This page provides some very basic examples on the features of PDB2PQR. It is under continual development and suggestions are appreciated!
This section outlines the basic process of adding hydrogens and assigning charge/radius parameters to an otherwise complete PDB structure.
This 3-finger toxin structure is available at high resolution (1.9 Å) and has all its heavy atoms present in the PDB file. We'll use one of the PDB2PQR servers to add hydrogens to this protein and optimize their positions.
1FAS into the PDB ID field. Once the calculations are complete, you should get a web page with a link to the new PQR file. You can download this PQR file and view it in your favorite molecular visualization package (e.g., VMD, PyMOL, or PMV). For comparison, you might download the the original PDB file and compare the PDB2PQR-generated structure with the original to see where hydrogens were placed.
This kinase structure is available at somewhat lower (2.5 Å) resolution and is missing several sidechain atoms as well as portions of its sequence. We'll use this example to demonstrate how PDB2PQR can add missing sidechain atoms to an imcomplete structure but cannot fill in missing regions of the backbone. In particular, we'll use PDB2PQR to add/optimize hydrogens, reconstruct sidechains K53, N65, R140, E154, Q192, Y195, E221, N222, K225, E228, K232, and Q272 from model geometries, and assign parameters.
1A06 into the PDB ID field. Once the calculations are complete, you should see a web page with a link to the new PQR file and warnings about incomplete and poorly-positioned portions of the PQR structure. In particular, PDB2PQR will complain about missing regions between K53 and E64 and between F163 and P182. PDB2PQR may also complain "Unable to debump VAL A 189", referring to bad contacts between V189 and other residues that it was unable to resolve. You can download the resulting PQR file and view it in your favorite molecular visualization package (e.g., VMD, PyMOL, or PMV). For comparison, you might download the the original PDB file and compare the PDB2PQR-generated structure with the original to see where hydrogens were placed.
Interested users should read Li H, Robertson AD, Jensen JH. Very Fast Empirical Prediction and Rationalization of Protein pKa Values. Proteins, 61, 704-721 (2005). for a much more complete description and analysis of titration state assignment using PROPKA. The examples here are taken from this paper. Nearly all of these examples can be reproduced using PDB2PQR/PROPKA, we give a single example here for demonstration purposes.
The PDB structure 1HPX includes HIV-1 protease complexed with an inhibitor at 2.0 Å resolution. HIV-1 protease has two chains; residue D25 is anionic on one chain and neutral on the other -- these titration states are important in the role of D25 as an acid in the catalytic mechanism.
1HPX into the PDB ID field. Once the calculations are complete, you should see a web page with a link to the PROPKA output, a new PQR file, and warnings about the ligand KNI (since we didn't choose to parameterize it in this calculation -- see below). You can download the resulting PQR file and view it in your favorite molecular visualization package (e.g., VMD, PyMOL, or PMV). For comparison, you might download the the original PDB file and compare the PDB2PQR-generated structure with the original to see where hydrogens were placed.
This section outlines the parameterization of ligands using the PEOE_PB methods (see Czodrowski P, Dramburg I, Sotriffer CA, Klebe G. Development, validation, and application of adapted peoe charges to estimate pka values of functional groups in protein-ligand complexes. Proteins. 65 (2), 424-37, 2006 for more information).
As described in the PDB2PQR user guide and on the PDB2PQR server page, ligand parameterization currently requires a MOL2-format representation of the ligand to provide the necessary bonding information. MOL2-format files can be obtained through the free PRODRG web server or some molecular modeling software packages. Please note that PRODRG provides documentation as well as several examples on ligand preparation on its web page; please refer to the PRODRG documentation for questions about ligand MOL2 file preparation.Mixing things up a little bit from above, we're now ready to look at the 1HPV crystal structure (HIV-1 protease) and parameterize its ligand, KNI-272. We're going to
1HPX into the PDB ID field. Once the calculations are complete, you should see a web page with a link to the new PQR file with a warning about debumping P81 (but no warnings about ligand parameterization!). You can download the resulting PQR file and view it in your favorite molecular visualization package (e.g., VMD, PyMOL, or PMV). For comparison, you might download the the original PDB file and compare the PDB2PQR-generated structure with the original to see where hydrogens were placed and how the ligand is bound to the active site.
Our next example uses PDB structure 1ABF of L-arabinose binding protein in complex with a sugar ligand at 1.90 Å resolution. To parameterize both this protein and its ligand:
1ABF into the PDB ID field. Once the calculations are complete, you should see a web page with a link to the new PQR file with a warning about debumping P66, K295, and K306 (but no warnings about ligand parameterization!). You can download the resulting PQR file and view it in your favorite molecular visualization package (e.g., VMD, PyMOL, or PMV). For comparison, you might download the the original PDB file and compare the PDB2PQR-generated structure with the original to see where hydrogens were placed and how the ligand is bound to the active site.