Gaussian Processes: A Method for Automatic QSAR Modeling of ADME Properties

By O. Obrezanova, Gábor Csányi, Joelle M. R. Gola, Matthew D. Segall. 2007

196 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Molecular Biology, Spectroscopy

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Ruslan Salakhutdinov
2. Matthew D. Segall
3. Bryan L. Roth
4. A. Hopkins
5. Navdeep Jaitly
6. George E. Dahl
7. Dougal Maclaurin
8. Abhinav Vishnu

In Molecular Biology

1. A. Hopkins
2. Bryan L. Roth
3. Ruslan Salakhutdinov
4. Matthew D. Segall
5. Navdeep Jaitly
6. George E. Dahl
7. Dougal Maclaurin
8. Abhinav Vishnu

In Spectroscopy

1. Matthew D. Segall
2. A. Hopkins

Citations in the corpus

• Multi-task Neural Networks for QSAR Predictions (2014, background): “Gaussian process regression models can be viewed as an infinite hidden layer limit of Bayesian neural networks, but can still be quite computationally expensive, often requiring time cubic in the number of training cases [Obrezanova et al., 2007].”
• Automated design of ligands to polypharmacological profiles (2012, methodology): “ADME properties and CNS penetration are calculated using previously publishing Gaussian Process models 57,58 as implemented in StarDrop (Optibrium Ltd).”
• Can we really do computer-aided drug design? (2011, background): “But how can we further improve the accuracy of predictive methods? The computational algorithms for statistical modelling have grown increasingly sophisticated and now include a wide range of powerful non-linear fitting techniques, such as Random Forests [13], Support Vector Machines [14], Artificia”