Our platform technology enables us to detect relationships between complex molecular patterns and specific biological activity of molecules.
It also helps us to develop models which can prioritise and rank molecules depending on their biological activity to specific drug targets. Only the best models with the highest prediction ability are included into our expert systems.
Science & technology
Our proprietary platform technology contains tens of millions of substances together with experimental data, tens of thousands of 3D-models and a huge amount of related scientific data. PharmaInformatic´s ADME/Tox-Database contains an impressive and continuously increasing amount of pharmacokinetic data obtained from clinical trials in humans and preclinical trials in animals.
A lot of work has been done in writing novel descriptors, which can describe drug patterns effectively and accurately. We have developed novel methods and algorithm in the field of chemInformatics, bioInformatics and data mining, which helped us to integrate, analyse and “understand” all the data. We have also used our long-term experience in artificial intelligence to build up data driven models.
See machine learning.
Other predictive models can be built up by using our rule based docking approach, see structure based drug design. We have access to multiple interaction site models for a wide variety of biologically relevant targets. Integrating different models results in an impressive prediction ability.
PharmaInformatic and UNIZYME Laboratories A/S sign cooperation agreement (more).