Learning and Predicting DNA Sequences in DNA-nanotube Conjugates with High Response to Serotonin

DNA-wrapped single walled carbon nanotube (SWNT) conjugates have emerged as promising optical tools for imaging and sensing important small biological molecules, such as neurotransmitters. In this presentation, I will describe how we use experimental near-infrared (nIR) fluorescence response datasets for ~100 DNA-SWNT conjugates, to train machine learning (ML) models to predict new unique DNA sequences with strong optical responses to the neurotransmitter serotonin. First, classifier models based on convolutional neural networks (CNN) are trained on sequence features to classify DNA ligands as either high response or low response to serotonin. Second, support vector machine (SVM) regression models are trained to predict relative optical response values for DNA sequences. Finally, we demonstrate with validation experiments that integrating the predictions of multiple highest quality CNN classifiers and SVM regression models leads to the best predictions of both high and low response sequences. With our ML approaches, we discovered five new DNA-SWNT sensors with higher fluorescence intensity response to serotonin than obtained previously. These new approaches for discovering functional DNA molecules could be applied to developing tools to detect other biological molecules of interest.