Algorithms for the identifications of three-dimensional maximal common substructures
Journal of Chemical Information & Computer Sciences
The nature of statistical learning theory
The nature of statistical learning theory
Solving the multiple instance problem with axis-parallel rectangles
Artificial Intelligence
Pharmacophore Discovery Using the Inductive Logic Programming System PROGOL
Machine Learning - Special issue on applications of machine learning and the knowledge discovery process
ACM SIGKDD Explorations Newsletter
ICML '01 Proceedings of the Eighteenth International Conference on Machine Learning
Learning from ambiguity
Quantitative pharmacophore models with inductive logic programming
Machine Learning
kFOIL: learning simple relational kernels
AAAI'06 Proceedings of the 21st national conference on Artificial intelligence - Volume 1
nFOIL: integrating Naïve Bayes and FOIL
AAAI'05 Proceedings of the 20th national conference on Artificial intelligence - Volume 2
An integrated approach to learning bayesian networks of rules
ECML'05 Proceedings of the 16th European conference on Machine Learning
Top-Down Induction of Relational Model Trees in Multi-instance Learning
ILP '08 Proceedings of the 18th international conference on Inductive Logic Programming
A numerical refinement operator based on multi-instance learning
ILP'10 Proceedings of the 20th international conference on Inductive logic programming
Hi-index | 0.00 |
We present a new machine learning approach for 3D-QSAR, the task of predicting binding affinities of molecules to target proteins based on 3D structure. Our approach predicts binding affinity by using regression on substructures discovered by relational learning. We make two contributions to the state-of-the-art. First, we use multiple-instance (MI) regression, which represents a molecule as a set of 3D conformations, to model activity. Second, the relational learning component employs the "Score As You Use" (SAYU) method to select substructures for their ability to improve the regression model. This is the first application of SAYU to multiple-instance, real-valued prediction. We evaluate our approach on three tasks and demonstrate that (i) SAYU outperforms standard coverage measures when selecting features for regression, (ii) the MI representation improves accuracy over standard single feature-vector encodings and (iii) combining SAYU with MI regression is more accurate for 3D-QSAR than either approach by itself.