Britta Velten

Statistical tests for bivariate spatial association across multi-omics data with disjoint coordinates

Autocorrelation-aware bivariate Moran's I, bivariate Gaussian process tests and generalized additive models (R package sbivar) test cross-modality colocalisation on disjoint coordinates; many existing permutation tests inflate false positives.

msBayesImpute: A Versatile Framework for Addressing Missing Values in Biomedical Mass Spectrometry Proteomics Data

msBayesImpute improves missing value handling in proteomics data by integrating Bayesian factorization and probabilistic dropout models, outperforming existing methods in imputation and normalization.

Genome-scale mapping of variant, enhancer and gene function in primary human CD4+ T cells

The paper maps variant, enhancer, and gene function in primary human CD4+ T cells by combining Perturb-seq of regulatory elements and genes, identifying CRE-gene pairs and linking variants to disease-related programs.

Comparison of Interventional Causal Structure Learning Algorithms for Gene Regulatory Network Inference

This paper benchmarks causal structure learning algorithms for gene regulatory network inference using synthetic and CRISPR perturbation data, finding reliable performance on synthetic data but limitations on real data due to noise and scalability.

A genome-scale single cell CRISPRi map of trans gene regulation across human pluripotent stem cell lines

The paper presents a genome-scale CRISPRi perturbation map in human pluripotent stem cells using single-cell RNA sequencing to study trans gene regulation across diverse genetic backgrounds.

Guide assignment in single-cell CRISPR screens using crispat

The paper introduces crispat, a tool for guide assignment in single-cell CRISPR screens, showing differences in results across strategies.

Multicellular factor analysis of single-cell data for a tissue-centric understanding of disease

The paper introduces a framework for analyzing cross-condition single-cell atlases to identify multicellular disease programs, validated across multiple datasets.

FISHFactor: A Probabilistic Factor Model for Spatial Transcriptomics Data with Subcellular Resolution

FISHFactor is a probabilistic factor model for spatial transcriptomics data that improves accuracy by modeling single molecule resolution data and sharing information across cells.

Spatially resolved single-cell multiomics map of human trophoblast differentiation in early pregnancy

This study creates a spatially resolved single-cell multiomics atlas of human trophoblast differentiation in early pregnancy, revealing key regulatory programs and cell-cell communication events involved in EVT invasion and arterial transformation.

Identifying temporal and spatial patterns of variation from multi-modal data using MEFISTO

MEFISTO enables spatio-temporal dimensionality reduction and pattern separation in multi-modal genomic data, revealing conserved and diverged developmental programs and microbiome heterogeneity.

Developmental gene expression differences between humans and mammalian models

The paper identifies developmental gene expression differences between humans and mammalian models, finding that half of human genes differ from their mouse orthologs in temporal trajectories, particularly for disease-related genes.

MOFA+: a probabilistic framework for comprehensive integration of structured single-cell data

MOFA+ integrates structured single-cell multi-modal data by modeling group and view structures, revealing coordinated gene expression and epigenetic changes during mouse gastrulation.

Multi-Omics factor analysis - a framework for unsupervised integration of multi-omic data sets

MOFA, a sparse Bayesian factor model with mixed likelihoods, separates shared from modality-specific multi-omic variation, recovering IGHV, trisomy 12 and oxidative-stress axes in 200 leukaemia samples.