Predicting 3D Structures of Lasso Peptides.

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• Lasso peptides (LaPs) are a large class of ribosomally synthesized and post-translationally modified peptides (RiPPs) with unique slipknot-like structures.
• Examples of LaPs exhibit diverse functions, including antibiotic, enzyme inhibition, and molecular switching activities.
• Despite numerous predicted sequences, only about 50 distinct LaP structures have been characterized due to prediction challenges.

Purpose of the Study:

• To develop a computational tool for accurate prediction of lasso peptide 3D structures.
• To create the largest in silico-predicted lasso peptide structure database.

Main Methods:

• Developed LassoPred, a novel computational tool featuring a classifier for sequence annotation (ring, loop, tail) and a constructor for 3D structure generation.
• Utilized LassoPred to predict 3D structures for 4,749 unique LaP core sequences.

Main Results:

• Successfully predicted 3D structures for a large cohort of lasso peptides.
• Created the most extensive database of in silico-predicted LaP structures to date.
• LassoPred demonstrates superior performance in predicting LaP structures compared to existing tools.

Conclusions:

• LassoPred effectively addresses the challenge of predicting complex LaP structures.
• The generated database and tool will accelerate structure-function relationship studies for LaPs.
• This resource will aid in the discovery of novel functional lasso peptides for chemical and biomedical applications.