Artificial intelligence methods and classical statistics are being applied, in combination, to the task of analyzing and optimizing the cancer drug development pipeline. The Anti-Cancer Agent Predictions Working Group (ACAP) has been formed to coordinate these efforts. initial results are as follows: (1) Development of neural networks (MecNet) capable of predicting mechanism of drug action on the basis of patterns of activity in the 60-cell line cancer drug screen; (2) Development of a program (DISCOVER) for identification of new structural and functional motifs among the tens of thousands of agents tested in the screen; (3) Use of clustering in combination with neural networks and discriminant analysis to identify candidate cell lines for replacement in the screen (by breast, prostate, target-selected, target-transfected, and non-malignant lines, as appropriate). (4) Preliminary predictions of the clinical activity of phase II-evaluable drugs on the basis of patterns of activity in the screen.

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Intramural Research (Z01)
Project #
1Z01CB008398-01
Application #
3796478
Study Section
Project Start
Project End
Budget Start
Budget End
Support Year
1
Fiscal Year
1992
Total Cost
Indirect Cost
Name
Division of Cancer Biology and Diagnosis
Department
Type
DUNS #
City
State
Country
United States
Zip Code