Design of intrinsically disordered protein variants with diverse structural properties

Area of Science:

• Biochemistry
• Structural Biology
• Computational Biology

Background:

• Intrinsically disordered proteins (IDPs) are crucial for diverse biological functions.
• Designing IDPs with specific conformational properties presents significant challenges due to their inherent dynamics and complexity.
• Expanding the functional repertoire of proteins through design is a key goal in biotechnology.

Purpose of the Study:

• To develop a general computational algorithm for designing intrinsically disordered proteins (IDPs) with specific structural properties.
• To demonstrate the algorithm's capability by generating and validating modified naturally occurring IDPs.
• To explore the relationship between sequence features and protein conformation in IDPs.

Main Methods:

• Development of a general computational algorithm for IDP design.
• Generation of IDP variants with altered compaction, long-range contacts, and phase separation propensity.
• Experimental validation of designed IDPs.
• Analysis of sequence features influencing IDP conformations.
• Application of a machine learning model to capture and accelerate the design process.

Main Results:

• Successful design and experimental validation of IDP variants with predictable structural properties.
• Identification of sequence features that dictate IDP conformations.
• Demonstration that a machine learning model can accurately represent and accelerate the IDP design algorithm.
• Significant expansion of the computational protein design toolbox.

Conclusions:

• The developed algorithm provides a powerful new tool for the computational design of intrinsically disordered proteins.
• This advancement facilitates the creation of novel proteins with functions leveraging the unique properties of protein disorder.
• The integration of machine learning enhances the efficiency and applicability of protein design strategies.