Functional Protein Structure Annotation Using a Deep Convolutional Generative Adversarial Network
This work addresses a computationally intensive challenge in molecular engineering and biology for researchers in protein structure prediction, but it appears incremental as it applies an existing DCGAN method to protein data without major methodological breakthroughs.
The paper tackles the problem of identifying novel functional protein structures by using a Deep Convolutional Generative Adversarial Network (DCGAN) to classify and generate 3D protein structures, achieving robust annotation against adversarial samples with loss convergence to a local minimum.
Identifying novel functional protein structures is at the heart of molecular engineering and molecular biology, requiring an often computationally exhaustive search. We introduce the use of a Deep Convolutional Generative Adversarial Network (DCGAN) to classify protein structures based on their functionality by encoding each sample in a grid object structure using three features in each object: the generic atom type, the position atom type, and its occupancy relative to a given atom. We train DCGAN on 3-dimensional (3D) decoy and native protein structures in order to generate and discriminate 3D protein structures. At the end of our training, loss converges to a local minimum and our DCGAN can annotate functional proteins robustly against adversarial protein samples. In the future we hope to extend the novel structures we found from the generator in our DCGAN with more samples to explore more granular functionality with varying functions. We hope that our effort will advance the field of protein structure prediction.