Cancer syndromes are occur due to DNA repair gene mutations of the mismatch repair (MMR) genes, nucleotide excision repair (NER) group, DNA cross linking repair, and plenty of others. Epigenetic gene inactivation processes influencing DNA repair genes can inactivate or reduce the effectiveness of a DNA repair system as an alternate to genetic mutation. Machine learning algorithms, being one of the computational techniques can be trained using data from a large number of patients, but individual physicians and researchers would find it impossible to obtain such experience throughout the course of their careers or study. The digital libraries including IEEE Xplore Digital Library, Google Scholar, Science Direct, PubMed, Elsevier, ACM Digital library, and Springer were explored and literature reviewed. Overall, the machine learning framework showed to be a successful technique of identifying and generating therapy strategies based on RNA repair cancer genes as the complexity and biological granularity (cancer subtypes) of the machine learning increased. This implies that there is some sort of batch effect at work. The findings should be investigated further to see if the machine learning is detecting biologically significant trends in RNA repair gene participation or only learning to distinguish anything in the sequencing process that is characteristic of the data source.