Malte D. Luecken

HARP integrates PBMC data from multiple studies to create a reference atlas with consensus annotations and identifies age, sex, and disease-related cellular modules.

The paper integrates multiple single-cell and single-nucleus RNA-sequencing datasets to create a comprehensive Human Kidney Cell Atlas, identifying cell types, states, and disease-associated populations.

The paper introduces scAtlasTb, a framework for building and optimizing single-cell reference atlases, demonstrating its application on three Human Cell Atlas projects.

The paper constructs a healthy human pancreas reference atlas to contextualize disease states and benchmark models.

The paper benchmarks gene expression reconstruction from single-cell latent representations, finding autoencoders perform best at low dimensionality and foundation models depend on decoder architecture.

Cellpin imputes and denoises spatial transcriptomes using a variational autoencoder trained on single-cell RNA-seq, outperforming existing methods without requiring cross-modality alignment.

Human Developmental Cell Atlas integrates ~4.6 million cells/nuclei from 4-22 post-conception weeks into 454 refined cell types and maps 114 spatial niches across 6PCW whole-embryo sections.

The paper presents a longitudinal multimodal benchmarking dataset and competition to evaluate modality prediction methods for gene regulatory inference, showing top models outperform existing approaches.

This paper introduces scTRAM, a framework to quantitatively evaluate single-cell trajectory representations, revealing model-specific trade-offs and structural biases in embeddings.

This study evaluates the Xenium In Situ spatial transcriptomics platform using mouse brain and human breast cancer datasets, comparing it to other technologies and benchmarking analysis tools.

geneRNIB provides a benchmark for gene regulatory network inference, showing simpler models often outperform complex ones in perturbation-informed metrics.

Spapros selects optimal probe sets for targeted spatial transcriptomics to improve cell type identification and spatial variation detection across multiple datasets.

ArchMap provides a no-code tool for reference-based analysis of single-cell datasets, enabling automated mapping, annotation, and collaboration across reference atlases.

The paper creates a comprehensive human retina cell atlas using single-cell multi-omics data, revealing over 110 cell types and their gene regulatory networks, and enables fine-mapping of GWAS and eQTL variants.

A NeurIPS 2021 benchmarking competition evaluated over 2600 submissions from 280 competitors across three multimodal single-cell integration tasks using a purpose-built benchmarking dataset and cloud evaluation pipeline.

BBKNN, Scanorama, and scVI lead complex scRNA-seq integration in a 38-setup, nine-task, 14-metric benchmark; highly variable gene selection helps, whereas scaling favors batch removal over bio-conservation.

This paper introduces scArches, a deep learning method for mapping query datasets onto reference single-cell atlases using transfer learning, effectively reducing batch effects and preserving biological information with fewer parameters.

The paper examines ACE2 expression in kidney cells in relation to RAAS blockers and age/sex, finding age-related differences and associations with drug use.

This study uses single-cell RNA-seq and computational methods to analyze macrophage states in COPD patients, revealing transcriptional changes and metabolic shifts associated with the disease.

Across 1,176,683 airway and lung cells from 22 datasets, ACE2 rises with age, in males and in smokers' multiciliated cells, and is lowest in young children.

Benchmark-derived recommendations per single-cell RNA-seq step (scran, ComBat, UMAP, Louvain, Slingshot/PAGA, MAST/limma), guidance on which processed data suit each analysis, and a mouse intestinal case study.