Searching for three-dimensional secondary structural patterns in proteins with ProSMoS.

Area of Science:

• Structural bioinformatics
• Computational biology
• Protein structure analysis

Background:

• Identifying functional links between evolutionarily distant proteins relies on structural comparisons.
• Existing methods often focus on precise coordinate matching, missing broader structural similarities.
• A need exists for tools that capture protein fold architecture irrespective of minor 3D variations.

Purpose of the Study:

• To develop a computational tool, ProSMoS, that emulates expert structural motif identification.
• To enable the discovery of proteins with similar secondary structures, topological connections, and spatial architecture.
• To overcome limitations of existing methods by tolerating deviations in 3D coordinates.

Main Methods:

• ProSMoS utilizes pre-defined secondary structural elements from a given protein structure.
• It constructs a meta-matrix representing interactions between elements, including orientation, handedness, and length.
• This meta-matrix serves as a general, yet informative, definition of a protein fold for searching.

Main Results:

• ProSMoS successfully identifies proteins matching a query structural motif's meta-matrix within the Protein Data Bank (PDB).
• The program's performance was evaluated and compared against other existing tools.
• An analysis of beta-Grasp containing proteins demonstrated the tool's utility.

Conclusions:

• ProSMoS provides a novel approach to protein structure motif searching by abstracting structural information into meta-matrices.
• The program effectively identifies proteins with conserved structural motifs, regardless of precise 3D structural differences.
• ProSMoS is available for non-commercial use, facilitating further research in structural bioinformatics.