Secondary Structure Prediction for Intrinsically Disordered Proteins

Abstract

Intrinsically disordered proteins (IDPs) or intrinsically disordered regions (IDRs), without well-defined three-dimensional structures, play vital roles in various biological processes, e.g., protein-/nucleic acid-binding, cellular signaling, and protein regulation. One such role of IDPs/IDRs is the intracellular compartmentalization by forming membraneless organelles via liquid-liquid phase separation (LLPS). These phase-separated protein assemblies are also vital in the design of biomimetic protein-based materials. We recently developed a residue-level coarse-grained (CG) model that has yielded valuable information regarding LLPS of various IDPs [1-2]. However, the current model treats protein chains as flexible molecules in which harmonic bonds link amino acid interaction sites. The model lacks chain rigidity and the ability to populate common secondary structure elements due to the lack of angle and dihedral backbone potentials. Here, we present a new CG model which includes these additional backbone potentials to help capture the transient secondary structures of IDPs. The parameters for the pseudo backbone dihedral potential are transferable and obtained from the context-independent helical propensities of twenty naturally occurring amino acids using NMR [3]. We show that the new model can accurately predict the regions of high helical propensities of IDPs such as TDP-43 as well as PPII propensities of proline-rich regions when compared to NMR and all-atom simulation data. Looking forward, this model can be an essential tool in understanding the sequence-specific LLPS of IDPs and high-throughput sequence protein design. 1. G. L. Dignon, W. Zheng, R. B. Best, Y. C. Kim, and J. Mittal, Proc. Nat. Acad. Sci. 115, 9929-9934, 2018. 2. G. L. Dignon, W. Zheng, Y. C. Kim, and J. Mittal, ACS Cent. Sci. 5, 821-830, 2019. 3. R. J. Moreau, et al., J. Am. Chem. Soc. 131, 13107-13116, 2009.