Utveckling av ett kandidatprediktionssystem för DNA-strukturer vid läkemedelsleverans : Ett programmatiskt sätt att förutsäga celluptagningsförmågan för DNA-sekvenser

DNA nanostructures offer a promising platform for precision medicine, with the potential to target specific cells, reduce off-target effects, and improve therapeutic efficacy. In this paper, I present the development of a high-throughput candidate prediction system designed to rank DNA sequences based on their predicted levels of cell uptake—addressing a significant challenge in modern DNA research. The system extracts valuable features from DNA sequence reads and their simulated structures to efficiently reduce and prioritize the most promising candidate sequences within a given library for drug delivery applications. This is achieved by developing a predictive machine learning model trained to rank DNA sequences according to their predicted cellular uptake. Additionally, the system integrates advanced computational and machine learning tools into the pipeline, aligning with broader trends in computational healthcare innovation.