Image analysis – Applications – 3-d or stereo imaging analysis
Reexamination Certificate
1995-02-17
2002-09-17
Johnson, Timothy M. (Department: 2623)
Image analysis
Applications
3-d or stereo imaging analysis
C382S128000, C382S209000, C382S305000, C702S019000
Reexamination Certificate
active
06453064
ABSTRACT:
BACKGROUND OF THE INVENTION
1) Field of the Invention
This invention relates to a common structure extraction apparatus which extracts a common structure from two three-dimensional structures each formed from a set of sequenced points, and more particularly to an apparatus which retrieves and extracts analogous common portions from a plurality of substances having different three-dimensional structures.
2) Description of the Related Art
In the fields of physics and chemistry, in order to investigate a property of a novel or unknown substance or artificially produce a new substance, a molecular structure is analyzed to make clear a mechanism for manifestation of a function of the substance.
Thanks to results of investigations in the past, it is known that a function and a three-dimensional structure of a substance have a close relation to each other, and it is considered that a structurally analogous (or specific) portion contributes very much to a function of a substance.
Thus, three-dimensional structures of various substances have been made clear and determined by such techniques as an X-ray crystal analysis or an NMR (Nuclear Magnetic Resonance) method, and data bases are produced for the three-dimensional structures which have thus become clear.
When a research worker tries to retrieve and extract analogous portions between three-dimensional structures from such a data base as described above, a series of cumbersome operations must be performed. If such retrieval and extraction can be performed automatically, then the burden of the series of operations to the research worker can be reduced.
In recent years, in order to assist clarification and production of a novel substance and modification to a function of a known substance, much effort has been and is being directed to operations to determine a three-dimensional structure of an object substance by such a technique as an X-ray crystal analysis or an NMR method and store the thus determined three-dimensional structure into a data base. One of representative data bases which are spread world-wide is the Protein Data Bank (PDB) in which three-dimensional structures of proteins, ribonucleic acids and like substances are registered. Further, the Cambridge Structural Database (CSD) is known as a data base in which chemical substances are registered.
A protein is constituted from a plurality of amino acids connected to each other by way of peptide linkages like a chain folded in vivo to form a three-dimensional structure and manifests various functions. The individual amino acids are represented by numbering them in order from 1 beginning with a terminal of N (nitrogen) and ending with the other terminal of C (carbon). The numbers are called amino acid numbers or amino acid residue numbers.
A protein is normally constituted from about 20 amino acids and is arranged stably including a portion having an a helix structure, another portion having a beta structure which extends generally linearly in a zigzag pattern and a further portion of a disordered random coil structure at a variable rate. Meanwhile, each amino acid is constituted from a plurality of atoms depending upon the kind thereof. Accordingly, information including the name of a protein, a management number, the numbers of amino acids forming the protein, the kinds and three-dimensional coordinates of atoms constituting each of the amino acids is registered in the PDB.
Again, thanks to results of investigations in the past, it is known that a function and a three-dimensional structure of a substance have a close relation to each other, and much effort is directed to operations to make clear the relation between a function and a structure. Above all, since it is considered that a structurally analogous (or specific) portion between different substances having a same function contributes very much to the function of the substances, it is essentially required to find out an analogous structure which exists commonly between different three-dimensional structures.
Under present conditions, however, since no technique is available to directly extract a characteristic portion from three-dimensional coordinates of a three-dimensional structure of a substance, each research worker manually searches for a characteristic portion by displaying each three-dimensional structure by means of a 3D (three-dimensional) graphic system. Generally, there is no fixed method for deciding the orientation of a substance, and since a substance is rotated with reference to another substance to search for a characteristic portion thereof, much time is required for the operation.
When a research worker searches for an analogous three-dimensional structure, an rmsd (root mean square distance) value is used as a scale for the analogy between three-dimensional structures of substances. The rmsd value is a square root of a mean square distance between components of substances matched with each other. Empirically, where the rmsd value between two substances is smaller than 1 angstrom, it is considered that the two substances are very analogous to each other.
A popular method for calculation of an rmsd value will be described below with reference to FIGS.
79
(A) to
79
(D).
It is assumed that there are a substance A represented by such a point set P={p
1
, p
2
, . . . . , p
i
, . . . , p
N
} as shown in FIG.
79
(A) and another substance B represented by such a point set T={t
1
, t
2
, . . . . , t
j
, . . . , t
N
} as shown in FIG.
79
(B). The elements (points) constituting the substances A and B are matched with each other as shown in FIG.
79
(C), and the substance B is rotated or moved and superposed on the substance A as shown in FIG.
79
(D) so that the rmsd value between the thus matched elements may exhibit a lowest value. The rmsd value is calculated in accordance with the following equation:
rmsd
=
∑
k
=
1
N
(
w
k
(
Ut
k
-
p
k
)
)
2
N
where N is the number of the matched points, U is a rotation matrix, and w
k
is a weight at each of the matched points.
A technique for calculation of rotation or movement of a substance which minimizes the rmsd value between such matched points has been proposed by Kabsh et al. and is utilized widely at present.
However, since the technique compares different substances with each other in regard to an equal number of points, it is the existing state of the things that it is found out by trial and error of a research worker what matching between components of a substance and components of the other substance minimizes the rmsd value.
Further, in order to produce a novel substance, existing substances must be investigated. For example, when it is intended to increase the heat resisting property of a certain substance, a structure common to substances which are high in heat resisting property is searched out, and the structure is additionally provided to the substance to be produced newly to promote the function.
Accordingly, a function of retrieving a necessary structure from a data base is required. However, it is the existing state of the things that, because of a similar reason to that described above, a structure is searched out from a data base by trial and error of a research worker using a computer graphic system in a similar manner as described above.
Furthermore, in recent years, the importance of an analogous structure is recognized also in designing and improvement of a protein. One of example is an experiment for improvement in function of human lysozyme (HL). It has been found out that the three-dimensional structure of the protein HL which does not have an activity to couple a calcium ion includes a structure analogous to that of &agr; lactalbumin which is a protein which couples a calcium ion.
Thus, it has been reported that an experiment to replace an amino acid at a portion of the structure in HL with another amino acid of a different kind by a genetic recombination operation proved coupling of the resulted substance to a calcium ion (Kuroki R. et al., Proc. Natl. Acad. Sci, U.S.A., 86, pp.6,903-6,
Aikawa Seiichi
Matsuzawa Fumiko
Tomikawa Mayumi
Johnson Timothy M.
Staas & Halsey , LLP
LandOfFree
Common structure extraction apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Common structure extraction apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Common structure extraction apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2890806