Fast search algorithms for computational protein design

• Published in: Journal of computational chemistry 2016-05, Vol.37 (12), p.1048-1058
• Main Authors: Traoré, Seydou, Roberts, Kyle E., Allouche, David, Donald, Bruce R., André, Isabelle, Schiex, Thomas, Barbe, Sophie
• Format: Article
• Language: English
• Subjects: Algorithms; Amino Acid Sequence; Backbone; Bioinformatics; Buildings; Chain branching; Chains; Chemical Sciences; Cheminformatics; Chemistry; Combinatorial analysis; Compounding; Computational Biology; computational protein design; Computer Science; computer-aided protein design; Cost function; cost function networks; Design; deterministic search methods; Drug Design; Enumeration; exact combinatorial optimization; global minimum energy conformation; Lower bounds; near-optimal solutions; Optimization; Protein Conformation; Proteins; Proteins - chemistry; Search algorithms; search heuristics; Searching; State of the art; Technology
• ISSN: 0192-8651; 1096-987X
• DOI: 10.1002/jcc.24290

One of the main challenges in computational protein design (CPD) is the huge size of the protein sequence and conformational space that has to be computationally explored. Recently, we showed that state‐of‐the‐art combinatorial optimization technologies based on Cost Function Network (CFN) processing allow speeding up provable rigid backbone protein design methods by several orders of magnitudes. Building up on this, we improved and injected CFN technology into the well‐established CPD package Osprey to allow all Osprey CPD algorithms to benefit from associated speedups. Because Osprey fundamentally relies on the ability of A* to produce conformations in increasing order of energy, we defined new A* strategies combining CFN lower bounds, with new side‐chain positioning‐based branching scheme. Beyond the speedups obtained in the new A*‐CFN combination, this novel branching scheme enables a much faster enumeration of suboptimal sequences, far beyond what is reachable without it. Together with the immediate and important speedups provided by CFN technology, these developments directly benefit to all the algorithms that previously relied on the DEE/ A* combination inside Osprey* and make it possible to solve larger CPD problems with provable algorithms. © 2016 Wiley Periodicals, Inc. Computational protein design (CPD) through Cost Function Networks (CFN) provides important speedups to explore large sequence‐conformation spaces and provably identifies the sequence with the conformation of optimal stability (Global Minimum Energy Conformation, GMEC). In addition to quickly finding the GMEC of highly complex protein design problems, CFN‐based methods also enable the efficient enumeration of suboptimal solutions. These approaches offer an attractive alternative to the usual CPD methods and were implemented in the well‐established CPD package Osprey.