery.

OMIDIENT: Multiomics Integration for Cancer by Dirichlet Auto-Encoder Networks

OMIDIENT is a multi-omics integration method that uses Dirichlet auto-encoders to improve clustering, classification, and data reconstruction across cancer types.

SpliSync: Genomic language model-driven splice site correction of long RNA sequencing reads

SpliSync improves RNA long-read alignment precision using a genomic language model, enhancing splice site correction but with limited impact on transcript reconstruction.

Differentiable Design for Morphogenesis I: Simulation and Simulacra

This paper introduces waxMorph, a differentiable framework for simulating and reconstructing 3D morphogenesis, showing it can infer shape programs and trajectories from biological data.

Connectome-scale self-supervised representation learning reveals neuronal organization beyond canonical labels

This paper develops a self-supervised framework for learning structure-connectivity representations in dense connectomes, revealing neuronal organization beyond canonical labels.

Neural representations of speech production in neocortical and cerebellar regions

This study investigates neural representations of speech production in both cortical and cerebellar regions using fMRI and finds that cerebellar representations align more closely with parietal operculum than primary motor cortex.

Prefrontal representations can specialize rather thangeneralize with experience

The paper shows that prefrontal cortex representations become specialized for specific task demands rather than generalizing with experience.

CodonBERT and ESM-2 Embedding Spaces Share an Evolutionarily Conserved Paired Geometry Encoding Synonymous Codon Information

The paper shows that CodonBERT and ESM-2 embeddings encode synonymous codon information through a shared geometric relationship across species.

Scalable multi-group nonnegative spatial factorization for spatial genomics data with cell-type heterogeneity

The paper introduces smNSF, a method that integrates spatial and cell-type information to uncover interpretable spatial gene patterns in spatial transcriptomics data.

A uniform tissue-clearing framework and mesoSPIM-ultra enable cm-scale single-neuron tracing

The paper introduces a tissue-clearing framework and mesoSPIM-ultra system for large-scale single-neuron tracing in cleared mouse tissue.

Learning-related population dynamics in right and left dorsal premotor cortex during typing skill acquisition

The paper shows that during typing skill acquisition using a BCI, neural population dynamics in the dorsal premotor cortex become more compact, with left hemisphere showing enhanced firing-rate modulation and generalization.

Segmentation-guided photon pooling enables robust single cell analysis and fast fluorescence lifetime imaging microscopy

This paper introduces a photon pooling technique for faster single-cell FLIM analysis without machine learning, achieving single-cell metabolic data every second and 15x faster mosaic acquisition.

Riemannian Metric Learning for Alignment of Spatial Multiomics

This paper introduces MGW, a metric-learning framework for aligning spatial multi-omics data across different modalities, showing improved alignment over existing methods.

Undersampling techniques for non-linear chemical space visualization

This study examines how undersampling techniques affect hyperparameter selection for non-linear dimensionality reduction methods in chemical space visualization, showing reduced computational costs and improved local structure preservation.

NeuroVLM: A generative vision-language framework for human neuroimaging

NeuroVLM generates and evaluates vision-language models for human neuroimaging using 30,826 neuroimage-text pairs to perform tasks like text-to-image generation and image-to-text interpretation.

CPS: Mapping Physical Coordinates to High-Fidelity Spatial Transcriptomics via Privileged Multi-Scale Context Distillation

CPS maps physical coordinates to high-fidelity spatial transcriptomics using a multi-scale context distillation strategy, achieving state-of-the-art imputation and denoising performance.

A Simple Subject Independent Channel Selection in EEG for Motor Imagery Task

This paper introduces a subject-independent EEG channel selection method for motor imagery tasks, showing improved performance over existing methods with fewer channels.

Shape Factor Analysis as a Quantitative Framework for Assessing Spheroid and Organoid Morphology and Invasiveness

This paper develops a shape factor analysis method to quantify spheroid and organoid morphology, showing it provides more reliable morphology assessments than standard shape descriptors.