Abstract : The work aims at designing and implementing a deep pharmacogenomic approach and proposing a secure sharing system of patient data. It has been found that there is a correlation between the cancer types affecting breast, ovary and cervical regions through gene expression analysis. Common mechanisms, phenotypes, chemicals (influencing these genes) have been identified for the major genes. Their corresponding genetic signatures were identified to study the proneness of breast cancer patients towards ovarian cancer and cervical cancer. A deep learning tool has been developed to predict the mutation profile from pathological images. Based on the expression levels of the corresponding genes, the type of cancer could be identified. The designed prediction model gave an initial accuracy of 70.3 %. © 2019, Editorial office of Journal of International Pharmaceutical Research. All rights reserved.