About this item:

64 Views | 47 Downloads

Author Notes:

Corresponding author at: Rollins School of Public Health, Emory University, Office: 1518 Clifton Rd. CNR 4049 (1518-002-AA), Atlanta, GA 30322, USA. E-mail address: bpearce@emory.edu (B.D. Pearce)

This work was made possible due the work of Ann E. Pulver PhD, who is the director of the Epidemiology/Genetics Research Program in Psychiatry at Johns Hopkins University. Dr. Pulver and her team recruited the participants, collected samples, and performed key evaluations through her epidemiology genetics program.

We are also grateful to Robert Yolken, MD (Theodore and Vada Stanley Distinguished Professor of Neurovirology in Pediatrics at Johns Hopkins) whose laboratory performed the serology assays.

See publication for list of contributions and competing interests.


Research Funding:

1R01MH092512 (BDP, DA), NIH R01MH094757, R01MH096764.


  • Science & Technology
  • Life Sciences & Biomedicine
  • Psychiatry

Polygenic risk scores differentiate schizophrenia patients with toxoplasma gondii compared to toxoplasma seronegative patients

Show all authors Show less authors


Journal Title:



Volume 107


, Pages 152236-152236

Type of Work:

Article | Final Publisher PDF


Schizophrenia (SCZ) is an etiologically heterogeneous disease with genetic and environmental risk factors (e.g., Toxoplasma gondii infection) differing among affected individuals. Distinguishing such risk factors may point to differences in pathophysiological pathways and facilitate the discovery of individualized treatments. Toxoplasma gondii (TOXO) has been implicated in increasing the risk of schizophrenia. To determine whether TOXO-positive individuals with SCZ have a different polygenic risk burden than uninfected people, we applied the SCZ polygenic risk score (SCZ-PRS) derived from the Psychiatric GWAS Consortium separately to the TOXO-positive and TOXO-negative subjects with the diagnosis of SCZ as the outcome variable. The SCZ-PRS does not include variants in the major histocompatibility complex. Of 790 subjects assessed for TOXO, the 662 TOXO-negative subjects (50.8% with SCZ) reached a Bonferroni corrected significant association (p = 0.00017, R2 = 0.023). In contrast, the 128 TOXO-positive individuals (53.1% with SCZ) showed no significant association (p = 0.354) for SCZ-PRS and had a much lower R2 (R2 = 0.007). To account for Type-2 error in the TOXO-positive dataset, we performed a random sampling of the TOXO-negative subpopulation (n = 130, repeated 100 times) to simulate equivalent power between groups: the p-value was <0.05 for SCZ-PRS 55% of the time but was rarely (6% of the time) comparable to the high p-value of the seropositive group at p > 0.354. We found intriguing evidence that the SCZ-PRS predicts SCZ in TOXO-negative subjects, as expected, but not in the TOXO-positive individuals. This result highlights the importance of considering environmental risk factors to distinguish a subgroup with independent or different genetic components involved in the development of SCZ.

Copyright information:

© 2021 Published by Elsevier Inc.

This is an Open Access work distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (https://creativecommons.org/licenses/by-nc-nd/4.0/).
Export to EndNote