Reaction similarity searching

The next step is to design a set of reactions to synthesize the compounds. One or more reaction databases can be searched to find whether any reactions give the desired structures as products or give structures that are similar to the desired ones. The chemist may also use reaction similarity searching (73)and searching across reaction schemes (e.g., if A + B c + D and C + E F + G a reaction scheme search wiU find the query A — F) (74). Once a reaction is found, the chemist needs to decide what reagents to use in the synthesis and where to obtain them. The selection of reagents will usually be based on a combination of physicochemical property considerations (i.e., QSAR and diversity), tempered by... 

Reaction similarity searching differs from conventional similarity searching in that it involves calculating the similarities between pairs of reactions. 

We have also defined and implemented a reaction similarity search by making use of the presence or absence of the 933 molecule features in the participants of the reactions to obtain a molecule component of the similarity value, and an additional set of of 230 features associated with the reacting centres and their immediate environment to obtain a reaction-centre component of the similarity value. 

We demonstrate the usefulness of similarity searching over both the reaction and the molecule domain as an important tool in the design and development of new chemical compounds. Reaction similarity searching is instrumental in designing viable synthetic routes to a proposed target molecule. Molecule similarity searching and its sub- and super-similarity variations are effectively used for derivatisation and lead optimisation studies. 

Only then can the full arsenal of processing reaction information, such as reaction center searching, reaction similarity perception, or reaction classification (see Section 3.5) be invoked. Figure 10.3-19 shows such a full-fledged reaction represen tation. 

As stated before, PGVL is too large to be fully enumerated practically. Therefore our strategy is to find a way to focus in a just-in-time manner on much smaller sub-regions ( 104) of PGVL for subsequent on-the-fly enumeration followed by standard similarity search against the same query molecule. It is intuitively evident that a virtual compound space built from parallel synthesis reaction protocols has inherent array structures in the form of implicit arrays of related just-in-time enumerated compounds, even if those compounds do not have their molecular structures yet enumerated at the time this inherent array structure is exploited. 

Identify suitable reactants most similar to the corresponding virtual reactants obtained from step 1 in order to focus on the most relevant sub-regions. But the disconnection does not necessarily result in bona fide known and available starting materials, after just step 1. Consider as an example a two-component reaction which in the PGVL has M suitable bona fide reactants for the first reaction component and N suitable bona fide reactants for the second reaction component. Two similarity searches are used in the step to select m (out of M) and n (out of N) reactants based on two virtual reactants as seeds, which arose from the exact disconnection of the query molecule. In most cases, M and N are 103, and m and n are 102. Here extra search parameters need to be specified and/or optimized for each reaction component. 

The output is a set of Basis Products with high asymmetric similarity (AS) values (the default cutoff value is set to 90%) when they are mapped into the query molecule. The reaction schemes and reactants encoded by those Basis Products are then extracted, ranked, and used to form sub-regions of PGVL for subsequent just-in-time enumeration and symmetric similarity search against the query molecule. 

The workflow with WODCA starts with entering a target structure, the reaction product. The software automatically performs an identity search in the database to identify suitable starting materials. If no starting materials are found, the user can start a similarity search in the database. Similarity searches include 40 different criteria, such as the following ... 

Searching for structures or reactions in an external repository is an alternative to entering the structure directly into the scientific document. This is especially helpful if the structure is complex and not easy to author. Another application is the search for structures stored in an external system — for instance, to find already performed identical or similar reactions. The search query is typically entered via commonly used structure editors that provide the required standard file format. The search query can be performed on individual databases or on all connected databases, including the internal one, at the same time. 

The toxicity of phosgene has spawned a lot of research into alternates for both MDI and TDI, as well as polycarbonates. In addition to safety, there are economic incentives for developing alternate routes. In the conventional MDI process, methylene diphenylmethane diamine (MDA) is formed by reacting aniline with formaldehyde. Separating excess aniline from crude MDA is an expensive operation. Also, by-product HCl formed in the conversion of MDA to MDI is an environmental issue. The final isocyanate product contains hydrolyzable chloride compounds that are difficult to separate and dispose of. The reactants must be kept bone dry to prevent corrosion, and the introduction of water can cause a runaway reaction. Similar concerns influence the search for nonphosgene routes for TDl. Conventional routes to polycarbonates also employ phosgene, which produces chlorine waste products, primarily sodium chloride, that present disposal problems. The elimination of chlorine from the polycarbonate process would constitute a major improvement. 

Rosowsky and co-workers sought to synthesize thiophene-containing derivatives of trimetrexate and piritrexim in their search for potential inhibitors of Pneumocystis carinii and Toxoplasma gondii dihydrofolate reductase. Their goal was to improve selectivity and limit the potential side effects found in trimetrexate and piritrexim. The thiophene moieties were installed by means of a Gewald aminothiophene reaction similar to those used in the creation of the C-nucleosides shown previously. Desired substitution patterns were accessed by the use of appropriate substrates as shown below. 

---