Peter , Gedeck, Skolnik, Suzanne and Rodde, Stephane (2017) Developing collaborative QSAR models without sharing structures. Journal of Chemical Information and Modeling, 57 (8). pp. 1847-1858. ISSN 1549-95961549-960X

Abstract

It is widely understood that QSAR models greatly improve if more data are used. However, irrespective of model quality, once chemical structures diverge too far from the initial data set, the predictive performance of a model degrades quickly. To increase the applicability domain we need to increase the diversity of the training set. This can be achieved by combining data from diverse sources.
In this contribution, we will present a method for the collaborative development of linear regression models. The method differs from other past approaches, because data are only shared in an aggregated form. This prohibits access to individual data points and therefore avoids the disclosure of confidential structural information. The final models are equivalent to models that were built with combined datasets.

Item Type:	Article
Date Deposited:	10 Oct 2017 00:45
Last Modified:	10 Oct 2017 00:45
URI:	https://oak.novartis.com/id/eprint/33162