Efficient Conformational Search Based on Structural Dissimilarity Sampling: Applications for Reproducing Structural...

Area of Science:

• Computational Biology
• Structural Biology
• Biophysics

Background:

• Protein conformational changes are crucial for biological function.
• Efficient methods are needed to explore diverse protein states.
• Traditional molecular dynamics can be time-consuming for capturing transitions.

Purpose of the Study:

• To introduce Structural Dissimilarity Sampling (SDS) as an efficient method for conformational search.
• To demonstrate SDS's ability to promote essential protein structural transitions.
• To validate SDS's effectiveness in sampling protein folding processes.

Main Methods:

• Structural Dissimilarity Sampling (SDS) involves selecting initial structures based on dissimilarity.
• Conformational resampling uses diverse structures at subspace edges for short molecular dynamics (MD) simulations.
• Iterative cycles of structure selection and resampling expand conformational sampling.

Main Results:

• SDS successfully reproduced the open-to-closed state transitions of maltodextrin binding protein (MBP) in nanosecond timescales.
• SDS achieved transitions in 25 cycles (250 ns), while a 500 ns conventional MD failed.
• SDS effectively sampled native states for fast-folding proteins like chignolin, Trp-cage, and villin.
• SDS showed robustness, with sampling efficiency being relatively insensitive to the number of base sets used for dissimilarity characterization.

Conclusions:

• Structural Dissimilarity Sampling (SDS) is a robust and efficient method for exploring protein conformational landscapes.
• SDS accelerates the discovery of functionally relevant protein states and folding pathways.
• The method holds promise for various applications in structural biology and drug discovery.