Generative aptamer discovery using RaptGen

Abstract Nucleic acid aptamers are generated by an in vitro molecular evolution method known as systematic of ligands exponential enrichment (SELEX). Various candidates limited actual sequencing data from experiment. Here we developed RaptGen, which is a variational autoencoder for silico aptamer generation. RaptGen exploits profile hidden Markov model decoder to represent motif sequences effectively. We showed that embedded simulation sequence into low-dimensional latent space on the basis information. also performed embedding using two independent SELEX datasets. successfully even though they were not included high-throughput sequencing. could generate truncated with short learning model. demonstrated be applied activity-guided generation according Bayesian optimization. concluded generative and representation useful discovery.