Fig. 4: Comparison of demographic, clinical and cognitive characteristics of biotypes (I).

Discovery set: biotype 1, N = 426; biotype 2, N = 497. Replication set: biotype 1, N = 110, N = 146. First-degree relatives: biotype 1, N = 153; biotype 2, N = 178, control biotype, N = 95. WMS Weschler Memory Scale, BACS Brief Assessment of Cognition in Schizophrenia, YMRS Young Mania Rating Scale, BSFS Birchwood Social Functioning Scale, SES Socioeconomic Status (Hollingshead index), SBS Schizo-bipolar Scale, PANSS Positive and Negative Syndrome Scale for Schizophrenia, Chlorpromazine equivalentes: Average daily chlorpromazine dose; MADRS Montgomery-Asberg Depression Rating Scale, BACS Brief Assessment of Cognition in Schizophrenia, BSFS Birchwood Social Functioning Scale, SES Socioeconomic status, SBS Schizo-bipolar Scale, Chlorpromazine equivalents: Average daily chlorpromazine dose; MADRS Montgomery-Asberg Depression Rating Scale, GAF Global Assessment of Functioning Scale; Z-scores are shown. BACS z-scores were obtained from normative data stratified by age and sex. * Non-adjusted p-value < 0.05 from a two-tailed t-test; ** False discovery rate correction for multiple testing p-value < 0.05. Comparisons with three biotypes in first-degree relatives were adjusted with Tukey method for comparing a family of three estimates. WMS: Statistics were obtained from linear models that also accounted for the influence of age, sex, race, ethnicity, site, and socioeconomic status. BACS: Statistics were obtained from linear models that also accounted for the influence of race, ethnicity, site, and socioeconomic status (age and sex accounted for when computing z-scores). Statistical details are shown in Tables S2, S3, and S4. First-degree relatives with a diagnosis of a psychotic disorder were excluded from these analyses.