Active learning for computational chemogenomics
This free research article brought to MedChemNet users by our sister journal Future Medicinal Chemistry, discusses the development of an active learning computational chemogenomic model that may assist in improving the rate of drug discovery.
In this free research article, the collaborative research efforts of Daniel Reker, Petra Schneider, Gisbert Schneider (all Swiss Federal Institute of Technology, Zurich, Switzerland) and J.B. Brown (Kyoto University Graduate School of Medicine, Kyoto, Japan) explore the development of an active learning computational chemogenomic model for drug discovery.
The research team's models display high predictive
performance on datasets many folds larger than those utilized for model
construction, leading to the implication that chemogenomic active
learning might actually be able to computationally identify the most
beneficial assays for subsequent execution and evaluation. These results indicate that it could
serve as a platform to iteratively include the experimental results in an actively
updating model, which consequently would lead to making strides in
improving discovery rates and reducing screening costs.
Read the full article here:
Reker D, Schneider P, Schneider G, Brown JB. Active learning for computational chemogenomics. Future Med. Chem. doi:10.4155/fmc-2016-0197 (Epub ahead of print) (2017)