Skip to main content
. Author manuscript; available in PMC: 2020 Apr 29.
Published in final edited form as: J Neuroimmune Pharmacol. 2018 Oct 1;13(4):532–540. doi: 10.1007/s11481-018-9811-8

Figure 3.

Figure 3.

Inclusion of PGSs from comorbid traits improved the performance of prediction models. Shown in the figure were receiver operating characteristics (ROC) curves from the schizophrenia sPGS calculated at P-value of 1e-3 alone and the schizophrenia sPGS plus sPGSs from other comorbid traits. Also shown were ROC curves from independent CATIE dataset. In both the MGS+SSCCS and CATIE datasets, the inclusion of sPGSs from comorbid traits improved the performance of the prediction models.