JPH07155169A - Method and device for displaying and extracting locally similar sequence of biopolymer - Google Patents

Abstract

(57) [Summary] [Object] To provide a method and a device for displaying and extracting a region having a high degree of similarity from a sequence alignment result selected as a locally similar sequence of a biopolymer composed of a sequence of constituent elements. [Arrangement] A juxtaposition result of arrays having locally similar arrays is calculated by a method such as a dynamic programming method. The alignment result is obtained as a graph in which the first axis is the element number indicating the order of bases or amino acids that are the elements of one of the aligned sequences, and the second axis is the cumulative value of the scores up to that element number, and output indicate. The gradient of the graph is large in a region having a high degree of similarity. [Effect] From the results of protein juxtaposition, the low similarity region,
The non-similarity region and the high similarity region are easily identified and further extracted automatically.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for comparing sequences of constituent elements of biopolymers such as nucleic acids and proteins, and in particular,
The present invention relates to a technique for extracting a local similar sequence stored in a sequence of constituent elements of a plurality of biopolymers.

[0002]

2. Description of the Related Art In recent years, sequence data of bases and amino acids, which are constituents of biopolymers such as nucleic acids and proteins, have been remarkably accumulated, and databases of these sequence data have been prepared. In order to extract biologically meaningful information from these databases, sequences of biopolymer constituents are compared with each other and grouped according to the similarity of the sequences, or multiple biomolecules are compared. It is useful to search for local subsequences that are conserved in the sequences of the constituents of the molecule and are particularly relevant for their physiological or chemical function.

Conventionally, a method for comparing the base sequences of nucleic acids and the amino acid sequences of proteins to search for their local similarity is to evaluate the similarity of the entire sequence of the constituent elements of the biopolymer. In the Needleman (Needlema
n) and Wunsch (Journal of Molecular Biology, 48, 444-
P. 453 (1970) (J. Mol. Biol. 48, 444-453.
(1970))), and then a method based on dynamic programming, which was improved by Smith and Waterman as a local similarity search method, is widely used (Journal of Molecular Biology, 147 volumes). 195-197 (1981)
(J. Mol. Biol. 147, 195-197 (1981))). In this method, the constituent elements of the sequences of two biopolymers to be compared are juxtaposed, and the locality is based on a score representing the match between the juxtaposed elements, the mismatch, and the introduction of a gap due to the insertion or deletion of the constituent elements. Compute the sum of scores in similar regions to derive the optimal juxtaposition. However, this dynamic programming-based method takes a very long time because the calculation time is proportional to the product of the array lengths of the components to be compared. Therefore, when comparing the newly determined sequence data of nucleic acids or amino acids with the sequence data of biopolymer components in the database, a faster method, FASTA (Proceedings of National Academy of Sciences) is used. USA,
Volume 85, pp. 2444-2448 (1988) (Proc.
Natl. Acad. Sci. USA, 85, 2444-2448 (1988)), or BLAST (Journal of Molecular Biology, 215, 403-410 (1990) (J.
Mol. Biol., 215, 403-410 (1990)) is often used.

However, in any of these methods, the arrangement of the constituent elements of the two biopolymers to be compared is juxtaposed in the region where the total score takes a maximum value from 0. Fig. 1 shows the alignment result of the amino acid sequence portion having the highest total score among the results of comparing the amino acid sequences of the Drosophila notch protein and the Xenopus xotch protein, which are highly similar proteins, by the BLAST method. Is shown. Hereinafter, a key array used as a reference in comparison and an array to be compared will be referred to as a target array. The juxtaposition of amino acid sequences of notch protein and xotch protein was used as a key sequence for comparison.
It is made for the amino acid sequences 271 to 689 of the notch protein. In the result of the BLAST method shown in FIG. 1, the column labeled Query is the amino acid sequence of the key sequence of the notch protein, Sbj
The columns labeled ct show the amino acid sequences of the target sequence, xotch protein, in one-letter representation. Between the Query column and the Sbjct column, when the key sequence and the target sequence match, the amino acid is indicated by a one-letter expression. In the case of disagreement, + is shown when the score has a positive value, and nothing is shown when the score has a negative value. In the score calculation in the BLAST method, a score matrix called PAM120 shown in FIG. 3 is used. The elements of this matrix indicate, when comparing two amino acid sequences, with what probability each pair of amino acid residues contained in each sequence arises from a sequence of unrelated amino acids. The values of the elements of the matrix are expressed as the logarithm of the probability that a pair of amino acid residues will result from a mutation from a common ancestor molecule divided by the probability that it will occur by chance from a sequence of unrelated amino acids. Therefore, when the values of the elements of the matrix are positive, the probability of having a common ancestor molecule is higher and the similarity is higher than it happens by chance, and the negative value is the opposite of the positive value. Respectively. Therefore, Qu in Figure 1
Between the ery column and the Sbjct column, the more the one-letter expression of the matching amino acids and the region with more + symbols, the higher the total score, and the higher the sequence similarity of the constituent elements. As the score matrix, PAM12 shown in FIG. 3 is used.
0 and PAM250 shown in FIG. 4 are widely used. 3 and 4 show 120 and 25 per 100 residues, respectively.
This is a calculation of the probability when 0 residue is mutated.

From the viewpoint of the degree of similarity between two amino acid sequences, FIG. 1 can be roughly divided into three regions. That is, the first from array number 271 to array number 335 in the Query column
Region, the second region from SEQ ID NO: 335 to SEQ ID NO: 456, the vicinity of SEQ ID NO: 456 to SEQ ID NO: 689
It can be roughly divided into the third area up to the turn. It seems that the first and third regions have a high degree of similarity, and the second region has no similarity. This is because the algorithm of the BLAST method does not allow the existence of a gap caused by the insertion or deletion of the constituent elements when the sequences of the constituent elements of the two biopolymers are aligned.

The alignment results obtained by comparing the amino acid sequences of the Drosophila notch protein and the Xenopus xotch protein by the FASTA method by selecting the same sequence comparison target as in FIG. 1 are shown in FIG. Similar to Figure 1,
In the amino acid sequence of FIG. 2, the upper column is the amino acid sequence (key sequence) represented by the one-letter notch protein, and the lower column is xo.
The amino acid sequences (target sequences) represented by one-letter expressions of tch proteins are shown. The FASTA method, like the dynamic programming method, is an algorithm that allows gaps caused by the insertion and deletion of components. In obtaining the results of FIG. 2, PAM250 was used as the score matrix, and 12 and 4 were used as the penalty scores for the introduction and extension of gaps caused by the insertion and deletion of the constituents, respectively. According to the result of FIG. 2, the target array corresponding to the key array 342 and the key array 45
It can be seen that there is a gap at position 7 and there is a gap. That is, in the result of the BLAST method shown in FIG. 1, the sequences of the key sequences from 342 to 457 are deviated from the originally expected alignment result by one amino acid residue, and it is judged that there is no similarity. Will be. That is, the result of juxtaposition by the BLAST method includes a region having no similarity at all. This is based on the principle that in the BLAST method, juxtaposition is performed in a region where the score total takes a maximum value from 0, and within this region, the score total can have any value larger than 0 and smaller than the maximum value. It is due.

[0007]

DISCLOSURE OF THE INVENTION A region having a high degree of similarity and a region having a low degree of similarity in the arrangement of constituent elements of a biopolymer are identified only by comparing the juxtaposition results of the constituent elements, and a local arrangement having a high degree of similarity is extracted. It's not easy. The first object of the present invention is to
A method of easily identifying a high similarity region having a high similarity from a juxtaposed result of an arrangement of components including a low similarity region having a low similarity degree selected as a local similarity sequence and a dissimilarity region having no similarity, To provide a device. A second object of the present invention is to provide a method and device for automatically extracting a highly similar region from the juxtaposed result of the arrangement of the constituent elements.

[0008]

In order to achieve the first object, in the present invention, the sequences of the constituent elements of biopolymers such as nucleic acids and proteins are compared, and the constituent elements are matched, mismatched, and introduced a gap. , Etc., and the obtained alignment result of locally similar sequence parts is calculated based on the sum of scores representing the similarity between the components of the two biopolymers. Alternatively, the element number indicating the sequence order of (amino acid) is taken on the first axis, and the predetermined parameter obtained from the score between the juxtaposed components is the second parameter.
A multi-dimensional graph is displayed on the axis of. This graph is smoothed. As the predetermined parameter, the cumulative value of the scores from the predetermined element number on the first axis to each element number or the score value itself of each element number is taken. As the locally similar sequence portion, for example, a result obtained by a dynamic programming method, FASTA method, BLAST method, or the like is used.

Further, in order to realize the second object,
In the present invention, the gradient of the graph having the cumulative value of the scores from the predetermined element number on the first axis to each element number as a predetermined parameter has a value continuously equal to or larger than the predetermined value with respect to the first axis. The area is identified and extracted, and the cumulative score value and the arrangement of the constituent elements in the area are juxtaposed and output and displayed. Alternatively, an area in which the score value of the second axis of the graph in which the score value of the element number itself is a predetermined parameter is continuously larger than the predetermined value with respect to the first axis is identified and extracted, and the area is extracted. The cumulative score value in and the array of constituent elements are juxtaposed and output.

[0010]

As described above, when the cumulative value of the scores from the predetermined element number of the first axis to each element number is used as the predetermined parameter, the score between the elements of the array of constituent elements is set in the high similarity region. Often takes a large positive value, so the gradient of the graph becomes large. On the other hand, in the low similarity region, the gradient of the graph becomes small, and in the dissimilarity region where there is no particular similarity, the score between each element of the array of constituent elements often takes a negative value, and the gradient of the graph is negative. become. Therefore, the low similarity region and the dissimilarity high similarity region degree can be identified in the juxtaposed result of the arrangement of the constituent elements only by looking at the obtained graph.

Further, when the score value itself of the element number is used as a predetermined parameter, the score between each element of the arrangement of the constituent elements often takes a large positive value in the high similarity region, so that the second graph The value of the axis becomes large. On the other hand, in the low similarity region, the value of the second axis of the graph becomes small, and in the dissimilarity region, the score between each element of the array of constituent elements often takes a negative value, and the value of the second axis of the graph The value will be negative.
Therefore, the low similarity region and the dissimilarity high similarity region degree can be identified in the juxtaposed result of the arrangement of the constituent elements only by looking at the obtained graph. Further, since the graph is smoothed, it is effective in identifying a subtle difference in similarity in a region where the similarity is low as a whole.

In order to automatically extract the high similarity region, when the cumulative value of the score from the predetermined element number of the first axis to each element number is used as the predetermined parameter, the region having a large gradient in the graph is used. When the score value itself of the element number is used as the predetermined parameter, it is only necessary to set threshold values for the areas having large values on the second axis of the graph and extract them. This threshold value may be set in advance with a constant gradient or score value as a reference value, or it may be calculated based on the average value of the gradient or score values of the entire graph obtained from the juxtaposed result of the arrangement of the constituent elements. You may enable it to set automatically. In order to efficiently and automatically extract the high similarity region, it is effective that the graph is smoothed and is less affected by the variation in the score value of each element of the arrangement of the constituent elements.

[0013]

EXAMPLE An example of the present invention will be described with reference to FIG. In this example, obtained by the BLAST method shown in FIG.
The alignment of the amino acid sequences of the Drosophila notch protein and the Xenopus xotch protein is shown graphically. The horizontal axis represents the number indicating the order of amino acids, and the vertical axis represents the cumulative value obtained by adding the score value of PAM120, the score matrix shown in FIG. 3 to each amino acid pair from the start position of the alignment of amino acid sequences. . The horizontal axis is the key sequence, which is the starting point of the alignment of amino acid sequences, and 2 of notch protein.
The number 71 is set to 1. As can be seen at a glance in the graph of FIG. 5, the values (numbers) on the horizontal axis are 1 to 64, and 187 to 4
In the region up to 15, the graph has a positive gradient and a high degree of similarity, and from 65 to 186 has a negative gradient and a low degree of similarity can be easily determined. In the present embodiment, the region having a positive gradient and the region having a negative gradient can be roughly classified, so that a local array having a high degree of similarity can be extracted by identifying a portion having a positive gradient. However, depending on the amino acid pair, there are also pairs with a negative value even in regions with high similarity,
In order to automatically extract local similar sequences, it is desirable to perform smoothing processing at several points along the horizontal axis. The cumulative score of automatically extracted local similar sequences was calculated from the increase in the positive slope portion of the graph, and the cumulative score of the number of matching amino acid pairs in the aligned portion and the total number of aligned amino acid pairs in the aligned portion of the amino acid sequence were calculated. By obtaining the degree of coincidence from the ratio and outputting and displaying it together with the juxtaposed portion of the amino acid sequence, it is possible to obtain information on the partial sequence having a high degree of similarity.

An embodiment of the present invention will be described with reference to FIGS. In FIG. 6, the alignment result of the amino acid sequences shown in FIG. 1 is shown by a number indicating the order of amino acids on the horizontal axis, a score matrix for amino acid pairs at each position on the vertical axis, and PA.
The score value of M120 is shown. The horizontal axis is the key sequence that is the starting point of the alignment of amino acid sequences, 271 of notch protein.
The amino acid residue numbered was 1. As can be seen from the graph of FIG. 6, the values on the horizontal axis are from 1 to 64, and 187.
It can be discriminated that the region from No. to 415 has a positive value and a high similarity, and the region from 65 to 186 has a negative value and a low similarity. In FIG. 6, regions having an average positive score value and regions having an average negative score value can be roughly classified, so that a local sequence having a high degree of similarity can be extracted by identifying a portion having a positive score value. However, in this case as well, there are some pairs having a negative value depending on the amino acid pair, so in order to automatically extract the local similar sequence, it is desirable to perform smoothing processing at several points along the horizontal axis. FIG. 7 shows the result of arithmetic mean of FIG. 6 at 11 points along the horizontal axis. According to FIG. 7, the variation due to the score of the amino acid pair is suppressed by averaging, so that the region with high similarity and the region with low similarity can be easily distinguished. Further, depending on whether the score is positive or negative, it is possible to automatically extract a region having a high degree of similarity.

An embodiment of the present invention will be described with reference to FIG.
The example of FIG. 8 shows the amino acid sequences of a human calmodulin protein belonging to the superfamily of proteins called calmodulin (proteins classified according to the degree of sequence similarity) and a protein called dnaK of E. coli.
The results obtained by comparison by the method of Smith-Waterman are displayed in a graph. Here, the matrix shown in FIG. 4, PAM250, is used for score calculation. Further, 12 and 4 were used as the penalty scores for introducing and extending the gap with respect to the amino acid sequence, respectively. It is known that the amino acid sequence of a protein called dnaK of Escherichia coli has the largest total score among proteins not belonging to the calmodulin superfamily when compared by the Smith-Waterman method or the FASTA method. When human calmodulin consisting of 149 amino acid residues is used as a reference sequence for comparison, the amino acid sequence can be juxtaposed with E. coli dnaK protein over almost its entire length. The alignment length of the amino acid sequence extends to 120 amino acid residues including insertions and deletions. However, as can be seen from FIG. 8, it is only the region of 30 amino acid residues that contributes to the score value of optimal alignment. This region corresponds to one of the four EF hands of human calmodulin, which is a motif involved in binding to calcium ions.
Also according to this embodiment, a region having a high degree of similarity can be easily identified from the result selected as the locally similar sequence.

An embodiment of the present invention will be described with reference to FIG.
In this example, K1HUAG, which is a protein belonging to the immunoglobulin V region superfamily, is used as a key sequence, and another protein K2 belonging to the immunoglobulin V region is used.
The results obtained by performing the sequence comparison of DGGM, KVRBB1 and S09230 (all of which are entry names of the protein sequence database PIR) by the dynamic programming method based on the Smith-Waterman algorithm are shown in FIG.
Graphs are shown in (a), FIG. 9 (b), and FIG. 9 (c), respectively. In the score calculation, the score matrix PAM250 shown in FIG. 4 was used. In addition, 12 and 4 were used as penalty scores for introducing and extending a gap with respect to an amino acid sequence, respectively. As a result of the sequence comparison, the degree of amino acid identity between the two proteins was 51% to 53% in the optimal alignment of the amino acid sequences. In FIG. 9, the abscissa axis represents the number of amino acid residues of K1HUAG as a key sequence, and the ordinate axis represents the cumulative value of the score for each amino acid pair from the alignment start position of the amino acid sequences. The V region of immunoglobulin light chain is composed of four skeletons (F
W) and three loop structure parts (CDR). The loop structure part (CDR) is a part related to antigen binding, and is a part with a large amino acid mutation due to the necessity of diversity of antigen recognition. On the other hand, the skeletal part (FW) is a part that determines the structure of immunoglobulin protein, and its amino acid sequence is relatively well conserved. FIG. 9 also shows these positions. As can be seen from FIG. 9, the gradient is relatively large in the FW region and small in the CDR region. That is, the above-mentioned knowledge about the conservation of the sequence and the division of each region can be easily obtained.

An embodiment of the present invention will be described with reference to FIGS. In FIG. 9, there are three locally large gradient regions indicated by elliptical frames 8, 9, and 10. The number of proteins belonging to the immunoglobulin V region superfamily is large, and their sequences are also diverse. Smith-Waterm
Regarding the comparison of the amino acid sequence similarity of different proteins by the method of an, the method of FASTA, the method of BLAST, etc.,
When searching the amino acid sequence database of proteins, the sensitivity and selectivity of selecting the proteins that belong to the superfamily as much as possible and not selecting the proteins that do not belong to the superfamily are measures of the algorithm performance. However, whichever algorithm is adopted, it is impossible to determine whether or not the sequence selected only from the optimal alignment score belongs to the same superfamily as the reference sequence. Therefore, in addition to the score, the elliptical frames 8, 9,
It is conceivable to add a characteristic local site as shown in 10 to the criterion. 9 (a) to 10 (e), FIG.
Similarly to the above, using K1HUAG as a key sequence, among the results obtained by the method of Smith-Waterman, some alignment results showing the lowest score among those belonging to the same superfamily are shown by a graph. 27% to 19% amino acid sequence identity
Met. These graphs also show the area outside the region of optimal juxtaposition. Therefore, the score value on the vertical axis is a relative value.
Although almost no similarities are seen as a whole, regions 8 ′, 9 ′, and 10 ′ corresponding to the elliptical regions shown in FIG. 9 are stored in many examples. Therefore, the combination of the characteristic pattern of the graph as shown here and the score value of the optimal juxtaposition suggests that the superfamily members can be identified with high accuracy.

An embodiment of an apparatus for carrying out the method of the present invention described above will be described with reference to FIG. In this embodiment, biopolymer array input means 1 for obtaining the optimum alignment, alignment operation means 2 for obtaining the optimum alignment, graphing operation means 3 for obtaining the graph illustrated in FIG. 5 from the optimum alignment result, and this graph is output and displayed. And a calculation program for obtaining the optimum alignment, a program for plotting the optimum alignment result, and the like, specifically, hardware such as a computer main body, an external storage device, a CRT, a printer, and a keyboard. Composed of software.

The input means inputs data such as a keyword for specifying a protein for which optimum alignment is to be obtained from a keyboard of a computer, and the amino acid sequence of the protein is stored in an amino acid sequence database 5 held in an external storage device of the computer. It is composed of means for retrieving data and means for copying the obtained array data in the internal storage device of the juxtaposition arithmetic means. If the amino acid sequence is known in advance,
Amino acid sequence data may be directly input from the keyboard.

The juxtaposition calculating means finds the optimum juxtaposition of the amino acid sequence data copied in the internal storage device by the input means and the sequence data sequentially copied from the amino acid sequence database in the external storage device to the internal storage device in the juxtaposition calculating means. ,
It comprises means for obtaining based on the scores of agreement, disagreement, gap introduction, etc. between each amino acid, and selection means for selecting a significant sequence pair based on the value of the score of optimum alignment obtained. The obtained optimum juxtaposition result is transferred to the internal storage device of the graphing calculation means. When the pair of arrays to be compared is known in advance, the array data of the array pair to be compared may be copied to the internal storage device of the juxtaposition calculation means by the input means. In this case, the result obtained by the optimum juxtaposition calculation may be directly transferred to the internal storage device of the graphing calculation means without passing through the selection means.

The graphing calculation means calculates the graph by the method for obtaining the graph shown in each of the above-described embodiments based on the optimum juxtaposition result calculated and selected by the juxtaposition calculation means and transferred to the internal storage device. Ask for. The obtained graph data is transferred to the output means and displayed for output. In this embodiment, the output means is composed of a CRT and a printer.

An embodiment of an apparatus for carrying out the method of the present invention will be described with reference to FIG. In this embodiment, in addition to the means constituting the apparatus of the above-mentioned embodiment, a selection calculation means 6 for selecting an area having particularly high similarity from the obtained graph, a cumulative score of the selected area, and a degree of coincidence. The output value calculating means 7 to be obtained and the output means 4 for outputting the cumulative score, the degree of coincidence, and the juxtaposition result are included. The selection calculation means is based on the graph data transferred to the internal storage device of the selection calculation means, based on the graph data, an area in which the gradient of the graph is continuously a certain value or more, or a score value of the graph is a value continuously more than a certain value. It is constituted by an arithmetic means for selecting a certain area and determining the boundary of the area.
The output value calculation means calculates the cumulative score value of the region selected by the selection calculation means and the degree of coincidence indicating the amino acid coincidence rate. The obtained cumulative score value, the degree of coincidence, and the juxtaposed result of the selection areas are transferred to the output means and displayed for output. Also in this embodiment, the output means is composed of a CRT and a printer. As for the display, a plurality of target arrays can be displayed in duplicate in two-dimensional coordinates or in multi-dimensional coordinates for one key layout. At this time, the data on each target array can be displayed in color to facilitate the identification. Furthermore, the slope in the cumulative score graphical display, or the score value in the score graphical display,
The positive and negative portions of can be displayed in different colors to facilitate identification.

In the above examples, the explanation has been centered on the results relating to the amino acid sequences of proteins, but it is clear that the present invention is also effective when comparing the sequences of nucleic acids. In particular,
When comparing genomic DNA composed of exons and introns with cDNA sequences, conventional local similar sequence extraction methods such as dynamic programming method, FASTA method and BLAST method cannot distinguish the boundaries of exons and introns. In some cases, not only sequences that are completely matched in exons but also sequences that have no similarity in introns may be extracted as similar sequences. There was a problem when assessing sex. However, if the present invention is used, for example, it is possible to clearly discriminate the completely matched portion and the unmatched portion from the gradient of the graph, automatically extract the completely matched portion, and output and display the score value, the matching degree, and the juxtaposition result. Thus, the above identity can be evaluated without error.

[0024]

EFFECTS OF THE INVENTION According to the present invention, a region having a high degree of similarity can be easily identified from the optimal alignment result of similar nucleic acids having a low degree of similarity and a dissimilarity region, and a protein can be automatically extracted. Is possible. Furthermore, the position and boundary of each domain of a protein composed of a plurality of domains or the positions and boundaries of nucleic acid introns and exons can be easily identified.

[Brief description of drawings]

FIG. 1 is an example of one output result according to one prior art related to the present invention.

FIG. 2 shows an example of one output result according to one prior art related to the present invention.

FIG. 3 is an example of a score matrix used in the present invention.

FIG. 4 is an example of a score matrix used in the present invention.

FIG. 5 is a graph showing an example of an output result according to the present invention.

FIG. 6 is a graph showing an example of an output result according to the present invention.

FIG. 7 is a graph showing an example of an output result according to the present invention.

FIG. 8 is a graph showing an example of an output result according to the present invention.

FIG. 9 is a graph showing an example of an output result according to the present invention.

FIG. 10 is a graph showing an example of an output result according to the present invention.

FIG. 11 is a block diagram of an embodiment of an apparatus in which the method of the present invention is implemented.

FIG. 12 is a block diagram of an embodiment of an apparatus in which the method of the present invention is implemented.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Input means, 2 ... juxtaposition calculation means, 3 ... graphing calculation means, 4 ... output means, 5 ... sequence database, 6 ... selection calculation means, 7 ... output value calculation means, 8, 9, 10 ... locally Area with a large gradient, 8 '... Area 8 with a large gradient locally
Region corresponding to the region 9 '... Region having a large gradient locally 10' ... Region having a large gradient 10
Area corresponding to.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Tetsuo Nishikawa 1-280, Higashi Koikekubo, Kokubunji, Tokyo Inside Hitachi Central Research Laboratory (72) Inventor Hisamitsu Kawaguchi 1099, Ozenji, Aso-ku, Kawasaki-shi, Kanagawa Hitachi, Ltd. In-house System Development Laboratory (72) Inventor Susumu Hiraoka 1099 Ozenji, Aso-ku, Kawasaki-shi, Kanagawa Hitachi Ltd. System Development Laboratory (72) Inventor Naoko Kasahara 1-280 Higashi-Kengokubo, Kokubunji, Tokyo Hitachi Ltd. Central Research Laboratory (72) Inventor Hideki Kamihara 1-280 Higashi Koigokubo, Kokubunji, Tokyo Hitachi Ltd. Central Research Laboratory (72) Inventor Ritsuji Okayama 6-81 Onoue-cho, Naka-ku, Yokohama, Kanagawa Hitachi Software Engineering Co., Ltd. Meeting

Claims (26)

[Claims] 1. A method for displaying a locally similar sequence of a biopolymer composed of an array of a plurality of constituent elements, the method comprising the step of comparing the arrangement of the constituent elements of at least two of the biopolymers with each other. , A step of obtaining a score for evaluating information regarding the similarity of sequences including a mismatch, a gap introduction, and a gap deletion, and the sequences of the constituent elements of a plurality of the biopolymers are juxtaposed and compared based on the score. In the method for displaying a locally similar biopolymer array, the first axis has an element number indicating an array order of the constituent elements of the one of the aligned biopolymers, and the second axis A predetermined parameter obtained from the score between the juxtaposed constituents on the axis of, and converting the juxtaposed result into graph data; Display method biopolymer local similarity sequence, characterized by having the steps of multidimensional display the data as a graph, a. 2. The method according to claim 1, further comprising a step of performing a smoothing process on the graph data, to display a biopolymer locally similar sequence. 3. The method according to claim 1 or 2, wherein the predetermined parameter is a cumulative value of the scores from the predetermined element number to each of the element numbers. Method for displaying locally similar sequence of body macromolecule. 4. The method according to claim 1 or 2, wherein the predetermined parameter is a score value at each of the element numbers. 5. The method according to claim 3, wherein the gradient of the graph data is continuous with respect to the first axis, and the value of the second axis is equal to or larger than a predetermined constant value. A biopolymer, further comprising a step of extracting a region, wherein at least one of the cumulative value of the score, the degree of coincidence of the arrangement of the constituents, and the juxtaposed result in the region is output and displayed. How to display locally similar sequences. 6. The method according to claim 4, wherein in the graph data, a region in which the value of the score is continuously greater than or equal to a predetermined constant value in relation to the first axis is extracted. And displaying at least one of the cumulative value of the score, the degree of coincidence of the sequences of the constituents, and the juxtaposed result in the region, Method. 7. A method for extracting a locally similar sequence of a biopolymer comprising a sequence of a plurality of constituent elements, the method comprising the step of comparing the arrangement of the constituent elements of at least two of the biopolymers with each other. , A step of obtaining a score for evaluating information regarding the similarity of sequences including a mismatch, a gap introduction, and a gap deletion, and the sequences of the constituent elements of a plurality of the biopolymers are juxtaposed and compared based on the score. In the method for extracting a locally similar biopolymer sequence, the first axis has an element number indicating the sequence of the arrangement of the constituent elements of the one of the aligned biopolymers, and the second axis A predetermined parameter obtained from the score between the juxtaposed constituents on the axis of, and converting the juxtaposed result into graph data; Steps and extraction method biopolymer local similarity sequence, characterized by having a step, the extracting a portion having a predetermined similarity from the result the juxtaposed to display multi-dimensional display the data as a graph. 8. The method according to claim 7, further comprising a step of performing a smoothing process on the graph data. 9. The method according to claim 7 or 8, wherein the predetermined parameter is a cumulative value of the scores from the predetermined element number to each of the element numbers. A method for extracting locally similar sequences of a polymer. 10. The method according to claim 7 or 8, wherein the predetermined parameter is a score value at each of the element numbers. 11. The method according to claim 9, wherein the gradient of the graph data is continuous with respect to the first axis, and the value of the second axis is equal to or larger than a predetermined constant value. A method for extracting a locally similar biopolymer sequence, further comprising the step of extracting a region. 12. The method according to claim 10, wherein a region in which the score value is continuously greater than or equal to a predetermined constant value in relation to the first axis is extracted from the graph data. A method for extracting a locally similar sequence of biopolymer, further comprising: 13. A display device of a locally similar array of biopolymers comprising an array of a plurality of constituent elements, wherein the array of the constituent elements of at least two biopolymers is compared,
Means for obtaining a score for evaluating information regarding the similarity of sequences including coincidence, disagreement, gap introduction, gap deletion between the constituent elements, and arrangement of the constituent elements of the plurality of biopolymers based on the score In a display device of a biopolymer locally similar sequence, which has at least a means for juxtaposing and comparing, a means for obtaining a predetermined parameter from a score between the juxtaposed constituent elements, and a juxtaposed first axis. One of the biopolymers is an element number indicating the order of arrangement of the constituent elements, and the second axis is the predetermined parameter, and the hand for converting the juxtaposed results into graph data, and the graph data A method for displaying a locally similar biopolymer array, comprising: a display means for displaying a multi-dimensional graph. 14. The display device according to claim 13, further comprising means for smoothing the graph data. 15. The apparatus according to claim 13 or 14, wherein the predetermined parameter is a cumulative value of the scores from the predetermined element number to each of the element numbers. A display device for locally similar sequences of polymer. 16. A display device for a biopolymer locally similar array according to claim 13 or 14, wherein the predetermined parameter is a score value at each of the element numbers. 17. The apparatus according to claim 15, wherein the gradient of the graph data is continuous with respect to the first axis, and the value of the second axis is equal to or larger than a predetermined constant value. A biopolymer, further comprising means for extracting a region, wherein at least one of the cumulative value of the score, the degree of coincidence of the arrangement of the constituent elements, and the juxtaposed result in the region is output and displayed. Display device of local similar array. 18. The device according to claim 16, wherein a means for extracting a region in which the score value is continuously greater than or equal to a predetermined constant value in the graph data in association with the first axis. And displaying at least one of the cumulative value of the score, the degree of coincidence of the sequences of the constituents, and the juxtaposed result in the region, apparatus. 19. A device for extracting a locally similar array of biopolymers, which comprises an array of a plurality of constituent elements, wherein the arrangement of the constituent elements of at least two biopolymers is compared,
Means for obtaining a score for evaluating information regarding the similarity of sequences including coincidence, disagreement, gap introduction, gap deletion between the constituent elements, and arrangement of the constituent elements of the plurality of biopolymers based on the score In a device for extracting a biopolymer local similar sequence, which comprises at least a means for juxtaposing and comparing, and a means for obtaining a predetermined parameter from a score between the juxtaposed constituent elements, and a juxtaposed first axis. One of the biopolymers is an element number indicating the order of arrangement of the constituent elements, and the second axis is the predetermined parameter, and the hand for converting the juxtaposed results into graph data, and the graph data A living body height component, comprising: a display unit that multidimensionally displays as a graph, and a unit that extracts a portion having a predetermined degree of similarity from the juxtaposed results. Extractor local similar sequences. 20. The apparatus according to claim 19, further comprising means for performing a smoothing process on the graph data, the apparatus for extracting locally similar biopolymer sequences. 21. The apparatus according to claim 19 or 20, wherein the predetermined parameter is a cumulative value of the scores from the predetermined element number to each of the element numbers. Device for extracting locally similar sequences of body macromolecules. 22. The apparatus for extracting a biopolymer locally similar sequence according to claim 19 or 20, wherein the predetermined parameter is a score value at each of the element numbers. 23. The apparatus according to claim 21, wherein the gradient of the graph data is continuous with respect to the first axis, and the value of the second axis is equal to or larger than a predetermined constant value. An apparatus for extracting a biopolymer locally similar sequence, further comprising means for extracting a region. 24. The apparatus according to claim 21, wherein the gradient of the graph data is continuous with respect to the first axis, and the value of the second axis is equal to or larger than a predetermined constant value. Extracting from the juxtaposed result, further comprising means for extracting a region, further comprising means for obtaining a cumulative value of the score in the region and means for obtaining a degree of coincidence of the arrangement of the constituent elements in the region And a display unit for displaying at least one of the accumulated region, the cumulative value of the score, and the degree of coincidence. 25. The apparatus according to claim 22, wherein a means for extracting a region in the graph data, in which the score value is continuously greater than or equal to a predetermined constant value, in association with the first axis. An apparatus for extracting a locally similar biopolymer sequence, further comprising: 26. The apparatus according to claim 22, wherein means for extracting a region in the graph data, in which the score value is continuously greater than or equal to a predetermined constant value, in association with the first axis. Further comprising: means for obtaining a cumulative value of the score in the region, and means for obtaining a degree of coincidence of the arrangement of the constituent elements in the region, the region extracted from the juxtaposed results, A display unit for displaying at least one of the cumulative value of the score and the degree of coincidence, and a device for extracting a locally similar biopolymer sequence.