Schizophrenia is a disabling disorder affecting approximately 1% of the population worldwide. Twins studies and family studies estimated that genetic factors contributes to 64%-80% of the liability of schizophrenia. With the development of Genome-Wide Association Study (GWAS), a total of 108 common genetic loci associated with schizophrenia has been identified through the meta-analysis conducted by the Schizophrenia Working Group of Psychiatric Genomics Consortium (PGC). Using the summary statistic from the PGC schizophrenia GWAS, B. K. Bulik- Sullivan et al., (2015) estimated that the common variant contributes to around 55.5% of the liability of schizophrenia using LD SCore regression (LDSC). However, studies suggest that rare mutations, structural variance, copy number variation (CNV) and also genetic-environment interaction are all contributing to the risk of schizophrenia. This suggest that the estimate from B. K. Bulik-Sullivan et al., (2015) might be too high. An independent estimation of the Single Nucleotide Polymorphism (SNP)-heritability of schizophrenia might provide insight as to whether if estimates from B. K. Bulik-Sullivan et al., (2015) are inflated. In this thesis, we developed SNP HeRitability Estimation Kit (SHREK), an alternative algorithm for the estimation of SNP-heritability. Our simulation results suggest that, SHREK provided a more robust estimate for oligogenic traits and for binary traits when no confounding variables was present when compared to LDSC. Most importantly, using the summary statistics from the schizophrenia GWAS, the SNP-heritability of schizophrenia is estimated to be 0.185 (SD=0.00450) by SHREK and 0.198 (SD=0.0057) by LDSC, suggesting that the estimates from B. K. Bulik- Sullivan et al., (2015) are inflated. Thus, the common variants contributes to less than 20% of the liability of schizophrenia and it is likely for other genetic variants such as rare mutations and epigenetic factors to contribute to the heritability of schizophrenia. In addition, previous studies have reported the interaction between genetic variation and prenatal infection in the etiology of schizophrenia. There are evidences that the effect of prenatal infection is mediated by maternal immune response, thus it is likely for the perturbation induced by maternal immune activation (MIA) to interact with genetic variations in the development of schizophrenia. We therefore performed a RNA-sequencing study to investigate whether there are any genetic overlaps between differential genes induced by MIA and genetic variations detected by schizophrenia GWAS using the polyriboinosinic-polyribocytidilic acid (PolyI:C) mouse model. We found that the functional gene sets associated with schizophrenia are also enriched in MIA. In addition, when investigating the treatment effect of n-3 polyunsaturated fatty acid (PUFA) rich diet in MIA, we found that the gene expression of Sgk1, a gene that regulates the glutamatergic system, is affected by the n-3 PUFA rich diet in the PolyI:C exposed mice. Sgk1 is therefore a potential mediator of treatment effect of n-3 PUFA rich diet in the MIA model. In conclusion, our results suggested that genes related to neural function and calcium ion signaling, as well as glutamate-related genes such as Sgk1, are the potential targets for future schizophrenia research.