UTRGAN: Learning to Generate 5′ UTR Sequences for Optimized Translation Efficiency and Gene Expression

The 5' untranslated region (5' UTR) of the messenger RNA plays a crucial role in the translatability and stability of the molecule. Thus, it is an important component in the design of synthetic biological circuits for high and stable expression of intermediate proteins. Several UTR sequences are patented and used frequently in laboratories. We present a novel model UTRGAN, a Generative Adversarial Network (GAN)-based model designed to generate 5' UTR sequences coupled with an optimization procedure to ensure a target feature such as high expression for a target gene sequence or high ribosome load and translation efficiency. We rigorously analyze and show that the model can generate sequences that mimic various properties of natural UTR sequences. Then, we show that the optimization procedure yields sequences that are expected to yield (i) 61% higher average expression (up to 5-fold) on a set of target genes, (ii) 53% higher mean ribosome load on average (up to 2-fold for the best 5' UTR), and (iii) a 34-fold increase on average translation efficiency, compared to the initially generated UTR sequences. We also demonstrate that when there is a single target gene of interest, the expected expression increases by at least 37% on average and up to 8-fold for certain genes (up to 32-fold for the best 5' UTR).