CIDRUX uses computational intelligence for its software products and services. What do we mean by this? Computational intelligence is the term for a set of key technologies made available by modern computer science over the past ten years. They include e.g., evolutionary algorithms, fuzzy logic, neural networks and other nonlinear search and modeling methods. The essence is that they can handle practical problems of extreme complexity regarding the amount of data as well as the analysis and understanding of such massive amounts of data.

That is something well known in pharmaceutical industry. Proprietary compound databases, either in the conventional 2D or the more demanding 3D format, have become the cornerstone in early drug discovery. Through the recent developments in combinatorial chemistry such proprietary databases have literally exploded. At present they comprise rather multi-million than, say, 100,000 data entries, which used to be an average number for a top-100 player in the drug industry.

Combined with other compound databases (from commercial vendors or public origin) they often are at the core of each existing and new drug discovery project. Important examples of the latter databases are the ACD (Available Chemicals Directory – some 300,000 compounds), the WDI (World Drug Index – over 50,000 marketed or clinically tested materials), the CSD (Cambridge Structural Database – over 200,000 crystal structures of small molecules), and the compound database of the NCI (National Cancer Institute – close to 250,000 submitted compounds, tested in cancer cell lines).

With the elucidation of the human and other genomes, e.g. of pathogenic bacteria, yet another class of high-density information databases is becoming available. This category already has and will have even more tremendous impact on the discovery of novel drug targets. Again, pharmaceutical industry is heavily investing in this area of functional genomics, often through strategic alliances with smaller biotech and bioinformatics companies. In this respect databases from Celera and from Incyte are among the best known, but there are many more specialist databases, also in the public domain (such as GPCRDB, the receptor database to which CIDRUX’s CEO, Ad IJzerman, has contributed significantly).

Adding content to and exploiting such databases also requires computational intelligence. For instance in the drug discovery process, the generation of virtual combinatorial libraries, the attempts to de novo drug design (with or without knowing the 3D structure of the target protein), the mining of DNA (e.g., SNPs), RNA and protein databases, are all possible only with, and powered by computational intelligence.

The Molecule Evoluator^TM and the Molecule Commander^TM are using state-of-the-art computational intelligence methods, in particular evolutionary algorithms and data mining.