Notice
2019.06.28 Team application deadline has been changed to 2019.08.21 (Wed).
2019.09.26 Added SMS Verification to winning conditions (Please see the "Evaluation" page).
Background
The development of new drugs requires such a long process from basic research to non-clinical studies, clinical studies, approval process and so forth, and also in general, it is said to take over a period of more than 10 years at a cost of several tens of billions of yen to one hundred billion yen. In addition, as it is necessary to optimize many factors such as drug efficacy, safety, and pharmacokinetics, it is said that the success rate of a lead compound to be launched on the market as a new drug is as low as 1 in a 30,000 possibility, even with the huge cost over a long period of time.
In recent years, methodologies from mathematical simulation and machine learning are attracting attentions from pharmaceutical industry and are expected to achieve development efficiency and cost reduction, for instance, by identifying novel candidate compounds from a large number of compounds rapidly. Also, applying these methodologies to big data analysis in the process of non-clinical studies to select candidate compounds efficiently is also expected to make the drugs reach patients in need of them as early as possible.
In this competition, participants are challenged to create an algorithm to predict pharmacokinetic parameters using multiple descriptors that represent structural chemical, physicochemical and biochemical characteristics of various compounds with the aim of improving the efficiency of new drug discovery process. This is definitely a rare opportunity to deal with the data in the pharmaceutical industry and to contribute to drug discovery.
We are looking forward to your participations.
* This competition is proposed and implemented by SIGNATE as a research project of Takeda Pharmaceutical Company's open innovation program (COCKPI-T® Funding).
Task
Predict pharmacokinetic parameters from compound information and experimental data in the screening test
* Please refer to "Data" page for details such as data format.