• Home
  • Publications
  • Presentations
  • Research
  • Funding
  • Resources
  • People
    • Dr. David L. Mobley
    • PostDoc
      • Lea El Khoury
      • Sukanya Sasmal
    • Graduate
      • Sam Gill
      • Victoria Lim
      • Kalistyn Burley
      • David Wych
      • Danielle (Teresa) Bergazin
      • Jessica Maat
      • Hannah Baumann
      • Oanh Tran
    • Undergraduate
      • Meghan Osato
      • Jordan Ehrman

Mobley Lab, UCI

Free energy methods for pharmaceutical drug discovery

You are here: Home / Papers / SAMPL6 challenge results from pKa predictions based on a general Gaussian process model

Papers, Our science

SAMPL6 challenge results from pKa predictions based on a general Gaussian process model

Caitlin C. Bannan, David L. Mobley, A. Geoffrey Skillman

A variety of fields would benefit from accurate pKa predictions, especially drug design due to the affect a change in ionization state can have on a molecules physiochemical properties. Participants in the recent SAMPL6 blind challenge were asked to submit predictions for microscopic and macroscopic pKas of 24 drug like small molecules. We recently built a general model for predicting pKas using a Gaussian process regression trained using physical and chemical features of each ionizable group. Our pipeline takes a molecular graph and uses the OpenEye Toolkits to calculate features describing the removal of a proton. These features are fed into a Scikit-learn Gaussian process to predict microscopic pKas which are then used to analytically determine macroscopic pKas. Our Gaussian process is trained on a set of 2,700 macroscopic pKas from monoprotic and select diprotic molecules. Here, we share our results for microscopic and macroscopic predictions in the SAMPL6 challenge. Overall, we ranked in the middle of the pack compared to other participants, but our fairly good agreement with experiment is still promising considering the challenge molecules are chemically diverse and often polyprotic while our training set is predominately monoprotic. Of particular importance to us when building this model was to include an uncertainty estimate based on the chemistry of the molecule that would reflect the likely accuracy of our prediction. Our model reports large uncertainties for the molecules that appear to have chemistry outside our domain of applicability, along with good agreement in quantile-quantile plots, indicating it can predict its own accuracy. The challenge highlighted a variety of means to improve our model, including adding more polyprotic molecules to our training set and more carefully considering what functional groups we do or do not identify as ionizable.
A preprint on chemRxiv, 2018. DOI

Share this:

  • Click to share on Twitter (Opens in new window)
  • Click to share on Facebook (Opens in new window)
  • Click to share on LinkedIn (Opens in new window)
  • Click to share on Reddit (Opens in new window)

Related

Posted: June 6, 2018 · Tags: machine learning, pKa

Categories

  • Papers (12)
  • News (29)
  • Commentary (1)
  • Meetings and workshops (1)
  • Null results (1)
  • Our science (12)
  • Positions (4)
  • Talks and posters (4)

People

  • Dr. David L. Mobley
  • PostDoc
    • Lea El Khoury
    • Sukanya Sasmal
  • Graduate
    • Sam Gill
    • Victoria Lim
    • Kalistyn Burley
    • David Wych
    • Jessica Maat
    • Danielle (Teresa) Bergazin
    • Hannah Baumann
    • Oanh Tran
  • Undergraduate
    • Meghan Osato
    • Jordan Ehrman

Recent Papers

Escaping Atom Types in Force Fields Using Direct Chemical Perception

David L. Mobley , Caitlin C. Bannan , Andrea … [Read More...]

Challenges in the use of atomistic simulations to predict solubilities of drug-like molecules

Guilherme Duarte Ramos Matos, David L. … [Read More...]

Open Force Field Consortium: Escaping atom types using direct chemical perception with SMIRNOFF v0.1

David Mobley, Caitlin C. Bannan, Andrea Rizzi, … [Read More...]

Binding modes of ligands using enhanced sampling (BLUES)

Samuel C. Gill, Nathan M. Lim, Patrick B. … [Read More...]

Synthesis facilitates an understanding of the structural basis for translation inhibition by the lissoclimides

Z. A. Könst, A. R. Szklarski, S. Pellegrino, S. E. … [Read More...]

RSS What we’re reading:

  • A reoptimization of the five-site water potential (TIP5P) for use with Ewald sums November 25, 2020
  • Characterization of the TIP4P-Ew water model: Vapor pressure and boiling point November 25, 2020
  • Development of an improved four-site water model for biomolecular simulations: TIP4P-Ew November 25, 2020
  • A computational investigation of thermodynamics, structure, dynamics and solvation behavior in modif November 25, 2020
  • The missing term in effective pair potentials November 25, 2020

Archives

Tags

Graduate Grants Benchmarks PostDoc Papers Open Force Field Undergraduate Alchemistry SAMPL Binding

Contact

David L. Mobley
dmobley@mobleylab.org
phone: 949.385.2436
office: 3134B Nat. Sci. I

Meta

  • Log in
  • Entries feed
  • Comments feed
  • WordPress.org
  • Home
  • Publications
  • Research
  • Funding
  • Resources
  • People
  • Presentations
  • Postdoc position with OpenFF
  • Oanh Tran

Copyright © 2021 · Education Pro on Genesis Framework · WordPress · Log in