Evaluating Spatial Encoding Strategies for Cell Type Annotation with Spatial Omics Data
Abstract
Recent spatial omics research leverages the assumption that spatial information enhances model performance on the cell type annotation task. This study investigates and challenges that assumption by conducting benchmark experiments comparing the performance of spatial and non-spatial models. We show that graph-based spatial models do not consistently outperform non-spatial models, provide theories to explain our findings, and make recommendations for future work on spatial encoding strategies.
Cite
Text
Kuijs et al. "Evaluating Spatial Encoding Strategies for Cell Type Annotation with Spatial Omics Data." ICLR 2024 Workshops: MLGenX, 2024.Markdown
[Kuijs et al. "Evaluating Spatial Encoding Strategies for Cell Type Annotation with Spatial Omics Data." ICLR 2024 Workshops: MLGenX, 2024.](https://mlanthology.org/iclrw/2024/kuijs2024iclrw-evaluating/)BibTeX
@inproceedings{kuijs2024iclrw-evaluating,
title = {{Evaluating Spatial Encoding Strategies for Cell Type Annotation with Spatial Omics Data}},
author = {Kuijs, Merel and Andersson, Alma and Hajiramezanali, Ehsan and Biancalani, Tommaso and BenTaieb, Aicha},
booktitle = {ICLR 2024 Workshops: MLGenX},
year = {2024},
url = {https://mlanthology.org/iclrw/2024/kuijs2024iclrw-evaluating/}
}