Isopeptor: a tool for detecting intramolecular isopeptide bonds in protein structures

Area of Science:

• Structural biology
• Computational biology
• Bioinformatics

Background:

• Intramolecular isopeptide bonds enhance protein structural stability, particularly in bacterial adhesins and pili.
• Current methods lack systematic detection of these bonds in new molecular structures, leading to potential mis-annotations and modeling errors.

Purpose of the Study:

• To develop and present Isopeptor, a computational tool for predicting intramolecular isopeptide bonds in experimentally determined protein structures.
• To address the limitations in identifying isopeptide bonds, thereby improving the accuracy of protein structural analysis.

Main Methods:

• Isopeptor employs structure-guided template matching using Jess software.
• A logistic regression classifier is utilized, incorporating root mean square deviation and relative solvent accessible area as key predictive features.

Main Results:

• Isopeptor achieved a precision of 1.0 and a recall of 0.947 in tests.
• The tool was validated on Protein Data Bank domains with known intramolecular isopeptide bonds but incorrectly modeled geometries.

Conclusions:

• Isopeptor provides a reliable method for detecting intramolecular isopeptide bonds.
• The tool enhances the accuracy of protein structure analysis and modeling, particularly in cases of mis-annotation.