Single-cell and spatial transcriptomics

Principal Genes: A PCA-based approach to highly variable genes selection for scRNA-Seq analysis

This paper introduces a PCA-based method for selecting highly variable genes in scRNA-Seq data, showing it is comparable or better than dispersion-based methods and faster.

Learning Discrete Cell and Niche Codes from Spatial Transcriptomics Using Dual Residual Vector Quantization

SQUINT learns discrete cell and niche codes from spatial transcriptomics data using residual vector quantization, outperforming baselines in identification and cross-section integration.

A modular architecture of human cellular aging across organs and diseases

This study identifies modular patterns of cellular aging across organs and diseases using single-cell transcriptomics and proteomics, revealing conserved and organ-specific aging programs.

Accurate quantification of spliced and unspliced transcripts for single-cell RNA sequencing with tidesurf

The paper introduces tidesurf, a tool for quantifying spliced and unspliced transcripts in scRNA-seq, showing improved accuracy over velocyto for 5' chemistry data.

scE2TM improves single-cell embedding interpretability and reveals cellular perturbation signatures

scE2TM improves single-cell embedding interpretability by integrating external knowledge, achieving better clustering and revealing biological perturbation signatures across datasets.

RobustCell: A Model Attack-Defense Framework for Robust Transcriptomic Data Analysis

RobustCell evaluates attack-defense methods in single-cell and spatial transcriptomic data analysis, showing attacks can impair clustering and annotation, while defenses can protect model performance.

Morphodynamic domains enable integration of live morphometrics and spatial transcriptomics

The paper introduces a framework to integrate spatial transcriptomics with live morphometrics to infer gene regulatory networks and validate gene function in Drosophila thorax development.

Benchmarking cell-type deconvolution in cross-platform transcriptomic data

This study benchmarks cell-type deconvolution methods across cross-platform transcriptomic data, finding that log-normal regression methods like SpatialDecon perform most reliably.

MuSpAn: A Toolbox for Multiscale Spatial Analysis

MuSpAn, a Python library of over 100 spatial methods, quantifies multiscale tissue organisation and maps cancer-associated fibroblast subtypes to distinct immune niches in mouse colorectal tumours.

PIANO: Probabilistic Inference Autoencoder Networks for multi-Omics enables robust generative modeling of gene expression and scales single-cell integration to 100 million cells

PIANO integrates single-cell data across species and modalities, models gene expression, and scales to 100 million cells for drug perturbation analysis.

AdaGeneBudget: Cell-Adaptive Gene-Token Allocation for Efficient Single-Cell Foundation Models

AdaGeneBudget reduces gene-token usage in single-cell foundation models while preserving annotation utility and fine-grained cell identities.

scOPE identifies which driver-associated expression programs transfer from bulk tumors to single cells

scOPE transfers driver-associated expression programs from bulk tumors to single cells, identifying selective programs with high predictive accuracy.

Reliable single-cell perturbations explain and improve model performance

The paper evaluates perturbation reliability in single-cell data and shows that using reliable perturbations improves model performance and reduces training data needs.

Spatial multi omics enables single cell transcriptome metabolome inference

CHIMERA infers single cell metabolomes from scRNA-seq data using spatially paired multi-omics data, revealing metabolic reprogramming and subclusters in liver disease models.

Virtual-cell verification enables self-auditing AI discovery for immune rejuvenation

The paper introduces a framework for self-auditing AI in drug discovery, using verifiers to validate predictions and adjust objectives based on experimental results.

A Time-Resolved Single-Cell Atlas Reveals Infection-Status, Age-, and Sex-Dependent Immune Responses Drive Viral Disease Severity

This study uses time-resolved single-cell RNA sequencing in mice to show age and sex influence immune responses and viral disease severity through altered gene expression and inflammation.

Moirai: single-cell trajectory inference grounded in gene-level expression dynamics

Moirai infers cell trajectories by focusing on gene-level expression dynamics, improving detection of branching points in developmental processes.

AnchorR: A QuPath and R interface for collaborative exploration of spatial transcriptomics and histology

AnchorR improves spatial alignment between H&E images and CosMx data by combining expert-defined landmarks, reducing landmark error from 121.5 m to 14.6 m.

Impact of Data Quality on Deep Learning Prediction of Spatial Transcriptomics from Histology Images

This study examines how data quality affects deep learning predictions of spatial gene expression from histology images, showing that improving data quality can enhance predictive performance.

Spatial and Multi-Omics Analysis of Human Breast Cancer Reveals the Spatiotemporal Dynamics of Basal Layer Disruption

The paper uses spatial and multi-omics data to study breast cancer progression, identifying key molecular changes and potential biomarkers associated with malignancy risk.

stCNASim: Allele-aware spatial RNA-seq simulator enables systematic benchmarking of copy number inference

stCNASim is a spatial RNA-seq simulator that benchmarks copy number inference methods in spatial transcriptomics.

Systematic assessment of the biological impact of cellular deconvolution on downstream analyses of disease transcriptomes

This study assesses how cellular deconvolution affects transcriptomic analyses of Alzheimer's disease, showing that correction reshapes differential gene expression and pathway enrichment, revealing more consistent cell intrinsic signatures.

bgnorm: A Generative Statistical Framework for Background Correction, Normalisation, and Quality Control in Multiplex Spatial Proteomics

bgnorm improves signal separation and marker positivity classification in multiplex spatial proteomics by addressing technical variation through background correction, normalization, and quality control.

MSGPCA: Multi-Slice Graph PCA for replicate-aware Spatial Omics analysis

MSGPCA improves spatial omics analysis by decomposing molecular variation into shared and slice-specific factors, enhancing the detection of tissue structure and microenvironmental variation.

Mapping disease traits onto spatiotemporal domain landscapes through scalable multi-sample integration of spatial transcriptomics

STUltra integrates spatial transcriptomics data to map disease traits onto spatiotemporal tissue landscapes, outperforming existing methods in scalability and speed.

Double Machine Learning with Multi-Gene Shared Backgroundfor Causal Inference in Single-Cell Data: Grouping Deviation Follows a Random Walk and the Accuracy-Compute Trade-Off

The paper introduces a Randomized Partition Strategy for efficient double machine learning in single-cell data, showing accuracy loss follows a random walk and compute savings are predictable.

RareCapsNet: An explainable capsule networks enable robust discovery of rare cell populations from large-scale single-cell transcriptomics

RareCapsNet identifies rare cell populations in single-cell RNA-seq data using capsule networks, outperforming existing methods in specificity and selectivity.

A Principled Statistical Framework for Analyzing Spatial Patterns in Spatially Resolved Multi-Omics

This paper introduces STORM, a statistical framework for analyzing spatial patterns in multi-omics data, demonstrating reliable performance in detecting spatial structures and supporting study design decisions.

scROMA: batch-aware pathway-activity inference and a ground-truth simulation framework for single-cell transcriptomics

scROMA infers pathway activity from single-cell transcriptomics data, accounting for batch effects, and provides a simulation framework for benchmarking.

Accurate prediction in reconstructed spatial transcriptomes does not ensure valid biological discovery

The paper introduces TIDEST, a framework that improves statistical inference in reconstructed spatial transcriptomes by correcting for prediction errors and spatial confounding, leading to more accurate biological interpretations.