On Highly Similar Molecules

Experimental results from an HTS assay are not the truth but merely an estimate of the true potency of a molecule. Because molecules are only measured one time in the HTS setting, there is a high degree of uncertainty. One thing that can be done is to look for highly similar molecules and treat them as pseudo replicates of the same parent molecule. See Goldberg (1978) for further discussion on estimating the error rate. 

Molecular identification numbers (ID) were introduced in 1984 [10] and have been used as molecular descriptors in QSAR [11] although they were constructed initially as a tool for identification of highly similar molecules. We should add that the problem of identification of highly similar molecules is different and distinct from the problem of graph isomorphism in the sense that for the former occasional occurrence of identical ID numbers for distinct structures is acceptable (as it may point to very similar structures), and in the latter, the occurrence of identical numerical values for different structures points to a failure of the approach. In the following section, we give more information on molecular ID numbers and their properties. 

Essentially, all computed similarity values lie on the unit interval [0,1] of the real line (more correctly the unit interval of the rational line). Highly similar molecules have values at the high end of this scale, while dissimilar molecules tend to lie at the lower end. Humans can, in general, assess the similarity of very similar objects, as chemists can assess the molecular similarity of molecules with similar structures. But what happens when molecules become less similar (more dissimilar) There... 

Acid-catalyzed hydrogen exchange is used as a measure of the comparative reactivity of different aromatic rings (see Table 5). These reactions take place on the neutral molecules or, at high acidities, on the cations. At the preferred positions the neutral isoxazole, isothiazole and pyrazole rings are all considerably more reactive than benzene. Although the 4-position of isothiazole is somewhat less reactive than the 4-position in thiophene, a similar situation does not exist with isoxazole-furan ring systems. 

Somsen et al. [796] have reported the use of SERR spectroscopy for the in situ selective determination and semi-quantitative analysis of structurally similar dyes separated by TLC. The limits of identification of the TLC-SERRS method (ca. 5ng applied) were sufficient for acquisition of spectra of impurities present in the certified dye standards. SERRS may also be used for in situ identification of highly fluorescent molecules on HPTLC plates. 

One of the simplest applications of the HSAB principle is related to solubility. The rule "like dissolves like" is a manifestation of the fact that solute particles interact best with solvent molecules which have similar characteristics. Small, highly charged particles or polar molecules are solvated best by solvents containing small, highly polar molecules. Large solute particles having low polarity are solvated best by solvent molecules having similar characteristics. Consequendy, NaCl is soluble in water, whereas sulfur, S8, is not. On the other hand, NaCl is insoluble in CS2, but S8 dissolves in CS2. 

Initially the substance at Rt 19.95 was identified as 2-nonen-l-ol based on mass spectrum library search. The comparison with a commercial 2-nonen-l-ol standard indeed revealed a high degree of similarity between the mass spectra, but a distinct deviation regarding the retention time suggesting a similar molecule with a chain length greater than 2-nonen-l-ol. The substance Rt 20.95 was tentatively identified as 6,10-dimethyl-5,9-undecadien-2-one which corresponds with the authentic standard regarding mass spectra and retention time. 

Molecular similarity (similarity-based virtual screening) on the basis of one or a small set of known actives, molecules showing a high similarity concerning specific features stored in a molecular descriptor are searched for. 

---