CoCoPRED Introduction

Coiled-coil is composed of two or more helices that are wound around each other. It widely exists in proteins and has been discovered to play a variety of critical roles in biology processes. While the periodicity of regular helix is 3.6 residues per turn, that of the helix in the coiled-coil is only 3.5. Due to the reduced periodicity, the coiled-coil has a supercoiled structure. Generally, there are three types of structural features in coiled-coil: coiled-coil domain (CCD), oligomeric state, and register. However, most of the developed computational tools only focus on one of them.

In this study, we describe a new model, CoCoPRED, which is based on convolutional layers, bidirectional Long Short-Term Memory, and attention mechanism. It has three networks, i.e., CCD network, oligomeric state network, and register network, corresponding to the three types of structural features in coiled-coil. This means CoCoPRED has the ability of fulfilling the comprehensive prediction. Through a rigid 5-fold cross-validation experiment, we demonstrate that CoCoPRED has largely exceeded the state-of-the-art models on both CCD prediction and oligomeric state prediction. Analysis on the prediction results suggests the CCD prediction may be a performance indicator of the oligomeric state prediction in CoCoPRED. The attention heads in CoCoPRED indicate that registers a, b, and e are more crucial for the oligomeric state prediction.