A Graph Based Neural Network Approach to Immune Profiling of Multiplexed Tissue Samples
This work addresses the analysis of complex multi-dimensional tissue samples for cancer research, presenting a novel method for immune profiling.
The authors tackled the problem of analyzing multiplexed immunofluorescence data to profile the tumor microenvironment across different tumor stages by using graph neural networks to combine tissue morphology and protein expression features, resulting in a new framework that overcomes key challenges and enables abstraction of biologically meaningful interactions.
Multiplexed immunofluorescence provides an unprecedented opportunity for studying specific cell-to-cell and cell microenvironment interactions. We employ graph neural networks to combine features obtained from tissue morphology with measurements of protein expression to profile the tumour microenvironment associated with different tumour stages. Our framework presents a new approach to analysing and processing these complex multi-dimensional datasets that overcomes some of the key challenges in analysing these data and opens up the opportunity to abstract biologically meaningful interactions.