A Data‐Driven Accelerated Sampling Method for Searching Functional States of Proteins. Issue 4 (4th January 2019)

• Record Type: Journal Article
• Authors: Zhu, Qiang; Yuan, Yigao; Ma, Jing; Dong, Hao
• Is Part Of: Advanced theory and simulations. Volume 2:Issue 4(2019)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2019-01-04
• Subjects: calmodulin -- enhanced sampling -- essential space -- molecular dynamics simulations -- principle component analysis; Science -- Simulation methods -- Periodicals; Science -- Methodology -- Periodicals; Engineering -- Simulation methods -- Periodicals; Engineering -- Methodology -- Periodicals; 507.21
• DOI: 10.1002/adts.201800171
• Languages: English
• ISSNs: 2513-0390

Abstract

Abstract: Protein exhibits distinct characteristics in different functional states. The lack of structural information for proteins hinders the understanding of their function. Here, a data‐driven accelerated (DA2) sampling method is proposed, which is capable of searching new functional states of protein from a known structure with high efficiency. The key function of DA2 sampling is to drive the conformational change of protein along its intrinsic motion without introducing biased potential/force, where principle component analysis is applied on‐the‐fly to reduce the highly redundant information generated by molecular dynamics simulations. In this work, the capacity and accuracy of DA2 sampling are validated by using alanine dipeptide. This protocol is then applied to search for the closed state of N‐terminal calmodulin (nCaM) from the open one. The identified structure resembles the crystal structure of nCaM in its closed state, with a root‐mean‐square deviation between the two of only 1.8 Å. Interestingly, independent DA2 samplings disclose different open‐to‐closed transition pathways for nCaM, which is likely to have implications for its biological functions. Therefore, DA2 sampling is expected to play important roles in exploring functional states of a broad spectrum of proteins at atomic level that are not easily determined experimentally. Abstract : The data‐driven accelerated (DA2) method is a new enhanced sampling method which drives the conformational change of protein along its intrinsic motion. Armed with normal mode analysis and principle component analysis, DA2 can sample the conformational space more efficiently than conventional molecular dynamics.