SEPAL: Spatial Gene Expression Prediction from Local Graphs

TL;DR

SEPAL is a novel graph neural network model that predicts gene expression from tissue images by leveraging local visual context, outperforming existing methods and establishing a new benchmark for spatial transcriptomics analysis.

Contribution

It introduces SEPAL, a graph neural network-based approach that directly supervises relative gene expression differences and utilizes local visual context for improved predictions.

Findings

SEPAL outperforms previous state-of-the-art methods.

The model effectively leverages local visual context.

A new benchmark for spatial gene expression prediction is proposed.

Abstract

Spatial transcriptomics is an emerging technology that aligns histopathology images with spatially resolved gene expression profiling. It holds the potential for understanding many diseases but faces significant bottlenecks such as specialized equipment and domain expertise. In this work, we present SEPAL, a new model for predicting genetic profiles from visual tissue appearance. Our method exploits the biological biases of the problem by directly supervising relative differences with respect to mean expression, and leverages local visual context at every coordinate to make predictions using a graph neural network. This approach closes the gap between complete locality and complete globality in current methods. In addition, we propose a novel benchmark that aims to better define the task by following current best practices in transcriptomics and restricting the prediction variables to only those with clear spatial patterns. Our extensive evaluation in two different human breast cancer datasets indicates that SEPAL outperforms previous state-of-the-art methods and other mechanisms of including spatial context.