Genetic disorders, intrinsic to DNA, can manifest unpredictably, often causing severe health implications. A lack of awareness about genetic disorders among people results in low genetic testing during the early stages which leads to an increase in the potential risk of disease severely affecting life of individuals. The severity and distribution of genetic disorders across the globe vary significantly. Current machine-learning diagnostic methods are limited to specific diseases leading to a constricted scope of prediction systems. There is a necessity for an efficient, non-invasive, and economical genetic disorder prediction system. The project aims to develop a predictive model leveraging Machine-Learning algorithms and model development techniques. The core objective is formulating a model proficient in pinpointing genetic disorders and disorder subclass, offering swift, accurate diagnostics leading to timely medical interventions. Uniquely, our approach merges a Genetic Algorithm to extract important features, which are used to perform model training incorporating the Random Forest algorithm. The development of a machine-learning predictive model enables predict susceptibilities to genetic disorders and the disorder subclass distinguishing it from traditional genetic sequencing, and gene testing diagnosis procedures.