Synthetic Promoter Design in Escherichia coli based on Generative Adversarial Network

Abstract

Synthetic promoters are commonly applied elements in circuit design for fine-tuning the protein expression levels. Promoter engineering was mostly focused on the random mutation or combination of regulation elements such as transcription factor binding sites. However, the size of promoter sequence space is still overwhelming and better navigation method is required. On the other hand, the generative adversarial network (GAN) is known for its great ability to reduce the searching space by learning to generate new data on the similar manifold of original data. Here, we applied WGAN-GP model into de novo promoter sequence design to generate entirely new promoter sequences. In total, 83 of model-generated promoter sequences were tested in promoter activity screening by regulating the expression of sfGFP gene in Escherichia coli. As a result, 26 out of 83 newly designed promoters were found functional and successfully expressed with varying activities, with similarity score to natural promoters all less than 0.7. Moreover, 3 of them showed higher promoter strength than the wild type promoters and their highly expression mutants. The much higher successful rate and promoter activity with much lower similarity score in our model-designed novel promoters confirmed the effectiveness of promoter sequence learning. Our work provides insights into an area of navigation of novel functional promoter sequence space automatically, as well as speeding up evolution process of naturally existing promoters, indicating the potential ability for deep generative models to be applied into genetic element designing in the future.

BP-GAN: Interpretable Human Branchpoint Prediction Using Attentive Generative Adversarial Networks

“…A GAN is a deep generative model with an exceptional capability to learn the structure of parameters and generate samples similar to true data [13]. It has initiated several recent studies in genomics [15]- [19].…”

“…In [15], [19], the approaches generated DNA sequences using a GAN. A GAN model in [17] aimed to generate RNA sequences for protein function analysis whereas other works focused on generating protein structures [16] and sequences [18].…”

“…2) Our method uses a GAN to learn the structure of latent variables for RNA sequences relevant to BP prediction. To the best of our knowledge, a GAN is suitable for generative tasks [15]- [19] in gene expression, however, this is the first GAN-based attempt to address a discriminative task. We believe that our approach is also applicable to other related downstream tasks.…”

Improving protein function prediction with synthetic feature samples created by generative adversarial networks

“…Summary of hyper-parameters of different classification algorithms for conducting the 5-fold cross validation-based grid-search by using the Scikit-learn. k-Nearest Neighbours n_neighbors: [1,3,5,7,9,11,13,15,17,19,21,23,25,27,29] weights: ['uniform', 'distance']…”

PFP-WGAN: Protein function prediction by discovering Gene Ontology term correlations with generative adversarial networks

“…The considerable performance of GANs in the field of image processing [35,36] motivates researchers to exploit them for other data types including biological ones. Works in [37,38] use GANs to analyze gene expression profiles and works in [39,40] attempt to synthesize genes and promoters by GANs. Recently authors in [41] have been proposed to perform data augmentation to generate synthetic training samples by a GAN to improve a classifier accuracy for annotating proteins.…”