Integrating structure-based machine learning and co-evolution to investigate specificity in plant sesquiterpene synthases
Fig 3
A. Thirty cation-specific residues found by the structure-based Clf-str predictor on the tobacco epi-aristolochene synthase (TEAS) structure, colored by region. Terpene synthase motifs are labelled in purple, Mg+2 ions in pink, and a substrate analog in orange and dark pink. The N-terminal domain is shaded with a lower opacity. B. Sequence conservation of Clf-str cation-specific residues across farnesyl and nerolidyl cation-specific STSs, labelled by region and residue position in the TEAS structure. The height of a letter represents its frequency in that position. An insertion/deletion is represented by a black ‘X’. Residue positions which are deleted in the TEAS structure are represented by ‘-’s and correspond to residue 627 and residue 687 respectively in the Abies grandis α-bisabolene synthase structure (PDB ID: 3SDU). Note that if, for a given position, the residues in both logos are similar, this indicates that in spite of similarity in sequence at this position, the farnesyl and nerolidyl cation-specific enzymes are structurally different.