De Novo Protein Design for Novel Folds Using Guided Conditional Wasserstein Generative Adversarial Networks

Abstract

Although massive data is quickly accumulating on protein sequence and structure, there is a small and limited number of protein architectural types (or structural folds). This study is addressing the following question: how well could one reveal underlying sequence–structure relationships and design protein sequences for an arbitrary, potentially novel, structural fold? In response to the question, we have developed novel deep generative models, namely, semisupervised gcWGAN (guided, conditional, Wasserstein Generative Adversarial Networks). To overcome training difficulties and improve design qualities, we build our models on conditional Wasserstein GAN (WGAN) that uses Wasserstein distance in the loss function. Our major contributions include (1) constructing a low-dimensional and generalizable representation of the fold space for the conditional input, (2) developing an ultrafast sequence-to-fold predictor (or oracle) and incorporating its feedback into WGAN as a loss to guide model training, and (3) exploiting sequence data with and without paired structures to enable a semisupervised training strategy. Assessed by the oracle over 100 novel folds not in the training set, gcWGAN generates more successful designs and covers 3.5 times more target folds compared to a competing data-driven method (cVAE). Assessed by sequence- and structure-based predictors, gcWGAN designs are physically and biologically sound. Assessed by a structure predictor over representative novel folds, including one not even part of basis folds, gcWGAN designs have comparable or better fold accuracy yet much more sequence diversity and novelty than cVAE. The ultrafast data-driven model is further shown to boost the success of a principle-driven de novo method (RosettaDesign), through generating design seeds and tailoring design space. In conclusion, gcWGAN explores uncharted sequence space to design proteins by learning generalizable principles from current sequence–structure data. Data, source codes, and trained models are available at

Quantitative Estimate Index for Early-Stage Screening of Compounds Targeting Protein-Protein Interactions

“…Additionally, in further studies, QEPPI can be used as a reward in sequence-based molecular generation models using reinforcement learning such as REINVENT [ 18 , 19 ], and as a condition for sequence-based molecular generation models using conditional Wasserstein generative adversarial networks (WGANs) and Variational Autoencoders (VAEs), such as gcWGAN [ 26 ] and CVAE [ 27 ], which will enable molecular design with high PPI-targeting compound properties.…”

Evolving scenario of big data and Artificial Intelligence (AI) in drug discovery

“…Hybrid approaches like LSTM-GAN (long short-term memory–generative adversarial network), DCGAN (deep convolutional generative adversarial network), gcWGAN (guided conditional Wasserstein GAN), which are constructed using different deep learning paradigms, have been successfully used in de novo protein design [ 112 , 113 ].…”

Recent advances in de novo protein design: Principles, methods, and applications

“…E , neural networks can predict the probabilities of sequences given a backbone structure ( 102 , 103 ) ( red ). Generative machine learning models design sequences by latent space sampling ( 104 , 105 , 106 , 107 , 108 ) ( green ). The TR-Rosetta neural network predicts the probability of the structure of a given sequence.…”

“…Generative models learn distributions of protein sequences and can generate new native-like protein sequences with or without input backbone structures. A number of generative models were developed for sequence design, including generative adversarial networks ( 104 ), variational autoencoders ( 105 , 106 ), and graph-based ( 107 , 108 ) models. Notably, the structure prediction neural network from TR-Rosetta ( 42 ) can be repurposed for sequence optimization ( 109 ).…”