BWM: A Novel, Provable, Ensemble-based Dynamic Programming Algorithm for Sparse Approximations of Computational Protein Design

被引:15
|
作者
Jou, Jonathan D. [1 ]
Jain, Swati [1 ,2 ,3 ,4 ]
Georgiev, Ivelin S. [1 ,6 ]
Donald, Bruce R. [1 ,2 ,5 ]
机构
[1] Duke Univ, Dept Comp Sci, Durham, NC 27708 USA
[2] Duke Univ, Med Ctr, Dept Biochem, Durham, NC 27708 USA
[3] Duke Univ, Dept Computat Biol, Durham, NC 27708 USA
[4] Duke Univ, Bioinformat Program, Durham, NC 27708 USA
[5] Duke Univ, Dept Chem, Durham, NC 27708 USA
[6] NIAID, Vaccine Res Ctr, NIH, 9000 Rockville Pike, Bethesda, MD 20892 USA
基金
美国国家卫生研究院;
关键词
protein design; sparse residue interaction graphs; provable algorithms; branch-decomposition; dynamic programming; ensemble-based algorithms; OSPREY; DEAD-END ELIMINATION; SIDE-CHAIN; GRAMICIDIN SYNTHETASE; SEARCH ALGORITHM; PREDICTION; REDESIGN; CONFORMATION; SPECIFICITY; RESISTANCE; POTENCY;
D O I
10.1089/cmb.2015.0194
中图分类号
Q5 [生物化学];
学科分类号
071010 ; 081704 ;
摘要
Sparse energy functions that ignore long range interactions between residue pairs are frequently used by protein design algorithms to reduce computational cost. Current dynamic programming algorithms that fully exploit the optimal substructure produced by these energy functions only compute the GMEC. This disproportionately favors the sequence of a single, static conformation and overlooks better binding sequences with multiple low-energy conformations. Provable, ensemble-based algorithms such as A* avoid this problem, but A* cannot guarantee better performance than exhaustive enumeration. We propose a novel, provable, dynamic programming algorithm called Branch-Width Minimization* (BWM*) to enumerate a gap-free ensemble of conformations in order of increasing energy. Given a branch-decomposition of branch-width w for an n-residue protein design with at most q discrete side-chain conformations per residue, BWM* returns the sparse GMEC in O(nw(2) q(3/2w)) time and enumerates each additional conformation in merely O(n log q) time. We define a new measure, Total Effective Search Space (TESS), which can be computed efficiently a priori before BWM* or A* is run. We ran BWM* on 67 protein design problems and found that TESS discriminated between BWM*-efficient and A*-efficient cases with 100% accuracy. As predicted by TESS and validated experimentally, BWM* outperforms A* in 73% of the cases and computes the full ensemble or a close approximation faster than A*, enumerating each additional conformation in milliseconds. Unlike A*, the performance of BWM* can be predicted in polynomial time before running the algorithm, which gives protein designers the power to choose the most efficient algorithm for their particular design problem.
引用
收藏
页码:413 / 424
页数:12
相关论文
共 50 条
  • [21] Heuristic Dynamic Programming Iterative Algorithm Design Based on BP Neural Network
    Zhao, Yu
    Yang, Jiye
    ADVANCES IN APPLIED SCIENCES AND MANUFACTURING, PTS 1 AND 2, 2014, 850-851 : 893 - 896
  • [22] Automatic Design Method of Dynamic Systems Based on Hungarian Algorithm and Genetic Programming
    Li Shaobo
    Yang Guanci
    Xie Qingsheng
    2008 4TH INTERNATIONAL CONFERENCE ON WIRELESS COMMUNICATIONS, NETWORKING AND MOBILE COMPUTING, VOLS 1-31, 2008, : 2294 - 2297
  • [23] Design and Implementation of Pairwise Sequence Alignment Algorithm Components Based on Dynamic Programming
    Shi H.
    Zhou W.
    Jisuanji Yanjiu yu Fazhan/Computer Research and Development, 2019, 56 (09): : 1907 - 1917
  • [24] A novel evolutionary algorithm based on orthogonal design for dynamic optimizations
    Zeng, SY
    Liu, R
    Yao, SZ
    Kang, LS
    PROGRESS IN INTELLIGENCE COMPUTATION & APPLICATIONS, 2005, : 275 - 284
  • [25] A novel technique for the design of hybrid composite laminates based on dynamic programming and dynamic tree trimming
    Sanz-Corretge, Javier
    Echeverria, Mikel
    STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION, 2018, 57 (04) : 1507 - 1521
  • [26] A novel technique for the design of hybrid composite laminates based on dynamic programming and dynamic tree trimming
    Javier Sanz-Corretge
    Mikel Echeverría
    Structural and Multidisciplinary Optimization, 2018, 57 : 1507 - 1521
  • [27] A novel density-based ensemble learning algorithm with application to protein structural classification
    Homayouni, Haleh
    Mansoori, Eghbal G.
    INTELLIGENT DATA ANALYSIS, 2017, 21 (01) : 167 - 179
  • [28] Boosting AND/OR-Based Computational Protein Design: Dynamic Heuristics and Generalizable UFO
    Pezeshki, Bobak
    Marinescu, Radu
    Ihler, Alexander
    Dechter, Rina
    UNCERTAINTY IN ARTIFICIAL INTELLIGENCE, 2023, 216 : 1662 - 1672
  • [29] A novel adaptive heuristic dynamic programming-based algorithm for aircraft confrontation games
    Mao, Yi
    Chen, Zhijie
    Yang, Yi
    Hu, Yuxin
    FUNDAMENTAL RESEARCH, 2021, 1 (06): : 792 - 799
  • [30] A novel contour extraction algorithm based on dynamic programming in sausage visual inspection system
    Gao, Liang
    Chen, Shuai
    Ma, Yue
    PROCEEDINGS OF THE 2ND ANNUAL INTERNATIONAL CONFERENCE ON ELECTRONICS, ELECTRICAL ENGINEERING AND INFORMATION SCIENCE (EEEIS 2016), 2016, 117 : 76 - 82