Rule-based filters

In addition to the two major categories of LEVS approaches, similarity searching and compound classification, database filtering techniques should also be added to the virtual screening spectrum. Rule-based filters that screen databases for compounds with desired (e.g., drug- or leadlike) or undesired (toxicity and instability) molecular properties are widely used, not to identify individual active compounds but rather to reduce the size of screening databases as much as possible for further studies using more sophisticated methods. 

To date, many of the reported ADME/Tox models have been rule based. For example, some research groups have used relatively simple filters like the rule of 5 [93] and others [94] to limit the types of molecules evaluated with in silico methods and to focus libraries for HTS. However, being designed as rapid computational alert tools aimed at a single property of interest, they cannot offer a comprehensive picture when it comes to understanding ADME properties. 

METEOR Rule-based Metabolite prediction software Predicts the metabolic fate of chemicals Displays results as a metabolic tree. User can filter results for likely metabolites. Links directly to MetaboLynx for analysis of mass spectrometry data www.lhasalimited.org... 

Permeability and liver microsome screens are not high throughput. To save time, researchers have sought simple methods for eliminating compounds that will be poor lead candidates. A common method involves calculated indices. The first and most widely recognized index-based filter was reported by Lipinski in 1997.19 This filter is called Lipinski s rules or the Rule of 5 (Table 10.1). 

While the nitrite must be used in exactly the amount theoretically required, an excess of acid is always used. For laboratory preparations, as a rule, the excess acid used is about 0.5 equivalent over the theoretically required 2 equivalents. The excess of acid must be increased to 1 to 3 equivalents in diazotizations of weakly basic amines whose salts are readily hydrolyzed (e.g., chloro- and nitroanilines). In some cases where even this larger excess of acid is insufficient to dissolve the base (e.g., nitrochloroaniline), the diazotization may be carried out in suspension provided that the base is present in a finely divided state. Thus, a solution of the base in concentrated sulfuric acid may be poured into ice water, the base filtered off and washed and then made up into a paste. Even this method fails to work with some bases such as dinitroaniline and m trodichloroaniline these compounds can be diazotized smoothly only in concentrated sulfiuric acid solution. 

There are, in addition to these simple functional group filters, a number of property-based filters that may be applied. These fdters take the form of calculated metrics, such as the Lipinski Rule of Five (LRoF Hydrogen-bond donors. Hydrogen-bond acceptors, Lipophilicity, Molecular weight). Solubility, total Polar Surface Area (tPS A), Blood-brain-barrier (BBB) Permeability, calculated metabolic filters (cADMET Absorption-Distribution-Metabolism-Excretion-Toxicity) and Bioavailability. 

Functional group filters are mainly utilized to remove unstable, reactive, toxic, or otherwise unsuitable compounds from compound libraries. The rapid elimination of swill (REOS) method introduced by Vertex was the first realization of this concept. REOS effectively combines physicochemical filters with a set of functional group filters. Databases are first subjected to property filtering similar to ROF that is followed by checking a set of rules based on the presence of functional groups expected to be problematic. Some examples of these rule-based functional group filters are illustrated in Fig. 2. It is important to note that REOS allows the user to customize each functional group filter as well as the set of rules applied. 

A Pharmacophore Point Filter contains a set of rules based on medicinal chemistry experience and on the observation that non-drug molecules are often underfunctionalized. Fundamental to this system is evaluating molecules for the occurrence of Pharmacophore Points, functional groups that potentially provide key interaction with the receptor or enzyme. The following rules constitute this filter ... 

The filter of Monge et al. includes some additional rules based on molecular structural features. In particular (a) compounds with atoms other than C, H, O, N, S, P, F, Cl, Br, I, Na, K, Mg, Ca, or Li are not allowed to pass the filter (b) no reactive functions (c) no perfluorinated chains (e.g., —CF2CF2CF3) (d) no rings with more than seven members (e) alkyl chains < (CH2)6CH3. 

Peukert et al. (2010a) propose the use of filter operators within match work-flows to prune dissimilar element pairs (whose similarity is below some minimal threshold) from intermediate match results. The threshold is either statically predetermined or dynamically derived from the similarity threshold used in the match workflow to finally select match correspondences. Peukert et al. (2010a) also propose a rule-based approach to rewrite match workflows for improved efficiency, particularly to place filter operators within sequences of matchers. 

---