Abstract
BackgroundProtein structure prediction (PSP), which is usually modeled as a computational optimization problem, remains one of the biggest challenges in computational biology. PSP encounters two difficult obstacles: the inaccurate energy function problem and the searching problem. Even if the lowest energy has been luckily found by the searching procedure, the correct protein structures are not guaranteed to obtain.ResultsA general parallel metaheuristic approach is presented to tackle the above two problems. Multi-energy functions are employed to simultaneously guide the parallel searching threads. Searching trajectories are in fact controlled by the parameters of heuristic algorithms. The parallel approach allows the parameters to be perturbed during the searching threads are running in parallel, while each thread is searching the lowest energy value determined by an individual energy function. By hybridizing the intelligences of parallel ant colonies and Monte Carlo Metropolis search, this paper demonstrates an implementation of our parallel approach for PSP. 16 classical instances were tested to show that the parallel approach is competitive for solving PSP problem.ConclusionsThis parallel approach combines various sources of both searching intelligences and energy functions, and thus predicts protein conformations with good quality jointly determined by all the parallel searching threads and energy functions. It provides a framework to combine different searching intelligence embedded in heuristic algorithms. It also constructs a container to hybridize different not-so-accurate objective functions which are usually derived from the domain expertise.
Highlights
Given the protein’s amino acid sequence, protein structure prediction (PSP) is to predict the tertiary structure of its native state
According to the hypothesis that the native structure always adapts to the status with the lowest free energy [3], Protein structure prediction (PSP) is usually converted to a single-objective optimization problem (SOP) which tries to minimize the free energy value of the predicted structure
There are two biggest major obstacles for solving PSP problem [1,7]: the first is that the searching is always inefficient even if the current computing power is increasing exponentially, and the second is that as the objective function for minimizing, energy function itself cannot accurately measure the free energy of a computer-generated conformation because we are lack of complete knowledge on measuring free energy based on protein conformation surrounded by the complex bio-environment
Summary
Given the protein’s amino acid sequence, protein structure prediction (PSP) is to predict the tertiary structure of its native state. According to the hypothesis that the native structure always adapts to the status with the lowest free energy [3], PSP is usually converted to a single-objective optimization problem (SOP) which tries to minimize the free energy value of the predicted structure (see the left part of Figure 1). Such problem and its variants have been proved as NP-hard problems [4,5]. Even if the lowest energy has been luckily found by the searching procedure, the correct protein structures are not guaranteed to obtain
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call