The invention relates to a spatial
omics tumor evolution prediction method and
system based on a graph neural network. The method comprises the following steps: acquiring different types of spatial
transcriptome data, and preprocessing and normalizing the spatial
transcriptome data; annotating and sequencing the
gene expression counting data according to the
genome position, and further grouping the genes; constructing a spatial neighbor graph, and calculating spatial statistics; constructing a spatial neighbor
graph based on the adjacent matrix, and training a graph neural network; identifying a credible diploid basic unit, and estimating a copy number; adding the estimated copy number into the spatial neighbor graph, and retraining the graph neural
network model to obtain potential embedding; according to the estimated copy number, determining a
breakpoint by using a segmentation
algorithm, and obtaining copy number contours of each basic unit and the space clone; and performing downstream
task analysis by utilizing potential embedding, and expanding to different spatial
omics samples for comparative analysis. Compared with the prior art, the method has the advantages of accurate tumor evolution mode prediction, high analysis efficiency and the like.