Schizophrenia is a severe common mental disease. Both genetic and environmental factors contribute to the schizophrenia development. DNA methylations, as the dynamic consequence of the interplay between genome and environment and further link with phenotypes, are investigated in schizophrenia patients using traditional methylation profiling methods. Many methylation patterns are discovered to deepen our understanding of the molecular etiology of schizophrenia. However, traditional whole-genome methylation profiling methods cannot reliably measure all-site methylation level, and leave a lot of methylation patterns uncovered, while novel methylation and genetic variants are needed urgently for schizophrenia. To overcome these limitations, this project focuses on novel biomarker discovery from genome-wide methylation/variant dissection in schizophrenia samples using Nanopore sequencing. To do this, (1) methylation detection on Nanopore sequencing will be substantially improved using novel self-supervised learning where unlabeled Nanopore data will substantially improve the learning process. The methylation detection error in the downstream analysis of schizophrenia is thus minimized; (2) 10 pairs of schizophrenia samples and control will be sequenced via Nanopore sequencing, and whole-genome millions of methylations will be detected using improved methylation detection. Since millions of methylation sites are reliably measured, novel methylation differentiation in schizophrenia will be explored; (3) Both wholegenome genetic variants and methylation are detected and investigated simultaneously to discover more methylation-variant patterns for schizophrenia. The completion of these aims will provide novel methylation and methylation-variant biomarkers for deepening our understanding of schizophrenia.