Skip directly to content

T-03: Personalized pharmacotherapy in psychiatry

Development of a chemoinformatic-based, data- and model-driven software module for individualized, optimized pharmacotherapy

The aims of this project are to minimize the unwanted effects of drugs or of interaction between drugs and to thus improve patient care while, at the same time, reduce costs in health services.

Currently available drug information systems and their electronic implementation still have considerable deficits. In today's systems, not enough consideration is given to the patient's individual circumstances (sex, tolerance or specific genetic features, especially in connection with polymedication). Often there is also a lack of data regarding the inhibition of active transport and the sub-cellular localization of drugs.

This project focuses on pharmacotherapy of patients suffering from psychiatric illnesses. The reason is that, in this field, termination of treatment due to unwanted effects represents a particular problem both in clinical terms and in terms of costs. Psychiatric illnesses are among the most common and serious illnesses that health services have to deal with.

In this project, a chemoinformatic approach will be taken to draw up a data and model-driven software module for individualized, optimized pharmacotherapy for both in-patients and out-patients. The approach is based on the integration of three pillars in a way that is transparent for the user:

1. On the existing knowledge of drugs, available in databases, regarding their effects and interaction with other drugs. This will be subjected to editorial scrutiny and, if appropriate, experimental data will be added.

2. On clinical patient data for individualized treatment.

3. On structure-based, chemoinformatic models created using the machine learning process. For the first time, such models will allow integration of chemo-pharmacological and clinical properties of substances.

Project Duration

01.07.2010 - 30.04.2015

Project Partners

  • Molecular Networks GmbH
  • Friedrich-Alexander-University Erlangen-Nuremberg (FAU)
  • University Hospital Erlangen


Lothar Terfloth
Molecular Networks GmbH

Henkestraße 91
91052 Erlangen

Kommentieren Sie diesen Beitrag und diskutieren Sie mit anderen Akteuren in der Medical Valley-Community auf Xing.
Wir freuen uns auf Sie!