Figure 1 | Scientific Reports

Figure 1

From: Machine learning to reveal hidden risk combinations for the trajectory of posttraumatic stress disorder symptoms

Figure 1The alternative text for this image may have been generated using AI.

Adjusted P values of combinational/single risk factors significantly associated with a poor prognosis for PTSD symptoms. The Y-axis indicates the negative logarithms of the P values of combinational/single risk factor(s) significantly associated with IES-R trajectory scores and the P value of each component of the significant combinations. As a measure to control the familywise error rate by correcting for multiple comparisons, the P values for combinations were adjusted by MP-LAMP, and the P values for single factors were adjusted by the Bonferroni correction. Combinations are represented by points connected by lines, and single factors are represented by points without lines. Among 56 significantly associated combinations, the representative combinations (including the combinations whose P values were the smallest for each component) are shown. MP-LAMP substantially increased the power to detect significant predictors by testing combinations, in which the components of significant combinations were not necessarily associated with the target variable as individual risk factors.

Back to article page