Generating and designing DNA with deep generative models
This work addresses the challenge of DNA sequence design for genomics research, representing an incremental application of existing deep learning methods to a new domain.
The paper tackled the problem of generating and designing DNA sequences with desired properties using deep generative models, resulting in new sequences estimated to be superior to training data for protein binding microarrays.
We propose generative neural network methods to generate DNA sequences and tune them to have desired properties. We present three approaches: creating synthetic DNA sequences using a generative adversarial network; a DNA-based variant of the activation maximization ("deep dream") design method; and a joint procedure which combines these two approaches together. We show that these tools capture important structures of the data and, when applied to designing probes for protein binding microarrays, allow us to generate new sequences whose properties are estimated to be superior to those found in the training data. We believe that these results open the door for applying deep generative models to advance genomics research.