Genome wide association studies (GWAS) has allowed the discovery of some interesting risk variants for schizophrenia (SCZ). However, this high-throughput approach presents some limitations, being the most important the necessity of highly restrictive statistical corrections as well as the loss of statistical power inherent to the use of a Single Nucleotide Polymorphism (SNP) analysis approach. These problems can be partially solved through the use of a polygenic approach. We performed a genotyping study in SCZ using 86 previously associated SNPs identified by GWAS of SCZ, bipolar disorder (BPD) and autistic spectrum disorder (ASD) patients. The sample consisted of 3063 independent cases with DSM-IV-TR diagnosis of SCZ and 2847 independent controls of European origin from Spain. A polygenic score analysis was also used to test the overall effect on the SCZ status. One SNP, rs12290811, located in the ODZ4 gene reached statistical significance (p = 1.7 × 10− 4, Allelic odds ratio = 1.21), a value very near to those reported in previous GWAS of BPD patients. In addition, 4 SNPs were close to the significant threshold: rs3850333, in the NRXN1 gene; rs6932590, at MHC; rs2314398, located in an intergenic region on chromosome 2; and rs1006737, in the CACNA1C gene. We also found that 74% of the studied SNPs showed the same tendency (risk or protection alleles) previously reported in the original GWAS (p < 0.001). Our data strengthen the polygenic component of susceptibility to SCZ. Our findings show ODZ4 as a risk gene for SCZ, emphasizing the existence of common vulnerability in psychosis.