The workshop scheduled for May 20—22, 2013 has been cancelled. Please visit this site again for future offerings.
University of Kansas School of Medicine • Kansas City, Kansas
- Experience a hands-on workshop
- Examine fuzzy logic-based schemes used in anti-cancer drug development
- Work with matlab software to train a neural network
- Review examples from bioinformatics research
The cost of drug discovery and development for each medication successfully brought to market is often hundreds of millions of dollars. The use of computational methods and bioinformatic tools can reduce this cost burden by concentrating on compounds with the greatest chance for success. In addition to lessening expenses, this approach can help biopharmaceutical companies bring drugs to market more quickly.
Contact Hours: 21
Participants in this seminar will receive a certificate of participation for 21 contact hours of instruction. It may be used to meet the CEU requirements of many professional associations.
This three-day course introduces the use of neural networks and fuzzy logic as aids not only in computer-assisted drug design projects, but also in broad bioinformatics research. Attendees will work with MATLAB software to train a neural network to achieve optimum performance. This hands-on experience will include working with the architectural design of the network and its optimization via iterative training and validation processes. The course will thoroughly examine applications of several neural network and fuzzy logic-based schemes used in anti-cancer drug development, together with examples from bioinformatics research. The advantages and disadvantages of different types of supervised and unsupervised neural networks will also be covered.
The course is intended for educators and scientists with a background equivalent to graduate studies in any branch of science or engineering. It is specially designed for those interested in learning non-analytical scientific approaches to solving complex issues of bioinformatics, including drug discovery. This course will also be of interest to experienced scientists who want to explore the utilization of neuro-fuzzy models to efficiently classify the potential of a drug while input data is incomplete and/or corrupted with noise.