Application of topological soliton in modeling protein folding: Recent progress and perspective
Xu-Biao Peng(彭绪彪)1,†, Jiao-Jiao Liu(刘娇娇)1, Jin Dai(戴劲)1,2, Antti J Niemi1,2,‡, and Jian-Feng He(何建锋)1,§
1School of Physics, Beijing Institute of Technology, Beijing 100081, China 2Nordita, Stockholm University, Roslagstullsbacken 23, SE-106 91 Stockholm, Sweden
Proteins are important biological molecules whose structures are closely related to their specific functions. Understanding how the protein folds under physical principles, known as the protein folding problem, is one of the main tasks in modern biophysics. Coarse-grained methods play an increasingly important role in the simulation of protein folding, especially for large proteins. In recent years, we proposed a novel coarse-grained method derived from the topological soliton model, in terms of the backbone Cα chain. In this review, we will first systematically address the theoretical method of topological soliton. Then some successful applications will be displayed, including the thermodynamics simulation of protein folding, the property analysis of dynamic conformations, and the multi-scale simulation scheme. Finally, we will give a perspective on the development and application of topological soliton.
Xu-Biao Peng(彭绪彪)†, Jiao-Jiao Liu(刘娇娇), Jin Dai(戴劲), Antti J Niemi‡, and Jian-Feng He(何建锋)§ Application of topological soliton in modeling protein folding: Recent progress and perspective 2020 Chin. Phys. B 29 108705
Fig. 1.
The Frenet frame vectors (ti, ni, bi) at the i-th Cα atom.
Fig. 2.
The virtual bond and torsion angles (κi, τi) along the backbone Cα chain.
Proteins
PDB ID
Length of sequence
RMSD/Å
Villin headpiece 35
1YRF
29 aa
0.38
Myoglobin
1ABS
154 aa
0.78
HIV-1 reverse transcriptase protein
3DLK
18 aa
1.13
λ-repressor
1LMB
84 aa
0.51
Human islet amyloid polypeptide
2L86
37 aa
1.17
Myc proto-oncogene protein
1NKP
88 aa
0.98
Amyloid intra-cellular domain
3DXC
28 aa
0.46
Engrailed homeodomain
2JWT
61 aa
0.67
Parvalbumin-β
2PVB
57 aa
1.28
Table 1.
The proteins that have been fitted using topological soliton model.
Fig. 3.
The radius of gyration evolution with temperature increasing. The gray, red, yellow dashed lines are corresponding to the real temperatures of 25 °C, 75 °C, and 90 °C, respectively. Reproduced with permission from Ref. [46].
Fig. 4.
The susceptibility of helical denucleation. Three transition temperatures are labeled as T1, T2, TE, representing the two transition temperatures for the radius of gyration and for the energy, respectively. The colored thick dash lines are the same as in Fig. 3. Reproduced with permission from Ref. [46].
Fig. 5.
The superimposition of the soliton model and PDB structures. Left panel is for 2L86 and right panel is for 3DXC. The light blue is from PDB structure and the red is from soliton model. Reproduced with permission from Refs. [35,36].
Fig. 6.
The conformational clusters for 2L86 at low temperature. Panel (a) is the conformational landscape, and panel (b) is the representative structures. Reproduced with permission from Ref. [35].
Fig. 7.
The conformational clusters for 3DXC at low temperature. Panel (a) is the energy landscape of the conformational ensemble, where the red triangle denotes the initial structure in PDB. Panel (b) is the representative structures whose energies are lower than the initial structures. Reproduced with permission from Ref. [36].
Fig. 8.
The corresponding soliton mobility of the clusters in 3DXC. Reproduced with permission from Ref. [36].
Fig. 9.
The conformational clusters for 1NKP at low temperature. Top panel is the energy landscape of the conformational ensemble, where the red triangle denotes the initial structure in PDB. The bottom panel is the corresponding conformational landscape projected from top panel. Reproduced with permission from Ref. [38].
Fig. 10.
The stability comparison among clusters in 1NKP, (a) comparison of the RMSD evolutions in MD simulations with initial conformations in clusters 1, 4, and 5 (denoted as PDB in the legend), (b) a comparison of the radius of gyration evolutions in MD simulations with initial conformations in clusters 1, 4, and 5 (denoted as PDB in the legend), (c) the conformational landscape evolution for MD simulations with initial conformation from cluster 1, (d) the conformational landscape evolution for MD simulations with initial conformation from cluster 5. Reproduced with permission from Ref. [38].
Fig. 11.
The folding index evolution in MD simulation. The top panel is the folding index in the entire MD simulation process, and bottom panel is a zoom in of top panel in frame 3950–4000. Reproduced with permission from Ref. [37].
Fig. 12.
The sidechain soliton motion in the N-terminal of the protein during the MD simulation. The two panels show the same data, but from different perspectives, for the first ten residues. Reproduced with permission from Ref. [37].
[1]
Dill K, Ozkan S B, Weikl T R, Chodera J D, Voelz V A 2007 Curr. Opin. Struc. Biol. 17 342 DOI: 10.1016/j.sbi.2007.06.001
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
