Abstract : Studies on nsSNPs have demonstrated their implications in several pathologies including hereditary diseases. In cancerous diseases, several genes have been the subject of studies in search of a possible contribution, particularly in hereditary cancers such as prostate cancer. The PTEN gene is one of the candidates frequently studied in this cancer, primarily due to its crucial role in the PIK3/AKT pathway, which is one of the fundamentals signaling pathways involved in carcinogenesis. Throughout this study we employed various computational approaches to identify nsSNPs that maybe deleterious to the structure and/or function of PTEN protein and potentially contributing to diseases such as cancers. The analysis was performed by executing various valid computational tools, including : SIFT, PANTHER, ALIGN GVGD, PolyPhen.2, I.Mutant.2, Fathmm, ConSurf, Interproscan, Raptor X, SWISS-MODEL. Out of 573 variants, 148 variants were found to be deleterious after analysis by the four tools (SIFT, Polypehen2, Panther, Align GVGD). Further evaluation of these variants by the three tools consurf, Fathmm and I.Mutant allowed the identification of 97 SNPs associated with cancer and decrease the stability of the protein. Analysis of the functional domains of the protein (InterProscan tool) revealed that 86% of the variants are located in the PTP domain, of which 14 variants correspond to 7 codons situated in the active site of the protein. Protein structure property prediction by Raptor X server showed significant differences in secondary structure (SS), solvent accessibility (ACC), and disorder regions (DISO). The results of this study support the probability of a pathogenic effect of the 14 variants of the PTEN protein : I122S, H123D, C124S, C124R, C124G, C124Y, G127R , G127V, G129R, G129V, R130G, R130P, G132D, G132V. With the emergence and current prominence of personalized medicine in the care of cancer patients, in vivo studies on the variants identified in this computational study are highly opportune, in particulary in association with the other risk factors and clinicopathologic parameters of the patients.