Proteins are responsible for almost all molecular processes that are essential for life on Earth. The properties of proteins found in nature have been optimized through natural evolution to meet the requirements of living organisms. This optimization does not facilitate the use of natural proteins outside of their natural niche or for alternative purposes. Protein engineering strategies have therefore been developed to customize proteins to the specific demands generated by applications in health, biotechnologies, green chemistry, and nanotechnologies, for example. Throughout the past decade, advances in technologies, algorithms, and high-performance computing have transformed the protein engineering field. Computational protein design approaches are now deep-rooted in the processes of tailoring proteins for various applications. Numerous computational tools and methods are developed and improved every year to meet the increasing demands and challenges of protein engineering. Moreover, the rapid progress of artificial intelligence techniques combined with the accumulation of sequence and structure protein data are currently shaping the field. More and more design methodologies exploit data in an unprecedented way. Especially, “deep learning” will be likely increasingly important for protein design.
This Research Topic aims at highlighting recent advances in computational approaches for rational or semi-rational protein design and their application to engineer proteins with improved or radically new capacities. More specifically, the goal is to gather articles that describe research aimed at improving protein design reliability, notably through the development of (i) methods able to efficiently take into account the backbone flexibility and molecular functional dynamics into protein design; ii) data-driven approaches based on protein structures, sequences or experimentally measured properties and using advanced machine learning algorithms, including deep neural networks; iii) original de novo design methods; iv) effective scoring functions for protein design. In this Research Topic, we will thus show, with examples of their applications, the most advanced methods that are anticipated to revolutionize protein design for health and bio(nano)technologies.
This collection welcomes Original Research articles, Reviews, Mini Review, and Perspectives on recent, novel, and promising computational and data-driven protein design approaches as well as their applications to address a variety of challenges in health and bio(nano)technologies. Areas to be covered in this Research Topic may include, but are not limited to:
• Molecular flexibility and dynamics in protein design
• Data-driven protein design approaches
• Fragment-based computational protein design
• AI approaches for designing new proteins
• Protein design with deep learning neural networks
• Computational methods and tools to improve protein stability, affinity, and specificity
• De novo computational protein design
• Scoring functions for protein design
• Computational protein design applications in health, biotechnology, and nanotechnology
Dr. Bruce Donald is the founder of Ten63 Therapeutics, Inc. and serves as a chair of the SAB
Proteins are responsible for almost all molecular processes that are essential for life on Earth. The properties of proteins found in nature have been optimized through natural evolution to meet the requirements of living organisms. This optimization does not facilitate the use of natural proteins outside of their natural niche or for alternative purposes. Protein engineering strategies have therefore been developed to customize proteins to the specific demands generated by applications in health, biotechnologies, green chemistry, and nanotechnologies, for example. Throughout the past decade, advances in technologies, algorithms, and high-performance computing have transformed the protein engineering field. Computational protein design approaches are now deep-rooted in the processes of tailoring proteins for various applications. Numerous computational tools and methods are developed and improved every year to meet the increasing demands and challenges of protein engineering. Moreover, the rapid progress of artificial intelligence techniques combined with the accumulation of sequence and structure protein data are currently shaping the field. More and more design methodologies exploit data in an unprecedented way. Especially, “deep learning” will be likely increasingly important for protein design.
This Research Topic aims at highlighting recent advances in computational approaches for rational or semi-rational protein design and their application to engineer proteins with improved or radically new capacities. More specifically, the goal is to gather articles that describe research aimed at improving protein design reliability, notably through the development of (i) methods able to efficiently take into account the backbone flexibility and molecular functional dynamics into protein design; ii) data-driven approaches based on protein structures, sequences or experimentally measured properties and using advanced machine learning algorithms, including deep neural networks; iii) original de novo design methods; iv) effective scoring functions for protein design. In this Research Topic, we will thus show, with examples of their applications, the most advanced methods that are anticipated to revolutionize protein design for health and bio(nano)technologies.
This collection welcomes Original Research articles, Reviews, Mini Review, and Perspectives on recent, novel, and promising computational and data-driven protein design approaches as well as their applications to address a variety of challenges in health and bio(nano)technologies. Areas to be covered in this Research Topic may include, but are not limited to:
• Molecular flexibility and dynamics in protein design
• Data-driven protein design approaches
• Fragment-based computational protein design
• AI approaches for designing new proteins
• Protein design with deep learning neural networks
• Computational methods and tools to improve protein stability, affinity, and specificity
• De novo computational protein design
• Scoring functions for protein design
• Computational protein design applications in health, biotechnology, and nanotechnology
Dr. Bruce Donald is the founder of Ten63 Therapeutics, Inc. and serves as a chair of the SAB
