Spatial transcriptome

Cell type identification

Graph-based deep learning

Graph neural networks

Survey

The SpaGCN, CCST, STAGATE, DeepST, conST, GraphST, SpaceFlow, and Spatial-MGCN benchmarking methods were discussed in this review. The codes for these frameworks of spatial clustering are available as follows: 1) SpaGCN: https://github.com/jianhuupenn/SpaGCN; 2) CCST: https://github.com/xiaoyeye/CCST; 3) STAGATE: https://github.com/zhanglabtools/STAGATE; 4) DeepST: https://github.com/JiangBioLab/DeepST; 5) conST: https://github.com/ys-zong/conST; 6) GraphST: https://github.com/JinmiaoChenLab/GraphST; 7) SpaceFlow: https://github.com/hongleir/SpaceFlow; 8) Spatial-MGCN: https://github.com/cs-wangbo/Spatial-MGCN.Eight datasets of spatial transcriptomics were used in this survey to evaluate the above baselines. We have provided the links to the original and the pre-processed resources for each. 1) Human DLPFC: The primary source: https://www.nature.com/articles/s41593-020-00787-0; the pre-processed source: https://github.com/LieberInstitute/spatialLIBD. 2) Human breast cancer: The primary source: https://www.10xgenomics.com/resources/datasets/human-breast-cancer-block-a-section-1-1-standard-1-1-0; the pre-processed source: https://github.com/JinmiaoChenLab/SEDR_analyses/. 3) Mouse brain: The primary source: https://support.10xgenomics.com/spatial-gene-expression/datasets; the pre-processed source: https://github.com/JinmiaoChenLab/SEDR_analyses/. 4) Data on the olfactory bulbs of mice for Slide-seqV2: The primary source: https://singlecell.broadinstitute.org/single_cell/study/SCP815/highly-sensitive-spatial-transcriptomics-at-near-cellular-resolution-with-slide-seqv2#study-summary; the processed version: https://stagate.readthedocs.io/en/latest/T3_Slide-seqV2.html. 5) Data on the olfactory bulbs of mice for Stereo-seq: The primary source: https://github.com/JinmiaoChenLab/SEDR_analyses/tree/master/data; the processed version: https://stagate.readthedocs.io/en/latest/T4_Stereo.html. 6) Mouse liver from MERFISH: the primary source: https://info.vizgen.com/mouse-liver-access; the processed version: https://squidpy.readthedocs.io/en/stable/notebooks/tutorials/tutorial_vizgen_mouse_liver.html#single-cell-clustering-of-vizgen-merfish-mouse-liver-data; 7) Mouse embryo from seqFISH: the primary source: https://crukci.shinyapps.io/SpatialMouseAtlas/; the processed version: https://squidpy.readthedocs.io/en/stable/notebooks/tutorials/tutorial_seqfish.html; 8) Human breasr cancer from 10x Xenium: the primary source: https://www.10xgenomics.com/products/xenium-in-situ/preview-dataset-human-breast; https://squidpy.readthedocs.io/en/stable/notebooks/tutorials/tutorial_xenium.html.The codes for this review have been summarized at https://github.com/narutoten520/Benchmark_SRT.

© 2023 The Author(s). Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology.