The first research project is dealing with endocannabinoid system. The aim is to utilize molecular modeling methods, including 3D-QSAR, comparative protein modeling and database screening in designing new lead compounds active in the cannabinoid system, and especially for the CB1-receptor and FAAH-enzyme. The second research topic is in the field of virtual high-throughput screening. This so-called "In-Silico" project, partially funded by Tekes, is trying to develop a computerized method that uses molecular fields to measure similarity between molecules. Within the project new algorithms are developed which allow fast and accurate field alignment and comparison.