Information Acquisition
The work of Information acquisition is highly dependent on the computer today. With the integration of modern sensors into chemical instrumentation, the volume of data that can be generated is enormous. In future, instrumentation will incorporate information that exists in chemical databases, employ modeling techniques, and analyze experimental data as they are generated. Such “smart instruments” will significantly improve the ability of the user to make intelligent decisions about the course of an experiment while the data are being collected and analyzed.
Currently, there are two complementary pathways for generating and collecting information in the chemical sciences: experimentation and computer simulation. The Traditional way of data gathering was from experiments which were done manually. The development of computers small enough to be purchased by individual laboratories, the phrase “computers in chemistry” arose to describe their use. Several decades ago this expression meant interfacing a computer to an experiment like a spectrometer or a chromatograph and collecting the data in real time for storage and later manipulation. While this is still being done with microprocessors built into the instruments themselves, a more encompassing label for the wide range of chemical activities involving computers is “computational chemistry.”
Computational chemistry seeks to predict quantitatively molecular and bimolecular structures, properties, and reactivity by computational methods alone. It uses modern chemical theory to predict the speed of unknown reactions and the synthetic sequences by which complex new molecules can be made most efficiently. Computational chemistry allows chemists to explore how things work at the atomic and molecular levels and to draw conclusions that are impossible to reach by experimentation alone. Thus, computational chemistry supplements experimentally derived data.
One aspect of computational chemistry is molecular modeling. Molecular modeling involves the investigation of three-dimensional molecular structures using classical and quantum mechanical methods assisted by computer graphics. Other molecular modeling techniques include quantitative structure-property relationships, which find applications in structure-based drug design, similarity searching, and molecular shape prediction. Molecular modeling techniques are utilized extensively in pharmaceutical research, especially to predict pharmacophores–the structural features of molecules required for particular biological activities. Molecular modeling is now used routinely to generate data concerning energetic, dynamics and other information at the molecular scale that is not amenable to experimentation.
Recent advances in combinatorial synthesis and high throughput screening technologies now allow for preparation and analysis of hundreds of thousands of or even millions of molecules in a very short period of time. Combinatorial chemistry techniques grew out of several disciplines, including organic, medicinal, and physical chemistry, engineering and robotics, computational chemistry, informatics, and screening technology. Robotics as used in combinatorial chemistry provides the drug industry a powerful tool with which to screen millions of potential compounds in a fraction of the time it would have taken to evaluate even a few dozen compounds a decade ago. Now widely employed in the pharmaceutical area, combinatorial chemistry has begun to find applications in materials science. Because so much information is being generated and collected from combinatorial technologies, there is a concomitant problem associated with storing and retrieving those data. That problem is now being addressed by those skilled in chemical informatics.