Conformational analysis of long spacers in PROSITE patterns

Area of Science:

• Structural bioinformatics
• Computational biology
• Biochemistry

Background:

• Sequence motifs, like PROSITE patterns, are crucial for identifying protein function.
• Understanding the 3D structure of these motifs, including variable regions, is key to accurate functional prediction.

Purpose of the Study:

• To investigate if variable spacer sequences within PROSITE patterns exhibit a consensus three-dimensional structure.
• To explore the relationship between sequence variability and backbone conformational heterogeneity in protein motifs.

Main Methods:

• Analysis of 3D protein structures containing PROSITE patterns with fixed-length spacers (>3 residues).
• Computation of backbone phi, psi, and side-chain chi1 dihedral order parameters to assess structural similarity.
• Introduction and application of a 'bias coefficient' to quantify residue-type variability at each pattern position.

Main Results:

• Backbone conformational heterogeneity does not always correlate with residue-type variability within a motif.
• Long spacer regions within motifs can adopt well-defined backbone conformations.
• PROSITE patterns can be refined into multiple regular expressions, each representing a distinct backbone conformation.

Conclusions:

• Variable regions in sequence motifs can possess defined structural properties.
• Observed structural consensus within motifs can enhance the accuracy of protein function prediction based on sequence alone.