Single-cell feature analysis methods, apparatuses, devices, and storage media

Abstract

The application discloses a single-cell feature analysis method, device, equipment and storage medium. The method comprises the following steps: obtaining cell feature data set of a target sample, wherein the cell feature data set comprises feature data of at least two cells, the feature data of various cells comprises at least one of cell proportion data corresponding to a cell type, gene expression data, gene interaction data, gene set score data and transcription factor expression data; wherein the cell proportion data is the percentage of the number of each type of cell in the target sample in the total number of cells; determining the first contribution score of each cell feature data in the cell feature data set to the classification prediction result of a target classification model; and determining a single-cell feature importance analysis result according to the first contribution score of each cell feature data to the classification prediction result of the target classification model. The application solves the problem of low accuracy of the existing single-cell feature analysis method and improves the accuracy of the single-cell feature analysis method.

Images