Exploring the potential of a structural alphabet-based tool for mining multiple target conformations and target flexibility insight
Fig 3
Identification and localization of uPA variable regions.
(A) neqAA (bottom graph) and neqSL (top graph) values for the 261 positions of the MSA computed using the uPA set (SA-conf step 4 output named Neq_graph.pdf). Bars presenting neqSL values are colored according to their secondary structure status: the MSA positions in which all chains have an α-helix conformation in red, a β-strand conformation in green, a loop conformation in gray, and where secondary structure changes occur in purple. In this figure, we added blue rectangles to localize the 9 variable regions highlighted during our analysis of the uPA set and defined as regions corresponding to variable positions with at least (i) four (l = 4) successive variable positions or (ii) one highest variable position (neqSL > 4): R1uPA (positions 33 − 40 + 42 − 45), R2uPA (positions 65–68), R3uPA (positions 101–106), R4uPA (positions 123–124), R5uPA (positions 177 − 181 + 183 − 186), R6uPA (positions 199–200), R7uPA (positions 212–218), R8uPA (positions 231–234), and R9uPA (positions 251–257). We also added red stars “*” to localize the residues involved in the binding site detected by the PockDrug webserver [46]. (B) Graphical representation of the human uPA domain complexed with a chemical inhibitor (PDB ID: 3IG6). The protein is displayed as a cartoon (left figure) and as a surface (right figure) and is colored according to the structural variability of the positions: structurally conserved positions are colored in magenta, weakly structurally variable positions are colored in cyan, structurally variable positions where SL changes do not imply secondary structure changes are colored in “navy blue”, and structurally variable positions where SL changes imply secondary structure changes are colored in dark blue. Residues located at mutated positions are displayed as sticks. Ligand 438 (HETATM code) is displayed in stick form with its Cα atoms colored in yellow. These figures were generated using PyMOL [61] and the script_pymol.pml generated during the step 5 output. A red triangle was added to easily locate the mutated positions. The presented protein corresponds to the first protein ID in the MSA file. Structurally variable regions are located at the protein surface, except R7uPA. R1uPA and R7uPA are within the beta-strand, R2uPA, R5uPA and R9uPA include α-helices and R3uPA, R4uPA, R5uPA, R6uPA and R8uPA are within long and solvent-accessible loops.