How to Construct a Bottom-Up Nomothetic Network Model and Disclose Novel Nosological Classes by Integrating Risk Resilience and Adverse Outcome Pathways with the Phenome of Schizophrenia.

Current case definitions of schizophrenia (DSM-5, ICD), made through a consensus among experts, are not cross-validated and lack construct reliability validity. The aim of this paper is to explain how to use bottom-up pattern recognition approaches to construct a reliable and replicable nomothetic network reflecting the direct effects of risk resilience (RR) factors, and direct and mediated effects of both RR and adverse outcome pathways (AOPs) on the schizophrenia phenome. This study was conducted using data from 40 healthy controls and 80 patients with schizophrenia. Using partial least squares (PLS) analysis, we found that 39.7% of the variance in the phenomenome (lowered self-reported quality of life) was explained by the unified effects of AOPs (IgA to tryptophan catabolites, LPS, and the paracellular pathway, cytokines, and oxidative stress biomarkers), the cognitome (memory and executive deficits), and symptomatome (negative symptoms, psychosis, hostility, excitation, mannerism, psychomotor retardation, formal thought disorders); 55.8% of the variance in the symptomatome was explained by a single trait extracted from AOPs and the cognitome; and 22.0% of the variance in the latter was explained by the RR (Q192R polymorphism and CMPAase activity, natural IgM, and IgM levels to zonulin). There were significant total effects (direct + mediated) of RR and AOPs on the symptomatome and the phenomenome. In the current study, we built a reliable nomothetic network that reflects the associations between RR, AOPs, and the phenome of schizophrenia and discovered new diagnostic subclasses of schizophrenia based on unified RR, AOPs, and phenome scores.