Spage2vec: Unsupervised representation of localized spatial gene expression signatures

Spage2vec is an unsupervised segmentation free approach for decrypting the spatial transcriptomic heterogeneity of complex tissues at subcellular resolution. Spage2vec represents the spatial transcriptomic landscape of tissue samples as a graph and leverage powerful machine learning graph representation technique to create a lower dimensional representation of local spatial gene expression. Here we visualize spage2vec localized gene expression signatures of different spatial transcriptomic datasets [1-4].

We thank Mats Nilsson, Sten Linnarsson and Xiaowei Zhuang for making their datasets publicly available.

[1] Qian, X., Harris, K. D., Hauling, T., Nicoloutsopoulos, D., Muñoz-Manchado, A. B., Skene, N., ... & Nilsson, M. (2020). Probabilistic cell typing enables fine mapping of closely related cell types in situ. Nature methods, 17(1), 101-106.
[2] Codeluppi, S., Borm, L. E., Zeisel, A., La Manno, G., van Lunteren, J. A., Svensson, C. I., & Linnarsson, S. (2018). Spatial organization of the somatosensory cortex revealed by osmFISH. Nature methods, 15(11), 932-935.
[3] Moffitt, J. R., Bambah-Mukku, D., Eichhorn, S. W., Vaughn, E., Shekhar, K., Perez, J. D., ... & Zhuang, X. (2018). Molecular, spatial, and functional single-cell profiling of the hypothalamic preoptic region. Science, 362(6416), eaau5324.
[4] Chen, K. H., Boettiger, A. N., Moffitt, J. R., Wang, S., & Zhuang, X. (2015). Spatially resolved, highly multiplexed RNA profiling in single cells. Science, 348(6233), aaa6090.