Background: Autism is a common psychiatric disorder in children. Since autism is a multifactorial disease, the genetic predisposition plays a significant role in the pathogenesis. However, numerous studies focused on genomic abnormalities in autism are unable to provide reproducible information about pathogenic processes causing this devastating disorder. The aim of the study: The identification of candidate genes by bioinformatic analysis of recurrent copy number variations (CNV) (5p15.33 duplications) revealed by molecular karyotyping in a clinical cohort. Materials and methods: Molecular karyotyping of 296 children with idiopathic autism, intellectual disability was performed by SNP-array. Bioinformatic analysis was made using an original algorithm. Results: Molecular karyotyping genome-wide analysis revealed 3 cases of 5p15.33 duplications. Bioinformatic analysis identified a candidate gene TPPP for brain dysfunction. TPPP is highly expressed in the brain; the gene encodes a protein catalyzing tubulin polymerization, which is important for oligodendrocytes myelination. Interactome analysis was performed to identify pathogenic processes associated with CNV involving TPPP. Expanded TPPP interactome network encompasses 37 proteins, 19 of which are associated with the synaptic plasticity and axonal guidance involved in the normal development and functioning of the brain. Changes in these processes may lead to autism and intellectual disability. Interestingly, clinical genetic databases have not previously associated this gene with a disease condition. Conclusion: Bioinformatic analysis of 5p15.33 CNV allowed us to show that TPPP is a candidate gene for alterations to the development and functioning of the brain. Accordingly, possible disease mechanisms leading to the development of autism with intellectual disability have been proposed. Since data on candidate processes is useful for personalized treatment, we conclude that molecular karyotyping complemented by our original in silico analysis of epigenome, proteome and metabolome is to become an important component for basic and applied research in psychiatric genetics.