Real world, chemical space

A domain object model is a model that describes key domain concepts and their relationships. Many of these concepts come from tangible objects in the real world of the problem domain. In the chemical informatics space, these are the objects that chemists are dealing with on a daily basis, such as compounds, structures, notebooks, and libraries. The domain analysis model being presented here focuses on those objects that are involved in the compound registration process. 

What became apparent was that the failures of the two theories and the ways they were amended had something in common. The simple theories did not initially incorporate all the dimensions that were part of our real world. Especially, noteworthy was the fact that neither had incorporated the evolutionary perspective, time s arrow, by which we who live in the 21st century make sense of large areas of the sciences and their interconnections. Structural theory in fact ignored time completely. Also neither theory in its classic form took account of the fact that atoms, molecules, and chemical bonds required space to exist, they were not points. Those were two crucial omissions. 

In-silico screening is one way to expand the chemical space beyond what has been reported after some 200 years of chemistry. While Chemical Abstracts and Beilstein might contain millions of chemical structures, there is another real chemical world that has yet to be mined and that is lost chemistry, the chemistry that was done but never reported. Current estimates are that on the order of 80% of the chemistry performed over the past years has not been published. It is to be found in files and lab books around the world, on dusty shelves, or destroyed long ago and lost forever. How some of the lost chemistry can be retrieved and evaluated is not our subject here. The visible chemistry that is real, the actual compounds in vials, is the center of the current focus. 

It is acknowledged that literal simulation of the combinatorial chemistry/ HTS process in silica still has substantial limitations. Although this will allow searching of chemical space orders of magnitude faster than can real world combinatorial synthesis, it can still explore only a minute fraction of even druglike chemical space. 

This section is a review of the macroscopic notions we are familiar with from chemical kinetics. Our final purpose is to build rate constants from the bottom up and therefore to describe the rate of chemical reactions in systems that are not in thermal equilibrium. Thermally equihbrated reactants are more typical of the laboratory than of the real world and, even in the laboratory, it takes care and attention to insure that the reactants are indeed thermally equilibrated. Outside of the laboratory, whether in the internal combustion engine (which fires many thousands of times per minute), in the atmosphere, or in outer regions of space, this is not the case. 

A limitation of the scope of the method is that each equation is structure-class specific. In real-world examples that do not readily fall into a previously studied class, model selection is a problem. Two methods have been described. In one approach, all the unique atom-based fragments obtained from all structure classes studied are projected into n-dimensional space (six and seven dimensions) based on topological environment vectors. A vector is then calculated for each carbon atom in the query compound. That model equation which corresponds to the atom fragment in n-dimensional space which is closest to the query atom fragment (Euclidean distance) is selected for predicting its chemical shift. Alternatively, a neural network is trained to relate the chemical environment of the set of carbon... 

The chemist must learn to live in, and to feel at home in, the world of molecules. It is not enough that he knows the chemical constitution and chemical reactions of the materials around him. To be really effective and successful, he must also develop an intimacy with the molecular world. He must fit himself into the molecular scale of things. He must put that first drummed-in chemical fact that molecules are small in the very back of his mind and replace it by a consciousness that molecules are real, intricate, structural arrangements of atoms in space. 

Very extensive computations have been carried out on the dynamics of reactions of the type A+BC. They are either founded on classical or on quantum mechanics, are either to be considered exact or involving more or less drastic approximations and have been based either in the real three dimensional world or in somewhat artificial spaces of lower dimensionality. These computations are thus attempts to solve the three-body problem more or less accurately. Other Chapters in this book extensively review this subject [41]. The papers, presented at a meeting celebrating "Fifty Years of Chemical Dynamics", held in Berlin in 1982 and published as an issue of the Berichte der Bunsen Gesellschaft in 1982, should be consulted, also for providing a historical perspective [42],... 

---