A Deep Learning Model for Predicting NGS Sequencing Depth from DNA Sequence

Abstract Targeted high-throughput DNA sequencing is a primary approach for genomics and molecular diagnostics, and more recently as a readout for DNA information storage. Oligonucleotide probes used to enriching gene loci of interest have different hybridization kinetics, resulting in non-uniform coverage that increases sequencing costs and decreases sequencing sensitivities. Here, we present a deep learning model (DLM) for predicting NGS sequencing depth from DNA probe sequence. Our DLM includes a bidirectional recurrent neural network that takes as input both DNA nucleotide identities as well as the calculated probability of the nucleotide being unpaired. We applied our DLM to two different NGS panels: a designed 7,373-plex panel for DNA information storage, and a 39,145-plex panel for human single nucleotide polymorphisms. In cross-validation, our DLM predicts sequencing depth to within a factor of 3 with 99% accuracy for the designed panel, and 93% accuracy for the human panel. The same model is also effective at predicting the measured single-plex kinetic rate constants of DNA hybridization and strand displacement.