Feature tree fragment space

Fig. 4.13 (a) Workflow of a Feature Tree fragment space set-up. The underlying fragment database can be derived from retrosynthetic analysis (RECAP) of compound databases (WDI) and/or from a selection of reagents from vendor databases (ACD). In addition, chemical rules for allowed fragment combi-... 

The Feature Tree fragment spaces screening concept was exemplified for a number of known actives across different target classes in the paper introducing the general concept [9], Starting from a single active compound as the query structure, the ability of the search method to construct and identify so-called... 

Fig. 4.15 Hit list from Feature Tree fragment space searches are grouped by a molecular framework analysis (y-axis) and the Feature Tree similarity score (x-axis). The search algorithm indeed produces compounds from...

The so-called feature tree fragment space (FTree-FS) method uses dynamic programming techniques and allows the detection of the molecule with the highest similarity to the query that can be found in the fragment space. The efficiency of searches in fragment spaces versus searches in product spaces can be easily shown. 

Lessel, U., Wellenzohn, B., Lilienthal, M., Claussen, H. (2009) Searching fragment spaces with feature trees. J Chem Inf Model 49, 270-279. 

Searching Combinatorial Fragment Spaces with Feature Trees I 99... 

In order to solve this task, we will again develop a dynamic programming scheme as in the case of pairwise Feature Tree comparisons. For every directed cut in the query tree and every link type in the fragment space, we would like to know the most similar fragment of the space under the assumption that the link of the fragment is matched to the atom behind the directed cut (see Fig. 4.12). For all pairs of directed cuts and link types, a list of the most similar... 

A popular way of creating fragment spaces is the retrosynthetic analysis of a compound set described in the context of Feature Trees in [9]. Here, the rules are used to break compounds into fragments at chemically meaningful positions and add the corresponding linker atoms. Fragmentation can be performed within relevant databases of drug-like chemical compounds such as the WDI,... 

The second problem appears when we browse through hit lists resulting from similarity searches in large spaces. Owing to the size of the space, the compounds in the hit list might be very similar to each other. It is therefore necessary to control the diversity within the hit list. Within Feature Trees, we have several criteria at hand for example, the number of common or different fragments between two molecules in the hit list can be limited. Alternatively, Feature Tree similarity within the hit fist can be used to increase diversity. 

---