Retinoblastoma is a pediatric retinal tumor initiated by biallelic inactivation of the retinoblastoma gene (RB1). Most of the alteration being unique and randomly distributed throughout the entire coding region. This study aimed firstly to identify mutations that may affect the RB1 gene in constitutional level witch contribute to the understanding of the molecular pathogenesis of this disease and for early detection of subject at risk and asymptomatic carriers. The detection of variations in 30 patient's DNAs was performed by high performance liquid chromatography (HPLC) followed by sequencing. Secondly, we developed a protocol in silico analysis using different software based on analytic methods (Align GVGD, Mutation tasting, SIFT, PolyPhen, I-Mutant and KD4V) and structural (Swiss-Pdb Viewer). This protocol was used to predict the deleterious effects of mutations on protein pRb. The spectrum of mutation included 2 missense mutations, 1 nonsense mutation, 1 deletion, 1 mutation affecting splice site and 2 polymorphisms. Among these mutations, some were identified in germinal level for children with no family history of the disease. Therefore, in silico analysis identified only three causal mutations, the first intronic mutation in the splice donor site caused probably a frame shift, giving rise to a truncated protein. The second missense mutation c.1903 G˃C was predicted to affect splicing processes. The third mutation c.1961T>A located in exon 20 may disrupt protein function and structure.In this study, in silico analysis has shown the effect of mutations on the structure and function of the protein. It is interesting to compare these results with others functional studies in order to evaluate the functionality of the mutations.