Fast design of arbitrary length loops in proteins using InteractiveRosetta.

Area of Science:

• Computational Biology
• Protein Engineering
• Structural Bioinformatics

Background:

• Exploring insertions and deletions (indels) is crucial for ab initio protein design.
• Interactive protein design faces challenges in speed and accuracy when handling variable-length indels.

Purpose of the Study:

• To introduce a new module, INDEL, for InteractiveRosetta.
• To enable users to specify desired indel lengths and rapidly obtain low-energy backbones.

Main Methods:

• Geometric hashing of loop anchor points using Cα-Cα and Cβ-Cβ distances for fast loop searching.
• Mapping hashed anchors to a pre-compiled file of non-redundant protein backbone coordinates.
• Filtering loops for collisions and returning poly-alanine structures for integration into design templates.

Main Results:

• The INDEL module successfully generated viable loops in 100% of 500 attempts for lengths ranging from 3 to 20 residues.
• The module returns low-energy backbones within seconds.
• Sidechain addition is facilitated using existing RosettaDesign tools.

Conclusions:

• INDEL provides an efficient method for exploring indel possibilities in protein design.
• The module was successfully applied to redesign a domain-swapping loop in T7-endonuclease I, altering its DNA specificity.
• This advancement supports the design of proteins with novel functions and structures.