Fabiola Curion

This paper evaluates Evo 2's in-context learning capabilities on genomic sequences, finding strong performance on short sequences but degradation with length, and no benefit from model scaling.

This paper creates a human endoderm organoid cell atlas by integrating single-cell transcriptomes from multiple organoid types to compare cell states with primary tissues and disease models.

This paper introduces a VAE-based model for integrating multimodal single-cell data, showing it can map queries across modalities and improve data imputation and biological discovery.

scDECAF identifies cell types, states, and programs in single-cell data by learning gene set representations, showing effectiveness in disease and immune response contexts.

hadge combines hashing and genotype-based deconvolution methods to recover previously discarded cells in single-cell data.

Panpipes provides computational workflows for multiomic single-cell and spatial transcriptomic analysis, enabling quality control, preprocessing, integration, clustering, and reference mapping.

The paper develops a method to generate cerebellar organoids from human iPSCs and shows they recapitulate key aspects of cerebellar development through single-cell profiling.