Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 May 2;10(9):2227-2234.
doi: 10.1021/acs.jpclett.9b00850. Epub 2019 Apr 22.

Evolution of All-Atom Protein Force Fields to Improve Local and Global Properties

Gül H Zerze  1 , Wenwei Zheng  2 , Robert B Best  3 , Jeetain Mittal  1
Affiliations

Evolution of All-Atom Protein Force Fields to Improve Local and Global Properties

Gül H Zerze et al. J Phys Chem Lett. .

Abstract

Experimental studies on intrinsically disordered and unfolded proteins have shown that in isolation they typically have low populations of secondary structure and exhibit distance scalings suggesting that they are at near-theta-solvent conditions. Until recently, however, all-atom force fields failed to reproduce these fundamental properties of intrinsically disordered proteins (IDPs). Recent improvements by refining against ensemble-averaged experimental observables for polypeptides in aqueous solution have addressed deficiencies including secondary structure bias, global conformational properties, and thermodynamic parameters of biophysical reactions such as folding and collapse. To date, studies utilizing these improved all-atom force fields have mostly been limited to a small set of unfolded or disordered proteins. Here, we present data generated for a diverse library of unfolded or disordered proteins using three progressively improved generations of Amber03 force fields, and we explore how global and local properties are affected by each successive change in the force field. We find that the most recent force field refinements significantly improve the agreement of the global properties such as radii of gyration and end-to-end distances with experimental estimates. However, these global properties are largely independent of the local secondary structure propensity. This result stresses the need to validate force fields with reference to a combination of experimental data providing information about both local and global structure formation.

PubMed Disclaimer

Figures

Figure 1:
Figure 1:
Chain size characteristics of the unfolded and disordered protein ensembles. A) Radius of gyration, Rg, with respect to peptide length, N for the deferent proteins in the data set. Symbols denote the ensemble-averaged values whereas solid lines represent power law _fits. Fit parameters are reported in Table 1. B) Normalized distribution of scaling exponents as evaluated from internal distance scaling. Broken red lines indicate the average scaling exponent, ν, for each force _field. Errors are calculated blocked standard error using two equal, non-overlapping blocks of data. Symbol sizes are larger than the error bars for the data points whose error bars are not visible.
Figure 2:
Figure 2:
DSSP based secondary propensities for three polypeptides, IAPP, TDP-43310-350and CSP. Top panels represent the total fractions of the structures sampled and the other panels show the per-residue secondary structure propensities including alpha, beta and disordered secondary structures. Note we have reduced the DSSP-defined secondary structures into three types for the clarity of interpreting the differences between the three force fields (see the text), black: ff03* (black), ff03w (red), and _03ws (green). Errors are calculated blocked standard error using two equal, non-overlapping blocks of data.
Figure 3:
Figure 3:
NMR order parameters of TDP-43310–350 are evaluated from simulation ensembles. The difference between experiment and simulation (Δ) is reported for all three force fields with the color code shown in the legend. Top panel shows the C α and C β chemical shift deviations (as ΔδCα - ΔδCβ). Average root mean square deviation, hRMSDi, between the simulation and experimental values are reported in the legends. Bottom panel shows the comparison between the scalar coupling constant 3JHNH_ from experiment and simulations.X2 values are shown in the legend. Brown dash lines indicate the errors associated with the prediction methods, to show the reliability of such prediction, the magnitude of which is larger than the differences between the three force fields.14
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. van der Lee R; Buljan M; Lang B; Weatheritt RJ; Daughdrill GW; Dunker AK; Fuxreiter M; Gough J; Gsponer J; Jones DT et al. Classification of intrinsically disordered regions and proteins. Chem. Rev 2014, 114, 6589–6631. - PMC - PubMed
    1. Wright PE; Dyson HJ Intrinsically disordered proteins in cellular signaling and regulation. Nat. Rev. Mol. Cell Biol 2015, 16, 18–29. - PMC - PubMed
    1. Ozenne V; Bauer F; Salmon L; Huang J.-r.; Jensen MR; Segard S; Bernadó P; Charavay C; Blackledge M. Flexible-meccano: a tool for the generation of explicit ensemble descriptions of intrinsically disordered proteins and their associated experimental observables. Bioinformatics 2012, 28, 1463–1470. - PubMed
    1. Schneider R; Huang J.-r.; Yao M; Communie G; Ozenne V; Mollica L; Salmon L; Jensen MR; Blackledge M. Towards a robust description of intrinsic protein disorder using nuclear magnetic resonance spectroscopy. Mol. BioSyst 2012, 8, 58–68. - PubMed
    1. Jensen MR; Ruigrok RW; Blackledge M Describing intrinsically disordered proteins at atomic resolution by NMR. Current opinion in structural biology 2013, 23, 426–435. - PubMed