Skip to main content
. 2026 Sep 11;14:1927235. doi: 10.3389/fcell.2026.1927235

FIGURE 6.

Two side-by-side bar charts labeled A and B display feature importance for a support vector machine (SVM) model. Chart A, titled "Global Feature Importance (SVM)," ranks features by mean SHAP value, with g__Thomasclavelia and g__Ruminococcaceae_unclassified as most important. Chart B, titled "Directional Feature Importance (SVM)," shows the same features with blue bars indicating negative (protective) impact and red bars indicating positive (risk) impact, highlighting g__Megamonas as most protective and g__Eggerthellaceae_unclassified as highest risk.

Interpretability analysis of the multi-omics diagnostic model using SHAP values (A) Global feature importance of the 18 core predictors, ranked by mean absolute SHAP value (mean (|SHAP|)) across all samples. (B) Directional SHAP summary plot showing the magnitude and direction of each feature’s contribution to model-predicted infertility risk. Red color indicates positive risk predictor and blue color indicated protective predictor.