QSAPR building
Silicos' powerful Spectrophore™ technology, in combination with Silicos' mix of people and expertise from different fields, makes it a very flexible partner to assist you in the different aspects of data analysis and QSAPR (Quantitative Structure Activity and Property Relations) building. Commonly applied technologies hereby include, for example, multi-linear and multi-nonlinear regression analysis, support vector machines, and neural networks. Datasets can be provided by the customer, or if desired, can also be assembled by Silicos.
An example is a model to predict the blood-brain barrier penetration of a molecule. In the figures below the performance of a generalised linear model and Support Vector Machine (SVM) with RBF kernel trained using Spectrophores™ is shown.
![]() |
![]() |

