Nano scale instance-based learning using non-specific hybridization of DNA sequences

DNA, or deoxyribonucleic acid, is a powerful molecule that plays a fundamental role in storing and processing genetic information of all living organisms. In recent years, scientists have harnessed hybridization powers between DNA molecules to perform various computing tasks in DNA computing and DNA storage. Unlike specific hybridization, non-specific hybridization provides a natural way to measure similarity between the objects represented by different DNA sequences. We utilize such property to build an instance-based learning model which recognizes an object by its similarity with other samples. The handwriting digit images in MNIST dataset are encoded by DNA sequences using a deep learning encoder. And the reverse complement sequence of a query image is used to hybridize with the training instance sequences. Simulation results by NUPACK show that this classification model by DNA could achieve 95% accuracy on average. Wet-lab experiments also validate the predicted yield is consistent with the hybridization strength. Our work proves that it is feasible to build an effective instance-based classification model for practical application.