Graph neural networks learn emergent tissue properties from spatial molecular profiles
Fischer, D. S.; Ali, M.; Richter, S.; Ertürk, A.; Theis, F. J.
Show abstract
Tissue phenotypes such as metabolic states, inflammation, and tumor properties are functions of molecular states of cells that constitute the tissue. Recent spatial molecular profiling assays measure tissue architecture motifs in a molecular and often unbiased way and thus can explain some aspects of emergence of these phenotypes. Here, we characterize the ability of graph neural networks to model tissue-level emergent phenotypes based on spatial data by evaluating phenotype prediction across model complexities. First, we show that immune cell dispersion in colorectal tumors, which is known to be predictive of disease outcome, can be captured by graph neural networks. Second, we show that breast cancer tumor classes can be predicted from gene expression alone without spatial information and are thus too simplistic a phenotype to require a complex model of emergence. Third, we show that representation learning approaches for spatial graphs of molecular profiles are limited by overfitting in the prevalent regime of up to 100s of images per study. We address overfitting with within-graph self-supervision and illustrate its promise for tissue representation learning as a constraint for node representations.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Learning tissue representation by identification of persistent local patterns in spatial omics data 97%
- STAIG: Spatial Transcriptomics Analysis via Image-Aided Graph Contrastive Learning for Domain Exploration and Alignment-Free Integration 97%
- stLearn: integrating spatial location, tissue morphology and gene expression to find cell types, cell-cell interactions and spatial trajectories within undissociated tissues 97%
Similar papers in this journal
- Inferring spatial single-cell-level interactions through interpreting cell state and niche correlations learned by self-supervised graph transformer 96%
- Delineating the Effective Use of Self-Supervised Learning in Single-Cell Genomics 95%
- INSCT: Integrating millions of single cells using batch-aware triplet neural networks 95%
Similar papers in this journal
- S3-CIMA: Supervised spatial single-cell image analysis for the identification of disease-associated cell type compositions in tissue 95%
- Bi-level Graph Learning Unveils Prognosis-Relevant Tumor Microenvironment Patterns in Breast Multiplexed Digital Pathology 95%
- CellContrast: Reconstructing Spatial Relationships in Single-Cell RNA Sequencing Data via Deep Contrastive Learning 95%
"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.