dimensionality reduction

Multi-modal data integration for machine learning applications

The paper introduces momics, a framework for integrating multi-omics genomic data, showing improved compression, speed, and performance in deep learning for chromatin accessibility prediction.

GPU-accelerated, self-optimizing processing for 3D multiplexed iterative RNA-FISH experiments

This paper introduces a GPU-accelerated framework for processing 3D multiplexed RNA-FISH data, quantifies information loss from axial sampling, and improves post-mortem sample quality.

Toggle-Untoggle: A cell segmentation tool with an interactive user verification interface.

Toggle-Untoggle improves cell segmentation accuracy by combining automated deep learning with an interactive interface for manual refinement without coding.

Joint multi-omics profiling of brain and body health in aging

This study identifies two brain-body health axes using multi-omics data from aging populations, linking metabolic features to cognitive deficits.

On ARGs, pedigrees, and genetic relatedness matrices

This paper introduces branch relatedness and efficient algorithms for computing branch GRMs using tree sequences, enabling large-scale genomic analysis.

Nanoscale spatial-omics via contrastive embedding of single-molecule localisation data

This paper introduces a contrastive learning framework for nanoscale spatial-omics, enabling analysis of single-molecule localisation data to study protein architecture and its changes under various conditions.

Towards Personalized Epigenomics: Learning Shared Chromatin Landscapes and Joint De-Noising of Histone Modification Assays

This paper introduces DecoDen, a method that learns shared chromatin landscapes and de-noises histone modification assays across multiple tissues.

Transforming Hyperspectral Data into Insight: The DREAM Approach for Pathology

The paper introduces DREAM, a method to enhance spectral visualization of autofluorescence microscopy data, improving colorfulness, sharpness, and consistency in unstained tissue samples.

SpaDiff: Denoising for Sequence-based Spatial Transcriptomics via Diffusion Process

SpaDiff improves spatial transcriptomics by reversing RNA molecule displacement, enhancing spatial specificity and data integrity for downstream analyses.

Understanding the high-order network plasticity mechanisms of ultrasound neuromodulation

This paper develops a whole-brain model to study how transcranial ultrasound stimulation (TUS) induces functional changes in the brain, identifying two mechanisms that predict plasticity-driven effects.

iBrAVE: a unified framework for 3D interactive and integrative analysis of brain atlas data across modalities and scales

iBrAVE integrates multimodal and multiscale brain atlas data across species, enabling interactive analysis of molecular, structural, and functional information.

Multi-immersion Oblique Plane Microscope (miOPM): A reconfigurable platform for high-resolution Light-Sheet Fluorescence Microscopy.

The paper introduces miOPM, a reconfigurable LSFM platform that enables high-resolution imaging across multiple spatial scales and refractive indices.

A Scalable Toolkit for Modeling 3D Surface-based Brain Geometry

The paper introduces a scalable toolkit for 3D surface-based brain analysis, demonstrating its utility in mapping subcortical shape variations across 21 international samples.

Return of the GEDAI: Unsupervised EEG Denoising based on Leadfield Filtering

GEDAI improves EEG denoising by removing artifacts without clean references, outperforming existing methods in simulations and enhancing neurobehavioral predictions.

Improved interpretability in LFADS models using a learned, context-dependent per-trial bias

This paper improves the interpretability of LFADS models by introducing a per-trial bias, enabling simpler contextual dynamics and better analysis of neural activity in simulated and real data.

SLUR(M)-py: A SLURM Powered Pythonic Pipeline for Parallel Processing of 3D (Epi)genomic Profiles

SLUR(M)-py is a Python pipeline using SLURM for parallel processing of multi-omic epigenomic data, demonstrating improved speed and completeness over existing HPC pipelines.

dFCExpert: Learning Dynamic Functional Connectivity Patterns with Modularity and State Experts

dFCExpert learns dynamic functional connectivity patterns from fMRI data using modularity and state experts, showing superior performance on large-scale datasets.

A knowledge-guided approach to recovering important rare signals from high-dimensional single-cell data

This paper introduces a framework that guides dimensionality reduction using knowledge-derived genes to recover rare but important signals in single-cell data, demonstrating its effectiveness in identifying rare cell subtypes.

Domain shifts in infant brain diffusion MRI: Quantification and mitigation for fODF estimation

This study evaluates deep learning models for fODF estimation in infant brain MRI, finding that U-Net performs best with fewer gradient directions, while MLP and Transformer improve with more directions.

Multimodal investigation of the neurocognitive deficits underlying dyslexia in adulthood

This study investigates neurocognitive deficits in dyslexia using multimodal MRI and behavioural assessments, finding that combining data improves classification accuracy compared to single-modality approaches.

NOVA: a novel R-package enabling multi-parameter analysis and visualization of neural activity in MEA recordings

NOVA is an R-package for analyzing MEA data, offering dimensionality reduction, clustering, and trajectory mapping, revealing network burst duration as a key activity metric.

Efficient integration of spatial omics data for joint domain detection, matching, and alignment with stMSA

This paper introduces stMSA, a deep learning model for integrating spatial omics data, improving domain detection, matching, and alignment across tissue slices.

xIV-LDDMM Toolkit: A Suite of Image-Varifold Based Technologies for Representing and Mapping 3D Imaging and Spatial-omics Data Simultaneously Across Scales

The paper introduces xIV-LDDMM, a toolkit for cross-scale and cross-modality mapping of 3D imaging and spatial-omics data using image-varifold measures.

Aposematic color patterns are the dominant axis of phenotypic diversification in Nymphalid butterflies

The paper uses computer vision to analyze wing color patterns in Nymphalid butterflies, finding that aposematism is the dominant axis of phenotypic diversification.

mcRigor: a statistical method to enhance the rigor of metacell partitioning in single-cell data analysis

mcRigor improves metacell partitioning reliability by detecting heterogeneous metacells and optimizing hyperparameters for single-cell RNA-seq and multiome data.

Integrate and generate single-cell proteomics from transcriptomics with cross-attention

This paper introduces scProca, a deep generative model that integrates and generates single-cell proteomics data from transcriptomics using cross-attention mechanisms.

Predicting chromatin conformation contact maps

The paper predicts missing chromatin conformation contact maps across cell types and assays, exploring how 3D genome architecture varies.

A Framework for Compressive On-chip Action Potential Recording

This paper introduces a compression framework for neural recording systems that reduces bandwidth demands while preserving spike detection accuracy.

Unsupervised machine learning reveals temporal components of gene expression in HeLa cells following release from cell cycle arrest

This study uses unsupervised machine learning to identify temporal gene expression components in HeLa cells after cell cycle release, revealing oscillatory and transient patterns linked to cell cycle phases and cancer-related processes.

st-DenseViT: A Weakly Supervised Spatiotemporal Vision Transformer for Dense Prediction of Dynamic Brain Networks

This paper introduces st-DenseViT, a weakly supervised spatiotemporal vision transformer that generates 4D dynamic brain networks from fMRI data, showing significant differences between schizophrenia patients and healthy controls.