Contributed data, collaboration and experience with ionization models – Part 1. Bayer Pharmaceuticals

Authors: ter Laak A, Gunther J
Software: ADMET Predictor®

World‘s best in class pKa prediction tool

  • In 2012 we joined forces with the company Simulations Plus with the aim to jointly develop the world‘s best in
    class pKa prediction tool
  • In 2013 the new pKa prediction tool was rolled out at Bayer and became also available world-wide via its
    integration into the ADMET Predictor software sold and distributed by Simulations Plus
  • The tool showed superb predictivity in our internal validation with ~13K new compounds (MUE=0.50, R2=0.93)*
  • In May 2018 the pKa tool won the SAMPL6 pKa prediction challenge as best of 32 participants#
  • pKa prediction can help to select and prioritize the right compounds for synthesis
  • This is even more so the case for ionizable groups for which pKa transitions cannot be detected experimentally

By Antonius ter Laak & Judith Günther