Get access

A fast empirical GAFF compatible partial atomic charge assignment scheme for modeling interactions of small molecules with biomolecular targets

Authors

  • Goutam Mukherjee,

    1. Department of Chemistry & Supercomputing Facility for Bioinformatics & Computational Biology, Indian Institute of Technology, Hauz Khas, New Delhi 110016, India
    Search for more papers by this author
  • Niladri Patra,

    1. Department of Chemistry & Supercomputing Facility for Bioinformatics & Computational Biology, Indian Institute of Technology, Hauz Khas, New Delhi 110016, India
    Search for more papers by this author
  • Poranjyoti Barua,

    1. Department of Chemistry & Supercomputing Facility for Bioinformatics & Computational Biology, Indian Institute of Technology, Hauz Khas, New Delhi 110016, India
    Search for more papers by this author
  • B. Jayaram

    Corresponding author
    1. Department of Chemistry & Supercomputing Facility for Bioinformatics & Computational Biology, Indian Institute of Technology, Hauz Khas, New Delhi 110016, India
    • Department of Chemistry & Supercomputing Facility for Bioinformatics & Computational Biology, Indian Institute of Technology, Hauz Khas, New Delhi 110016, India
    Search for more papers by this author

Abstract

We report here a new and fast approach [Transferable Partial Atomic Charge Model (TPACM4)—upto four bonds] for deriving the partial atomic charges of small molecules for use in protein/DNA-ligand docking and scoring. We have created a look-up table of 5302 atom types to cover the chemical space of C, H, O, N, S, P, F, Cl, and Br atoms in small molecules together with their quantum mechanical RESP fit charges. The atom types defined span diverse plausible chemical environments of each atom in a molecule. The partial charge on any atom in a given molecule is then assigned by a reference to the look-up table. We tested the sensitivity of the TPACM4 partial charges in estimates of hydrogen bond dimers energies, solvation free energies and protein-ligand binding free energies. An average error ±1.11 kcal/mol and a correlation coefficient of 0.90 is obtained in the calculated protein-ligand binding free energies vis-à-vis an RMS error of ±1.02 kcal/mol and a correlation coefficient of 0.92 obtained with RESP fit charges in comparison to experiment. Similar accuracies are realized in predictions of hydrogen bond energies and solvation free energies of small molecules. For a molecule containing 50–55 atoms, the method takes on the order of milliseconds on a single processor machine to assign partial atomic charges. The TPACM4 programme has been web-enabled and made freely accessible at http://www.scfbio-iitd.res.in/software/drugdesign/charge.jsp. © 2010 Wiley Periodicals, Inc. J Comput Chem, 2011.

Ancillary