Neighborhood behavior: A useful concept for validation of 'molecular diversity' descriptors

Journal of Medicinal Chemistry

Volumn 39, Issue 16, 1996, Pages 3049-3059

  Patterson, David E ^a   Cramer, Richard D ^a   Ferguson, Allan M ^a   Clark, Robert D ^a   Weinberger, Laurence E ^b  

^a TRIPOS INC   (United States)

^b NONE   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; CHEMICAL STRUCTURE; CORRELATION FUNCTION; HYDROGEN BOND; STRUCTURE ANALYSIS;

ALGORITHMS; CHEMISTRY, PHARMACEUTICAL; DRUG EVALUATION, PRECLINICAL; INFORMATION SYSTEMS; SOFTWARE; STRUCTURE-ACTIVITY RELATIONSHIP;

EID: 0029783934     PISSN: 00222623     EISSN: None     Source Type: Journal    
DOI: 10.1021/jm960290n     Document Type: Article

References (30)

1. A commercially available realization of these approaches is the Optiverse screening library of compounds, synthesized by Panlabs (Bothell, WA) and designed and distributed by Tripos, and associated screening, lead explosion, and lead optimization services. For more detail of the library design procedures themselves, see: Ferguson, A. M.; Patterson, D. E.; Garr, C. D.; Underiner, T. L. Designing Chemical Libraries for Lead Discovery. J. Biomol. Screen. 1996, accepted for publication.
2. An early review distinguishing the computational approaches for lead generation and optimization is Redl, G.; Cramer, R. D.; Berkoff, C. E. Quantitative Drug Design. Chem. Soc. Rev. 1974, 3, 273-292.
3. Patents pending.
4. Brown, R. D.; Bures, M. G.; Martin, Y. C. Similarity and cluster analysis applied to molecular diversity. American Chemical Society Meeting, Anaheim, CA, 1995; COMP 3.
5. 2/ expected LRT count where: expected LRT count = (LRT area/total area) x total count and LRT is an abbreviation of "lower right trapezoid".
6. The simpler choice of datapoint [x,y] did not behave well in practice.
7. Clark, M.; Cramer, R. D.; van Opdenbosch, N. Validation of the General Purpose Tripos 5.2 Force Field. J. Comput. Chem. 1989, 10, 982-1012.
8. This is the standard "2D screen" generation rules for a UNITY molecular structure database. More exactly, fragments of length 4 do include hydrogen, and 60 of the 988 available bits code for specific atoms or fragments and combinations.
9. BioByte, Inc., Pomona, CA.
10. Kier, L. B.; Hall, L. H. Molecular Connectivity in Chemistry and Drug Research; Academic Press: New York, 1976.
11. Carhart, R. E.; Smith, D. H.; Venkataraghavan, R. J. Chem. Inf. Comput. Sci. 1985, 25, 64-73.
12. Moreau, G.; Broto, P.; Vandycke, C. Nouv. J. Chim. 1980, 4, 359-360, 757-764; Eur. J. Med. Chem. Chim. Ther. 1984, 19, 66-78.
13. Moreau, G.; Broto, P.; Vandycke, C. Nouv. J. Chim. 1980, 4, 359-360, 757-764; Eur. J. Med. Chem. Chim. Ther. 1984, 19, 66-78.
14. Cramer, R. D.; Clark, R. D.; Patterson, D. E.; Ferguson, A. M. Bioisosterism as a Molecular Diversity Descriptor: Steric Fields of Single "Topomeric" Conformers. J. Med. Chem. 1996, 39, 3060-3069.
15. For any receptor atom locus, the nearest lattice point was identified, and then the 27 bits referenced by a cube centered on that point and 4 A wide were set to 1.
16. Martin, Y. C.; Bures, M.; Danaher, E.; DeLazzer, J.; Lico, I.; Pavlik, P. A Fast Approach to Pharmacophore Mapping and its Application to Dopaminergic and Benzodiazepine Agonists. J. Comput.-Aid. Mol. Des. 1993, 7, 93-102.
17. A careful observer will notice that there is a tendency within the bottom graphs for points to be less dense on the right side of the graph. This skewed distribution is expected for differences between the uniformly distributed values within an interval that a random number generator should produce. Note that this innate tendency of value differences opposes any tendency for a descriptor to exhibit a neighborhood enhancement.
18. Such artifactually nonuniform distributions may also produce positive neighborhood distribution enhancements that are actually spurious. This possibility was not directly investigated. However, as can be seen, for example, in the top panels of Figure 2, most of the enhancements appearing in Table 2B are not at all artifactual.
19. 2.
20. Willett, P. Similarity and Clustering in Chemical Information Systems; Research Studies Press: Letchworth, U.K., 1987.
21. To address a point of possible confusion, notice that the distance between two objects is always a single scalar number no matter how many dimensions the space containing those objects has. Thus, for example, the Tanimoto "distance" between two fingerprints is effectively a sum over all thousand-odd dimensions represented by the presence/absence of particular structural fragments, albeit normalized not by the total number of bits settable but instead by the bits actually set in either fingerprint.
22. A good example, which by comparing several libraries addresses the issue of library specific artifacts from factor analysis, is Cummins, D. J.; Andrews, C. W.; Bentley, J. A.; Cory, M. Molecular Diversity in Chemical Databases: Comparison of Medicinal Chemistry Knowledge Bases and Databases of Commercially Available Compounds. J. Chem. Inf. Comput. Sci., accepted for publication. Unfortunately the underlying descriptors used are closely related to those in Table 2C.
23. Martin, E. J.; Blaney, J. M.; Siani, M. A.; Spellmeyer, D. C.; Wong, A. K.; Moos, W. M. Measuring Diversity: Experimental Design of Combinatorial Libraries for Drug Discovery. J. Med. Chem. 1995, 38, 1431-1436.
24. It should however be cautioned that we have here considered only a small subset of the connectivity indices which can be calculated.
25. Madden, D.; Krchnak, V.; Lebl, M. Synthetic combinatorial libraries: Views on techniques and their applications. Perspect. Drug Discovery Des. 1995, 2, 269-285. Chapman, D.; Ross, M. J. Poster at the symposium Chemical and Biolmolecular Diversity, San Diego, CA, Dec 14-16, 1994; lecture at the symposium Exploiting Molecular Diversity: Small Molecule Libraries for Drug Discovery, La Jolla, CA, Jan. 23-25, 1995; conference summary available from Wendy Warr & Assoc., 6 Berwick Ct, Cheshire, U.K. CW4 7HZ.
26. Madden, D.; Krchnak, V.; Lebl, M. Synthetic combinatorial libraries: Views on techniques and their applications. Perspect. Drug Discovery Des. 1995, 2, 269-285. Chapman, D.; Ross, M. J. Poster at the symposium Chemical and Biolmolecular Diversity, San Diego, CA, Dec 14-16, 1994; lecture at the symposium Exploiting Molecular Diversity: Small Molecule Libraries for Drug Discovery, La Jolla, CA, Jan. 23-25, 1995; conference summary available from Wendy Warr & Assoc., 6 Berwick Ct, Cheshire, U.K. CW4 7HZ.
27. Madden, D.; Krchnak, V.; Lebl, M. Synthetic combinatorial libraries: Views on techniques and their applications. Perspect. Drug Discovery Des. 1995, 2, 269-285. Chapman, D.; Ross, M. J. Poster at the symposium Chemical and Biolmolecular Diversity, San Diego, CA, Dec 14-16, 1994; lecture at the symposium Exploiting Molecular Diversity: Small Molecule Libraries for Drug Discovery, La Jolla, CA, Jan. 23-25, 1995; conference summary available from Wendy Warr & Assoc., 6 Berwick Ct, Cheshire, U.K. CW4 7HZ.
28. Cramer, R. D.; Redl, G.; Berkoff, C. E. Substructural Analysis. A Novel Approach to the Problem of Drug Design. J. Med. Chem. 1974, 17, 533. Hodes, L. Clustering a Large Number of Compounds. 1. Establishing the Method on an Intial Sample. J. Chem. Inf. Comput. Sci. 1989, 29, 66-71.
29. Cramer, R. D.; Redl, G.; Berkoff, C. E. Substructural Analysis. A Novel Approach to the Problem of Drug Design. J. Med. Chem. 1974, 17, 533. Hodes, L. Clustering a Large Number of Compounds. 1. Establishing the Method on an Intial Sample. J. Chem. Inf. Comput. Sci. 1989, 29, 66-71.
30. Moreau, G. Proceedings of an IBC Meeting on Combinatorial Chemistry, Oct. 30-31, London, U.K.