DNA repair mutation detection using Deep learning strategy – A pharmacogenomic perspective

DNA repair is an aggregation of several mechanisms through which the cell identifies and retrogresses the damage that occurs in the DNA molecules. This function is facilitated by a specific set of genes called 'DNA repair genes'. Variations caused in DNA repair genes affect its ability to revert the mutations in the entire genome, thus making it an essential set of genes that aids the DNA repair mechanism. Hence DNA damage is considered as the key factor in the development and evolution of cancer cells. Prior detection of DNA damage and targeting the condition with a personalized approach is the main objective of the research. As discussed the cornerstone of the work is to design a mutation detection tool for early diagnosis of the `DNA repair specific mutations' from tissue biopsy images using one shot algorithm. The accuracy of the suggested tool is found to be 70% making it a potential model for prediction of DNA repair mutations, thereby helping us to monitor the cancer conditions effectively. To analyze and control the DNA damage a personalized approach has been followed using pharmacogenomic analysis, comprising the factors such as genomics, epigenomics, environmental genomics and metagenomics. The healthcare data acquired from the above sources has been suggested to be secured using Block chain algorithm.