Predictive Models for Suicide Attempts in Major Depressive Disorder and the Contribution of <i>EPHX2</i>: A Pilot Integrative Machine Learning Study

<div>The results from the three ML models and contributions of various features to identify MDD patients. (a) The ROC of three ML models in the validation set. (b) The DCA of three ML models in the validation set. (c) The global bar plot of SHAP values from D-Model B. The features are arranged in descending order based on their contributions from the D-Model B. (d) The beeswarm summary plot of SHAP values from D-Model B. The visualisation indicates the impact of these features on predictions, with colors representing the feature values from high (red) to low (blue). The horizontal position shows whether the feature value leads to a positive or negative prediction. Each point represents a SHAP value of the feature for a specific case.</div>

Depression and Anxiety

fig3

Figure 3

Figure 3: Predictive Models for Suicide Attempts in Major Depressive Disorder and the Contribution of <i>EPHX2</i>: A Pilot Integrative Machine Learning Study 

Figure 3 | Predictive Models for Suicide Attempts in Major Depressive Disorder and the Contribution of EPHX2: A Pilot Integrative Machine Learning Study 