Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2001-12-20
2002-12-03
Choules, Jack (Department: 2177)
Data processing: database and file management or data structures
Database design
Data structure types
Reexamination Certificate
active
06490588
ABSTRACT:
TECHNICAL FIELD
The present invention relates to a method of searching a three-dimensional structure databases, which can be utilized for finding novel lead-structures of medicaments, pesticides, and other biologically active compounds.
BACKGROUND ART
Medicaments generally exhibit their biological activities through strong interactions with target biopolymers. In recent years, three-dimensional structures of biopolymers, which play important roles in living bodies, have been revealed successively on the basis of progresses in X-ray crystallographic analyses and NMR technology. With the advance of these researches, methodologies of theoretical approach based on information about three-dimensional structures of biopolymers and their successful results have been reported with respect to lead generations of new drugs, which were conventionally achieved mostly by random screenings or accidental discoveries. Importance of such approach is recognized with increasing interest, and researches from various viewpoints have been conducted in all the world.
An object of these researches is to provide a process for creation of ligand candidate structures by means of a computer. Another object is to provide a process for searching for ligand structures from a three-dimensional structure database containing known compounds. The former (automatic structure construction approach) is advantageous in that the process can suggest a wide variety of possible ligand structures irrespective of known or unknown structures. On the other hand, the latter (database approach) is used in searching for novel biologically active compounds from databases comprised of accumulated structures of known compounds. When a search is applied to a database which consists of compounds stocked in an own company or commercially available compounds, this method is particularly advantageous in that compounds, this method is particularly advantageous in that compounds found to reach criteria of hit (“hit compounds”) can be obtained without any synthetic efforts, and their association constants to target biopolymers and biological activities can be measured.
Synthesis of a compound with a novel molecular skeleton generally needs several months even for an expert, and accordingly, years of and burden of work are required to synthesize a lot of promising structures and measuring their activities. However, if hit compounds are already available, measurements of association constants and biological activities can be easily performed for dozens of promising ligand candidate compounds. In addition, based on the structure of a compound found to have a considerable extent of high association constant or activity, an efficient lead generation is achievable by designing compounds with higher association constants and biological activities and synthesizing the compounds. For these reasons, approaches have recently been interested which comprises the steps of searching three-dimensional structure databases consisting of known compounds and discovering novel pharmacologically active compounds.
However, one of problems of this method is by what query the search is carried out. Generally, a three-dimensional structure database is searched by queries of whether or not given atomic groups or functional groups exist which are presumed as essential for the desired activity in typical biologically active compounds used as templates, or alternatively, whether or not their relative positions are similar to those of the template compounds. Where the structure of a target biopolymer is unknown, basically no approach can be relied on other than this process. However, since these queries are based on assumptions and hypotheses, hit compounds may often fail to have the same activity as that of the template compounds. Where two or more molecules with quite different structural features exhibit similar activities, it can only become possible to create more probable operating hypotheses, with reference to functional groups essential for the activity and their relative positions, by means of structural information about these molecules.
The most difficult problem in the search of three-dimensional structure databases lies in the handling of conformational flexibility of compounds. It has been known that the conformations of a ligand as it binds to a biopolymer (i.e., the most stable complex is formed) do not necessarily accord with any of stable structures of the molecule itself, in the state of a crystal or a solution, or a structure with the most stable energy obtained by energy calculation, and that one ligand molecule can form stable complexes in different conformations depending on target biopolymers. Generally, a set of atomic coordinates representing a three-dimensional structure of one conformation among multiple possible conformations is contained as for one compound in a database. Furthermore, except for databases derived from crystal structures, information of three-dimensional structure of each compound is obtained from two-dimensionally inputted structure through calculation. These three-dimensional structures often represent one of local minimum structures that can be taken by the molecule, per se.
Therefore, if a search is carried out merely on the basis of conformations contained in a database to judge whether or not compounds meet the criteria of hit, most compounds fail to be selected which should be hit if other conformations are considered. Although the number of conformations to be considered may vary depending on the number of rotatable bonds, as well as factors such as degrees of precise consideration of conformations, several tens to hundreds of thousands conformations should be taken into account for a moderately flexible molecule containing 3 to 6 rotatable bonds. In order to consider these possible conformations, available processes are limited to either of those comprising the steps of selecting promising conformations and inputting them in a database beforehand, or alternatively, generating conformations and examining at the time of conducting a search. In any events, enormous computer resources and calculation time are required.
Recently, Kearsley et al. with Merck prepared a database which comprises 20 conformations at most for a single compound and reported “FLOG,” a searching system in which a search is carried out using searching standard mainly consisting of relative positions of functional groups (M. D. Miller, S. K. Kearsley, D. J. Underwood, and R. P. Sheridan: Journal of Computer-Aided Molecular Design, 8, 1994, pp.153-174, FLOG: A system to select ‘quasi-flexible’ ligands complementary to a receptor of known three-dimensional structure). It was reported that the FLOG took approximately one week to complete searching a database containing 2,000,000 conformations for 100,000 compounds by means of a super computer CRAY. The twenty conformations in average for each compound are stored by selecting energetically stable local minimum structures through prior conformational analysis of each compound, for which enormous time and burden of work are needed. Nevertheless, twenty conformations per a single compound are insufficient.
In addition, the method adopts the query which comprise the presence or absence of functional groups presumably essential for the activity of a compound as a template, as well as distance, angle, or direction between the groups. These queries are most plain among possible queries, and although they have advantages to shorten calculation time using simple algorithm, high probability cannot be expected that hit compounds actually have the desired activity. The reason lies in that a molecule having inappropriate whole molecular shape and size fails to exhibit activity, even if functional groups merely have desired relative positions. If a three-dimensional structure of the target biopolymer is known, the most effective search can be achieved by means of such information, which provides high probability that hit compounds have the desired activity.
For example, Eyermann et al. with Dupont Merck performe
Itai Akiko
Mizutani Miho
LandOfFree
Method of searching novel ligand compounds from... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of searching novel ligand compounds from..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of searching novel ligand compounds from... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2951473