Giovanni Palla

The paper presents a multimodal perturbation atlas using CRISPR knockouts in A549 cells, combining microscopy, phase imaging, and scRNA-seq to assess phenotypic resolution.

This paper evaluates Tabular Foundation Models against specialized models for predicting cellular perturbation responses across multiple biological scales and tasks.

The paper introduces TranscriptFormer, a generative model trained on 112 million cells across 12 species, demonstrating cross-species cell type classification and disease state prediction.

Phoenix predicts spatially resolved single-cell gene expression from histology, identifying new biomarkers and validating them across cancers and treatment regimens.

SigSpace uses an LLM to interpret drug response signatures from cancer cell lines, generating summaries that outperform random baselines in matching to drug identity or mechanism of action.

moscot unifies optimal transport for temporal, spatial and spatiotemporal single-cell mapping, using multimodal cost spaces and low-rank solvers to scale to 1.7 million mouse embryo cells.

GraphCompass provides graph analysis methods to compare cell spatial organization across conditions using spatial omics data.

Pretrained on 57 million dissociated and 53 million spatial human and mouse cells, a 49.3M-parameter transformer predicts niche, region, neighborhood composition and density, and carries spatial labels onto scRNA-seq.

SpatialData, a Zarr/OME-NGFF-based format plus Python library, unifies images, labels, points, shapes and tables across spatial omics platforms with coordinate alignment, lazy loading and cross-modal aggregation.

This paper describes OME-Zarr, a cloud-optimized bioimaging file format developed through community collaboration to improve scalability and data access.

This paper introduces UPMM to model single-cell dynamics, showing it outperforms existing methods on real developmental data.

Squidpy extends Scanpy/AnnData with spatial neighbor graphs and a lazy-loading image container, enabling spatial statistics, ligand-receptor tests, image features, and ~10-fold faster neighborhood enrichment than Giotto.

The paper introduces GRaNIE and GRaNPA, tools for inferring and evaluating enhancer-mediated gene regulatory networks in macrophages, identifying PURA as a regulator of pro-inflammatory polarization.

The paper compares latent factor models on scRNA-seq data to identify disease-related gene signatures in autoimmune disease cell subsets.

diffTF quantifies differential transcription factor activity and classifies TFs as activators or repressors using multi-omics data, identifying known and novel TFs in two case studies.