Jain, A.N., Dietterich, T.G., Lathrop, R.H., Chapman, D., Critchlow, R.E., Bauer, B.E., Webster, T.A., Lozano-Perez, T. (1994). Compass: A shapebased machine learning tool for drug design. Computer Aided Molecular Design, 8 635-652.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....box that covers at least one example from each positive bag and no examples from any negative bag. Then they expand the resulting box (via a statistical technique) When appropriately tuned, their algorithm gives 89 accuracy on Musk2. The work of Dietterich et al. was preceded by the work of Jain et al. 1994) in which they presented COMPASS which as an APR like neural network algorithm which is robust to errors in the initial alignment of the molecules. While COMPASS can handle real valued labels, we are not aware of any reported results on any available real valued data sets. Auer (1997) presented ....

Jain, A.N., Dietterich, T.G., Lathrop, R.H., Chapman, D., Critchlow, R.E., Bauer, B.E., Webster, T.A., Lozano-Perez, T. (1994). Compass: A shapebased machine learning tool for drug design. Computer Aided Molecular Design, 8 635-652.

....do not provide the kind of structural information that makes it easy to design new molecules. These weaknesses have led researchers in academic institutions and the pharmaceutical industry to seek alternative approaches that allow the explicit representation of structure. The compass algorithm [19, 20] overcomes these problems by using a more sophisticated representation of molecular shape, neural network learning methods and adaptive alignment of conformations. The models produced can be used to predict the activity of new molecules, and visual representations of the results can aid compound ....

A.N. Jain, T.G. Dietterich, R.H. Lathrop, D. Chapman, R.E. Critchlow, B.E. Bauer, T.A. Webster, and T. Lozano-Perez. Compass: a shape-based machine learning tool for drug design. Journal of Computer-Aided Molecular Design, 8:635-652, 1994.

....be used by a synthetic chemist involved in devising new drugs. Such relational knowledge descriptions are known within the drug design literature as pharmacophores. Successful applications of a machine learning technique in problems related to pharmacophore discovery have been discussed previously [12, 13]. The domain expert (first author) in the present study suggested a more explicit representation for pharmacophores (Section 2.2) This paper describes a series of experiments providing insights at four different levels. First, regarding Angiotensin Converting Enzyme (ACE) inhibition the ....

....alignment, must be chosen in advance of the analysis. This is equivalent to deciding the manner in which the molecules interact with the target. If the molecules contain a large common structural element this alignment may be straightforward, but this is often not the case. The COMPASS algorithm [12, 13] overcomes these problems by using a more sophisticated representation of molecular shape, neural network learning methods and adaptation of the alignments. The models produced can be used to predict the activity of new molecules, and, again, visual representations of the results can aid compound ....

A.N. Jain, T.G. Dietterich, R.H. Lathrop, D. Chapman, R.E. Critchlow, B.E. Bauer, T.A. Webster, and T. Lozano-P'erez. Compass: a shape-based machine learning tool for drug design. Journal of Computer-Aided Molecular Design, 8:635--652, 1994.

....chemist involved in devising new drugs. Such relational knowledge descriptions are known within the drug design literature as pharmacophores (see Section 2. 2) Successful applications of a machine learning technique in problems related to pharamcophore discovery have been discussed previously [5, 6]. The domain expert (first author) in the present study suggested a more explicit representation for pharmacophores (Section 2.2) This paper attempts to provide detailed documentation of experiments carried out jointly between the Computational Chemistry Group at Pfizer and the Oxford University ....

....must be chosen in advance of the analysis. This is equivalent to deciding on the manner in which the molecules interact with the target. If the molecules contain a large common structural element this alignment may be straightforward, but this is often not the case. The COMPASS algorithm [5, 6] overcomes these problems by using a more sophisticated representation of molecular shape, neural network learning methods and adaption of the alignments. The models produced can be used to predict the activity of new molecules, and, again, visual representations of the results can aid compound ....

A.N. Jain, T.G. Dietterich, R.H. Lathrop, D. Chapman, R.E. Critchlow, B.E. Bauer, T.A. Webster, and T. Lozano-P'erez. Compass: a shape-based machine learning tool for drug design. Journal of Computer-Aided Molecular Design, 8:635--652, 1994.

....do not give the peak performance. Peak performance of 91.2 is attained for any of the following parameter values: 0:99, ffl = 0:012) 0:99, ffl = 0:014) and ( 0:995, ffl = 0:008) This matches the best performance reported for an APR like neural network algorithm on this same data set (Jain, Dietterich, Lathrop, et al. 1994), where parameter values were also chosen after cross validation. Performance of at least 89.2 is robust over a wide range of parameter values. Figure 22 shows a visualization of the binding hypothesis learned from the entire Musk Data Set 2 applied to classify conformation 18 of a molecule ....

....effect of some drugs can only be measured by qualitative response, there are usually quantitative measures of drug efficacy in human subjects and in laboratory assays. Hence, medicinal chemists are primarily interested in algorithms 36 for predicting real valued activites. As we mentioned above, Jain, Dietterich, Lathrop, et al. 1994), and Jain, Koile, Bauer, et al. 1994) describe an APR like neural network based method, called COMPASS, that can make quantitative activity predictions. Second, the algorithms in this paper assume that a conjunction of conditions must be satisfied for binding. This is not always the case. For ....

[Article contains additional citation context not shown here]

Jain, A. N., Dietterich, T. G., Lathrop, R. H., Chapman, D., Critchlow, R. E., Bauer, B. E., Webster, T. A., Lozano-Perez, T. (1994). Compass: A shape-based machine learning tool for drug design. Journal of Computer Aided Molecular Design, 8 (6), 635--652.

No context found.

Jain, A., Dietterich, T., Lathrop, R., Chapman, D., Critchlow, R., Bauer, B., Webster, T., and Lozano-P'erez, T. (1994a). Compass: a shape-based machine learning tool for drug design. Journal of Computer-Aided Molecular Design, 8:635--652.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC