Computational Modeling Overview and Evolution

Nowadays the drug discovery process is a complex workflow full of risks that even the majority of the big pharma company cannot deal with. In this view a big opportunity is given by the computational approaches. These are among the cheapest ways for a pharmaceutical company, biotech or even a university laboratory to work in the drug discovery field. In the past, the complexity of the rules to respect to reach the market with a new drug was lower, so the risk of fail was reduced, this is the reason why today a renewed interest engaged the molecular modeling field. In the 1979, the Tripos® International company was the first company to bring scientific computational drug discovery capabilities to the academic and industrial pharmaceutical world through its proprietary software called Sybyl®. It was able to help the drug design process suggesting the right moiety placement/replacement on a set of molecules with known activity against a biological target. It was able to run on really expensive computer, with dedicated architecture. Later, less but still expensive computers like the Silicon Graphics® workstations, running the IRIX® operating system where able to perform such a calculation. The big jump ahead was done by the distribution of the first two releases of the LINUX operating system in 1993, namely Slackware® and Debian®. This operating system was similar to the IRIX proprietary OS in terms of capabilities, but was free to download and install on normal computers, so it leads to the possibility to run modeling software on standard hardware. Interestingly the software complexity and features evolved during the years and now some of them can be run directly on Graphic Processing Unit, like the NVidia® GPUs, with outstanding performances. Moreover, the molecular modeling scenario was enriched by the release of open source and free modeling software.