SiteMotif: A graph-based algorithm for deriving structural motifs in Protein Ligand binding sites.

Area of Science:

• Biochemistry
• Structural Biology
• Bioinformatics

Background:

• Protein sequence alignment is common but limited for proteins with similar ligand binding but divergent sequences or folds.
• Existing methods compare binding sites but cannot derive independent structural motifs.
• Proteins with functional similarity often share structural similarities only in their ligand-binding sites.

Purpose of the Study:

• To develop a novel algorithm, SiteMotif, for deriving sequence-order independent structural motifs of protein binding sites.
• To enable the comparison of binding sites across proteins lacking sequence or fold similarity.
• To provide a tool for uncovering mechanistic insights into ligand interactions.

Main Methods:

• Development of the SiteMotif algorithm for comparing multiple protein binding sites.
• Derivation of sequence-order independent structural site motifs.
• Testing and benchmarking against three existing binding site comparison methods.

Main Results:

• SiteMotif successfully derives structural motifs from protein binding sites, independent of sequence or fold similarity.
• The algorithm demonstrates superior performance compared to three current binding site comparison methods.
• New structural motifs of spatially conserved residues were identified in proteins with no sequence or fold-level similarity.

Conclusions:

• SiteMotif effectively identifies functional relationships between proteins based on binding site structure, irrespective of sequence or fold homology.
• The algorithm can provide key mechanistic insights into ligand binding modes.
• SiteMotif has the potential to predict ligand types and enhance functional annotation accuracy for proteins.