Calculate Drug Molecule Similarity Derived by Maximum Common Substructure Search

Usage

calcDrugMCSSim(
  mol1,
  mol2,
  type = c("smile", "sdf"),
  plot = FALSE,
  al = 0,
  au = 0,
  bl = 0,
  bu = 0,
  matching.mode = "static",
  ...
)

Arguments

mol1: The first molecule. R character string object containing the molecule. See examples.
mol2: The second molecule. R character string object containing the molecule. See examples.
type: The input molecule format, 'smile' or 'sdf'.
plot: Logical. Should we plot the two molecules and their maximum common substructure?
al: Lower bound for the number of atom mismatches. Default is 0.
au: Upper bound for the number of atom mismatches. Default is 0.
bl: Lower bound for the number of bond mismatches. Default is 0.
bu: Upper bound for the number of bond mismatches. Default is 0.
matching.mode: Three modes for bond matching are supported: 'static', 'aromatic', and 'ring'.
...: Other graphical parameters

Value

A list containing the detail MCS information and similarity values. The numeric similarity value includes Tanimoto coefficient and overlap coefficient.

Details

This function calculate drug molecule similarity derived by maximum common substructure search. The maximum common substructure search algorithm is provided by the fmcsR package.

References

Wang, Y., Backman, T. W., Horan, K., & Girke, T. (2013). fmcsR: mismatch tolerant maximum common substructure searching in R. Bioinformatics, 29(21), 2792--2794.

Examples

mol1 = 'CC(C)CCCCCC(=O)NCC1=CC(=C(C=C1)O)OC'
mol2 = 'O=C(NCc1cc(OC)c(O)cc1)CCCC/C=C/C(C)C'
mol3 = readChar(system.file('compseq/DB00859.sdf', package = 'Rcpi'), nchars = 1e+6)
mol4 = readChar(system.file('compseq/DB00860.sdf', package = 'Rcpi'), nchars = 1e+6)
if (FALSE) {
sim1 = calcDrugMCSSim(mol1, mol2, type = 'smile')
sim2 = calcDrugMCSSim(mol3, mol4, type = 'sdf', plot = TRUE)
print(sim1[[2]])  # Tanimoto Coefficient
print(sim2[[3]])  # Overlap Coefficient
}