stFormer: a foundation model for spatial transcriptomics
Cao, S.; Yuan, Y.
Show abstract
Recent foundation models for single-cell transcriptomics data generate informative, context-aware representations for genes and cells. The Spatial Transcriptomics (ST) data offer extra positional insights, which were not considered by these single-cell models. Here, we introduce stFormer, a transformer model tailored for ST data. stFormer employs the cross-attention module to incorporate spatial ligand genes into the transformer encoder of single-cell transcriptomics. To unify different ST technologies with trade-off between resolution and gene coverage, we propose a biased cross-attention method that enables the model to do learning with single-cell resolution on whole-transcriptome but low-resolution Visium data. We collected human Visium datasets from a public ST database and performed cell type deconvolution, generating ~4.1 million pretraining samples. As a foundation model for ST data, stFormer improved performance upon the state-of-the-art single-cell foundation model, scFoundation, across a variety of tasks, including cell clustering, batch effect correction, cell type prediction, and gene function prediction. stFormer also revealed intercellular ligand-receptor signaling responses via in silico perturbation.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- OmicVerse: A single pipeline for exploring the entire transcriptome universe 98%
- Spatially informed clustering, integration, and deconvolution of spatial transcriptomics with GraphST 97%
- INSTINCT: Multi-sample integration of spatial chromatin accessibility sequencing data via stochastic domain translation 97%
Similar papers in this journal
- Identifying perturbations that boost T-cell infiltration into tumours via counterfactual learning of their spatial proteomic profiles 94%
- Ultra-fast Prediction of Somatic Structural Variations by Reduced Read Mapping via Pan-Genome k-mer Sets 93%
- Single-shot 3D photoacoustic tomography using a single-element detector for ultrafast imaging of hemodynamics 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.