Fig. 2

Evaluation and analysis of machine learning classifiers based on different feature selection strategies. A IFS results of five feature selection strategies in four machine learning algorithms. B UMAP shows the clustering of nine placental cell subpopulations in all gene sets (right) and TURF optimal gene sets (left). C The heatmap shows the correlation of subpopulations of placental cells. D Based on the TURF optimal gene set, the Confusion matrix of XGBoost on the independent dataset. E The bar graph shows the mean absolute value of the SHAP values of the first 20 genes for the TURF_XGB