Showing papers by "Henrik Boström published in 2003"

Improving structure-based virtual screening by multivariate analysis of scoring data.

Abstract: Three different multivariate statistical methods, PLS discriminant analysis, rule-based methods, and Bayesian classification, have been applied to multidimensional scoring data from four different target proteins: estrogen receptor α (ERα), matrix metalloprotease 3 (MMP3), factor Xa (fXa), and acetylcholine esterase (AChE) The purpose was to build classifiers able to discriminate between active and inactive compounds, given a structure-based virtual screen Seven different scoring functions were used to generate the scoring matrices The classifiers were compared to classical consensus scoring and single scoring functions The classifiers show a superior performance, with rule-based methods being most effective The precision of correctly predicting an active compound is about 90% for three of the targets and about 25% for acetylcholine esterase On the basis of these results, a new two-stage approach is suggested for structure-based virtual screening where limited activity information is available

Method for efficiently checking coverage of rules derived from a logical theory

Abstract: The method is used in a computer and includes the steps of providing a logical theory ( 12, 30 ) that has clauses. A rule ( 14 ) is generated that is a resolvent of clauses in the logical theory. An example ( 16 ) is retrieved. A proof tree ( 18, 40 ) is generated from the example ( 16 ) using the logical theory ( 12, 30 ). The proof tree ( 18, 40 ) is transformed into a database ( 20, 42 ) of a coverage check apparatus ( 28 ). The rule ( 14 ) is converted into a partial proof tree ( 60 ) that has nodes ( 62, 54, 66 ). The partial proof tree is transformed into a database query ( 22 ) of the coverage check apparatus ( 28 ). The query ( 22, 72 ) is executed to identify tuples in the database ( 20, 42 ) that correspond to the nodes of the partial proof tree.