Creative BioMart is a well-known expert in the development of a variety of computational analysis methods to predict the three-dimensional structure of proteins, and predict its functional conformation from the amino acid sequence of the protein. With years of experience, we provide customized fold recognition service to precisely meet customer requirements.

Introduction of Fold Recognition

Fold recognition, also known as protein threading. Used to model proteins that have the same fold as proteins of known structure but do not have homologous proteins of known structure. The number of protein fold types in nature is limited, and although many proteins share low sequence similarity, they may still have the same fold type, which is the theoretical basis for fold recognition. The continued development of such methods has had a major impact on structural biology, providing researchers with an increasing ability to accurately model 3D protein structures using very evolu-tionary distant fold template.

There are generally two steps to follow before folding recognition:

(1) Check for the absence of homologous proteins with known structures. If there are homologous structures, it is best to use homology modeling.

(2) Whether the secondary structure and accessibility of the protein are helpful in evaluating the fold recognition results.

Fig 1. Schematic representation of protein threading in simpler way. (Majumder P, 2020)

Services

For a sequence to be tested, if its corresponding fold type has been experimentally determined, how to find out its corresponding fold type by a suitable computational method is the core problem to be solved in fold recognition. As a leading service provider of protein engineering, Creative BioMart has successfully developed fold-recognition (or threading) methods to predict query protein structure, especially when the query protein shares low sequence-level identity (i.e. <25%) with other proteins with known structure. We provide a one-stop shop for protein fold recognition modeling:

• Construction of structural template database: after removing protein structures with high sequence similarity, we selected protein structures from databases such as PDB, FSSP, SCOP or CATH as structural templates.
• Design of scoring function: based on the knowledge of known relationships between structures and sequences, we design a good scoring function to measure the fitness between target sequences and templates, including mutation potential, environmental fitness potential, pairwise potential, secondary structure compatibility and vacancy penalty.
• Threading alignment: align the target sequence with each structural template by optimizing the designed scoring function.
• Threading prediction: a structural model of the target is constructed by placing the backbone atoms of the target sequence at the aligned backbone positions of the selected structural template.

Fold recognition does not need to predict secondary structure, and can directly predict three-dimensional structure, thus bypassing the current limit of secondary structure prediction accuracy of no more than 65%. Homology modeling applies to those targets that have homologous proteins of known structure, while fold recognition applies to those targets where only fold-level homology is found. When no significant homology is found, fold recognition can be predicted based on structural information.

Features of Fold Recognition

• Accurate sequence and structure alignment algorithms for long-range homology identification.
• 1D-3D profiles for similarity comparison.
• Template library construction.
• Using functional domain information and sequential evolution information.
• A multithreaded approach to identifying convergent solutions.
• Hybrid approach combining threading with other structural prediction methods.
• Validation of structure/sequence matches by energy calculatio.

Creative BioMart is committed to identifying the correct structural folding in the known template protein structure of the target protein, helping the development of different fields such as drug development and materials science. We will work with you to develop the most appropriate strategy and provide the most meaningful data for your research for accelerating the research of life sciences. If you are interested in our services, please do not hesitate to contact us for more information.

Reference

1. Majumder P. (2020) Computational Methods Used in Prediction of Protein Structure[M]. Statistical Modelling and Machine Learning Principles for Bioinformatics Techniques, Tools, and Applications. Springer, Singapore. 119-133.

• De Novo Prediction Service
• Homology Modeling Service