JP3793814B2 - Protein information processing apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a protein information processing apparatus used for discriminating a single transmembrane region-containing protein, and particularly relates to improving discrimination accuracy.
[0002]
[Prior art]
The functions of many proteins in the body are governed by modifications by sugar chains and lipids, and research on the identification and classification of proteins involved in these modifications (proteins that cause modification of other proteins) has been vigorously conducted. It has been broken. In particular, glycosyltransferase proteins have attracted attention from a medical viewpoint as one of the most important proteins involved in sugar chain modification. There is an urgent need to develop bioinformatics technology that uses computers to find new proteins involved in modification from a vast amount of human genome information.
[0003]
Conventionally, in order to find a new glycosyltransferase protein by a computer, a motif search technique is applied, and this motif search technique is performed by paying attention to the sequence on the C-terminal side of the protein.
[0004]
For example, Non-Patent Document 1 below discloses the latest example in which a glycosyltransferase motif on the C-terminal side is searched by a homology analysis algorithm BLAST and identified as a human glycosyltransferase. Non-Patent Document 2 discloses an example of C-terminal glycosyltransferase motif extraction using a Pfam motif database.
[0005]
[Non-Patent Document 1]
Gotoh M, Sato T, Akashima T, Iwasaki H, Kameyama A, Mochizuki H, Yada T, Inaba N, Zhang Y, Kikuchi N, Kwon YD, Togayachi A, Kudo T, Nishihara S, Watanabe H, Kimata K, Narimatsu H .: "Enzymatic synthesis of chondroitin with a novel chondroitin sulfate N-acetylgalactosaminyltransferase which tansfers N-acetylgalactosamine to glucuronic acid in initiation andelongation of chondroitin sulfate synthesis.", The Journal of Biological Chemistry Papers in Press, by The American Society for Biochemistry and Molecular Biology, Inc. August 2002, as Manuscript M203619200.
[Non-Patent Document 2]
Garinot-Schneider C, Lellouch AC, Geremia RA .: "Identification of essential amino acid residues in the Sinorhizobium meliloti glucosyltransferase ExoM.", The Journal of Biological Chemistry, by The American Society for Biochemistry and Molecular Biology, Inc., October 2000, Vol. 275, pp. 31407-31413.
[0006]
[Problems to be solved by the invention]
However, the C terminal motif search is low in accuracy and can only narrow down glycosyltransferase protein candidates to some extent. Therefore, it is required to be able to determine and predict whether or not a certain protein is a glycosyltransferase protein with higher accuracy.
[0007]
Here, when it sees about O-glycan type | mold glycosyltransferase protein, all of this kind of protein is a 1 time transmembrane region containing protein which exists on a Golgi apparatus membrane to the extent already known. In contrast, type II membrane protein localized on the cell membrane is also a single transmembrane region-containing protein, and the topology of both is similar. A protein having a signal peptide region also has a topology similar to that of a glycosyltransferase protein. Since there exists a protein having a similar topology in this way, it is not easy to obtain a glycosyltransferase protein with high accuracy by a conventional method of performing a motif search based on sequence homology for genomic information.
[0008]
In the present specification, the single transmembrane region-containing protein is typically a single transmembrane membrane protein, a type II membrane protein, and a protein similar to the single transmembrane membrane protein. Including. The glycosyltransferase protein, cell membrane localized type II membrane protein, and protein having a signal peptide region are included in the single transmembrane region-containing protein.
[0009]
In addition, the glycosyltransferase protein on the Golgi apparatus has been described above. However, from the viewpoint of bioinformatics, not only the glycosyltransferase protein present on the Golgi apparatus but also the cell membrane, vesicle membrane, nuclear membrane, mitochondria, etc. Highly accurate discrimination of any other existing single transmembrane domain-containing protein is also desired.
[0010]
The present invention has been made under the above-described background, and an object thereof is to provide a technique that makes it possible to accurately discriminate a target transmembrane region-containing protein.
[0011]
[Means for Solving the Problems]
In order to achieve the above object, the present inventor paid attention to the physicochemical properties of the transmembrane region of the single transmembrane region-containing protein. In the single transmembrane region-containing protein, the highly hydrophobic region on the N-terminal side corresponds to the transmembrane region. Discriminant reference score information for discriminating the hydrophobic region of the target protein and the control protein by analyzing the position-specific amino acid frequency of the hydrophobic region of the N-terminal with respect to the known target protein and the known control protein can get. When this discrimination reference score information is used, the target and the control are discriminated with high accuracy in the self-consistency test using the data of known proteins. This means that the candidate protein can also be discriminated with high accuracy by the discrimination criterion score information. The apparatus of the present invention stores the above-described discrimination score reference information, and obtains the discrimination score of the candidate protein from this discrimination score reference information. Therefore, it is possible to determine candidate proteins with high accuracy.
[0012]
  The protein information processing apparatus of one embodiment of the present invention includes:It is a protein information processing apparatus used for determining whether a candidate protein is a target single-transmembrane region-containing protein. This protein information processing deviceHighly hydrophobic region near the N-terminus of known target and control single transmembrane domain-containing proteinsFor each position of the amino acid residue unit ofA discrimination criterion information storage unit that stores discrimination criterion score information based on position-specific amino acid appearance frequency analysis, a candidate sequence acquisition unit that acquires amino acid sequence information of candidate proteins to be discriminated, and a storage in the discrimination criterion information storage unit And a score calculation unit that calculates the discrimination score of the highly hydrophobic region near the N-terminus of the candidate protein with reference to the discriminated discrimination score information. According to this aspect, as described above, highly accurate discrimination is possible by using the discrimination reference score information obtained from the amino acid appearance frequency analysis of the highly hydrophobic region.
[0013]
  aboveThe discriminant criterion score information stored in the discriminant criterion information storage unit includes a plurality of known values for each of the target and the control.With reference to a single transmembrane domain-containing protein,The area around the reference position when the highest hydrophobic position is set as the reference position.For each position of the amino acid residue unitIt is made based on frequency analysis to determine the appearance frequency of amino acids. By using the highest hydrophobic position as the reference position, the target protein can be discriminated with higher accuracy.
[0014]
  Preferably, the discrimination reference score information stored in the discrimination reference information storage unit is a peripheral region of the reference position.Depending on the type of amino acid at each position of the amino acid residue unitInformation indicating a score, wherein the score calculation unit is a candidate protein obtained from the discrimination reference score informationOf amino acid residuesA discrimination score for the highly hydrophobic region is obtained from the score at each position. In the present invention, as the reference position of the determination reference score information is set to the highest hydrophobic position, the reference position of the candidate protein to be processed is also adjusted to the highest hydrophobic position in order to improve the determination accuracy.
[0015]
Preferably, the target single transmembrane region-containing protein is a protein localized on the Golgi apparatus membrane, and the control single transmembrane region-containing protein is a protein localized on the cell membrane.
[0016]
Preferably, the target single transmembrane region-containing protein is a protein localized on the Golgi apparatus membrane, and the control single transmembrane region-containing protein is a protein containing a signal peptide.
[0017]
Preferably, the discrimination criterion information storage unit stores a plurality of types of discrimination criterion score information respectively corresponding to a plurality of types of control proteins in order to discriminate a target protein from a plurality of types of control proteins. Combination discriminating means for discriminating whether or not the candidate protein is a candidate for the target protein from a combination of a plurality of discriminant scores obtained from plural types of discriminant reference score information is provided. For example, the target protein is a protein localized on the Golgi apparatus membrane, and the plurality of types of control proteins are proteins including a protein localized on the cell membrane and a signal peptide. In this example, the protein on the Golgi apparatus membrane is discriminated from the two control proteins by the combination determination using the above-described plural types of discrimination reference score information. Thus, according to the present invention, the target protein can be discriminated from a plurality of control proteins.
[0018]
  The protein information processing apparatus according to another aspect of the present invention is a highly hydrophobic region near the N-terminal of the first and second types of known single-transmembrane region-containing proteins.For each position of the amino acid residue unit ofA discrimination criterion information storage unit that stores discrimination criterion score information based on position-specific amino acid appearance frequency analysis, a candidate sequence acquisition unit that acquires amino acid sequence information of candidate proteins to be discriminated, and a storage in the discrimination criterion information storage unit And a score calculation unit that calculates the discrimination score of the highly hydrophobic region near the N-terminus of the candidate protein with reference to the discriminated discrimination score information. For example, the first type corresponds to the target, and the second type corresponds to the control. Regardless of the target and control, multiple types of single transmembrane region-containing proteins may be identified.The discrimination criterion score information stored in the discrimination criterion information storage unit refers to the plurality of single transmembrane region-containing proteins known for each of the first type and the second type, and sets the highest hydrophobic position as the reference position. This is based on frequency analysis for determining the frequency of appearance of amino acids at each position of amino acid residue units in the region around the reference position.
[0019]
  Another aspect of the present invention is a discriminant reference score information creation method for generating discriminant reference score information used for discriminating whether or not a candidate protein is a target single transmembrane region-containing protein. This methodAcquire amino acid sequence information of known single-transmembrane region-containing proteins of target and control,A highly hydrophobic region in the vicinity of the N-terminal of a known single-transmembrane region-containing protein of the target and control is determined, and the highly hydrophobic region of each of the target protein and the control protein is determined.For each position of the amino acid residue unit ofAnalyzing the frequency of position-specific amino acids, a score representing the ratio of the position-specific amino acid frequency of both proteinsAs discrimination standard score informationAsk. Thereby, advantageous discrimination reference score information for discriminating whether or not the candidate protein is a target single transmembrane region-containing protein is obtained.
[0020]
  Another aspect of the present invention is a computer-readable recording medium that records discrimination criterion score information used to discriminate whether or not a candidate protein is a target single transmembrane region-containing protein. This recording medium obtains a highly hydrophobic region near the N-terminal of a target and control known single transmembrane region-containing protein, and for each of the target protein and the control protein, the highly hydrophobic regionFor each position of the amino acid residue unit ofThe position-specific amino acid appearance frequency is analyzed, and a score representing the ratio of the position-specific amino acid appearance frequencies of both proteins is obtained for each position and each amino acid is stored. Thereby, advantageous discrimination reference score information for discriminating whether or not the candidate protein is a target single transmembrane region-containing protein is obtained.
[0021]
The present invention is not limited to the above embodiment. Another aspect of the present invention is a protein information processing method using, for example, bioinformatics, and a computer program and a recording medium for realizing such a method.
[0022]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0023]
First, the protein information processing method of the present embodiment will be described, and then a suitable information processing apparatus for realizing the method will be described.
[0024]
The information processing of the present invention is used to determine whether a candidate protein is a target single transmembrane region-containing protein. In this information processing, the score of a candidate protein is obtained using discrimination reference score information prepared in advance.
[0025]
As described above, in the present specification, a single transmembrane region-containing protein (protein having a single transmembrane region) is typically a single transmembrane membrane protein, and a type II membrane protein. Yes, and includes proteins similar to single transmembrane membrane proteins. The glycosyltransferase protein, cell membrane localized type II membrane protein, and protein having a signal peptide region described below are included in the single transmembrane region-containing protein. Other type II membrane proteins and other similar proteins are also included in the single transmembrane region-containing protein.
[0026]
As a feature of the present invention, the discrimination criterion score information is created based on a position-specific amino acid appearance frequency analysis of a known single-transmembrane region-containing protein of the target and control, and in particular, a highly hydrophobic region near the N-terminus. It is made based on frequency analysis. Then, the discrimination score of the highly hydrophobic region near the N-terminus of the candidate protein is calculated using the discrimination reference score information.
[0027]
In the present embodiment, the target protein is a glycosyltransferase protein on the Golgi apparatus membrane, and the control protein is a type II membrane protein on the cell membrane. First, a method of creating discrimination criterion score information to be prepared in advance will be described in detail, and then a discrimination method by score calculation using the discrimination criterion score will be described.
[0028]
FIG. 1 shows a glycosyltransferase protein, which has a sugar chain modifying function. Glycosyltransferase protein is a type II membrane protein localized in the Golgi apparatus membrane and has one transmembrane region. The transmembrane region is a transmembrane helix. As shown in the figure, in the glycosyltransferase protein, the C-terminal region of the transmembrane region is long, whereas the N-terminal region is very short. That is, the transmembrane region exists near the N-terminus. Moreover, all the transmembrane regions exist in the first exon.
[0029]
A similar topology to the glycosyltransferase protein described above is also observed for type II membrane proteins on the cell membrane. That is, also in the type II membrane protein on the cell membrane, one transmembrane region exists near the N-terminal. Since there exists a protein having a similar topology in this way, it is not easy to obtain a glycosyltransferase protein with high accuracy by a conventional method of performing a motif search based on sequence homology for genomic information. Therefore, the new method of the present embodiment extracts the characteristics of glycosyltransferase protein from the comparison of physical properties of the transmembrane region, and applies the extraction result to discrimination prediction.
[0030]
This embodiment uses a known glycosyltransferase protein data set (target) and a known cell membrane localized type II membrane protein data set (control). Human known protein data is used. Such data can be obtained from a database. In the specific example of the present embodiment, the number of entries in the former is 126, and the number of entries in the latter is 48.
[0031]
Then, as shown in FIG. 2, a hydrophobic profile is created from each data of the above data set. In the transmembrane region, the hydrophobicity tends to be large. A hydrophobic profile is required to perform treatment based on the physicochemical properties of the transmembrane region using this fact. FIG. 2 shows the profile of some proteins in the data set, but it will be appreciated that profiles of other proteins may be determined as well.
[0032]
In the profile of FIG. 2, the horizontal axis is the position of each amino acid near the N-terminus on the amino acid sequence, and the vertical axis is the average hydrophobicity value of the amino acids around each position. In all glycosyltransferase proteins, the transmembrane region is contained within 100 residues from the N-terminus. Therefore, a profile of 100 residues from the N-terminus is created. The average hydrophobicity value on the vertical axis is determined as follows.
[0033]
FIG. 3 shows a formula for calculating the average hydrophobicity value. As shown, the average hydrophobicity value at a position is the average of the hydrophobicity values of amino acids in a predetermined range before and after that position. In the example shown in the figure, the window size is 23, and the average hydrophobicity value is obtained from the hydrophobicity values of 23 amino acids centered on the target position (Note that FIG. 2 shows the case where the window size is set to 11). Although profile data is shown, other data, score matrix, etc. of this embodiment are those when the window size for calculating the average hydrophobicity value is set to 23).
[0034]
Referring to FIG. 2 again, each protein profile has a highly hydrophobic region near the N-terminus, and this highly hydrophobic region corresponds to a transmembrane region. The sequence of this highly hydrophobic region is used in the following processing.
[0035]
Next, referring to FIG. 4, hydrophobic alignment is performed on the hydrophobic profiles of all entries obtained above. In the hydrophobic alignment, the highest hydrophobic position is set as the reference position. The highest hydrophobic position is the position where the hydrophobicity is highest in each profile. All profiles are aligned, i.e., aligned in the alignment direction, so that the highest hydrophobic positions coincide.
[0036]
The hydrophobic alignment described above is performed in order to realize frequency analysis based on the highest hydrophobic position, which is one of the features of the present invention. By setting the highest hydrophobic position as a reference, characteristics of amino acid appearance frequency in the highly hydrophobic region can be clearly obtained by frequency analysis.
[0037]
FIG. 5 shows a graph of the reliability and average of the profile group after the alignment of FIG. As illustrated, the reliability is low at a position far from the reference position toward the N-terminal. This is because the number of samples decreases. The reason why the number of samples decreases is that a sample cannot be obtained from a protein having a shorter number of residues on the N-terminal side than the reference position (the highest hydrophobic position).
[0038]
The reliability and average values in FIG. 5 are used to appropriately set the target region for position-specific amino acid appearance frequency analysis. A region around the reference position and having relatively high reliability and relatively high hydrophobicity is set as a frequency analysis target. In the present embodiment, the residue at the reference position and the 20 residues before and after are set as frequency analysis targets.
[0039]
Next, position-specific amino acid appearance frequency analysis is performed on the above-described region. The frequency of appearance of each position is determined for each of all types of amino acids. Each position is defined by a distance represented by the number of amino acid residues from the aforementioned reference position. Data of the target glycosyltransferase protein and the control cell membrane localized type II membrane protein are analyzed individually.
[0040]
FIG. 6 shows the frequency analysis results of four types of amino acids. Of course, the analysis results of the remaining amino acids can be obtained in the same manner.
[0041]
In FIG. 6, the horizontal axis is the position on the amino acid sequence, and the vertical axis is the appearance frequency. In each graph, the frequency of appearance of targets and controls is superimposed. In the part marked with a circle, the frequency difference between the target and the control is relatively remarkable.
[0042]
Next, discrimination reference score information is created from the frequency analysis result of FIG. In the present embodiment, the discrimination reference score information is expressed in a matrix format. This matrix is called a position specific score matrix (PSSM).
[0043]
7 to 9 show examples of the score matrix, and FIG. 10 shows the score matrix by mathematical expressions.
[0044]
In FIG. 10, i is an amino acid residue, and P is a position on the sequence. As described above, the position is represented by the number of residues from the reference position of the alignment. As shown in FIG. 10, the score of amino acid i and position P is expressed as a logarithm of the ratio of the target appearance frequency to the control appearance frequency. Since the logarithm is used, the score becomes positive when the target appearance frequency is greater than the control appearance frequency. Conversely, when the target appearance frequency is smaller than the control appearance frequency, the score becomes a negative value. And when both are equal, a score is set to 0.
[0045]
The score matrices in FIGS. 7 to 9 show the scores obtained from the equations in FIG. As shown, the score matrix has a score corresponding to the type of amino acid at each position.
[0046]
As described above, PSSM which is the discrimination reference score information of the present embodiment is obtained.
[0047]
FIG. 11 shows a series of the above-described discrimination reference score information creation processing. As shown in the figure, regarding the glycosyltransferase data set, all entries, sequence position numbers and amino acid residues are read (S10), and an average hydrophobicity value of 23 residues is calculated in order from the N-terminus (S12). The position having the highest average 23-residue hydrophobicity within the range of 100 residues or less is set as the reference point 0, and the position number is changed (S14). Then, the number of appearances is counted for each position and for each amino acid residue within a range of 20 residues before and after the reference point, and thus the appearance frequency is obtained (S16). On the other hand, the same processing is performed for the control data set (cell membrane localized type II) (S20 to S26). Then, a score is calculated for each position and each amino acid using the formula in the figure, and a score matrix is created (S30).
[0048]
Next, a method for determining whether or not a candidate protein is a glycosyltransferase protein using the score matrix obtained as described above will be described.
[0049]
First, amino acid sequence information of a candidate protein is acquired. This amino acid sequence information may be obtained from an existing protein sequence database or may be obtained from a cDNA database. Moreover, you may obtain | require a predicted amino acid sequence from the base sequence information obtained from a genome information database using various coding region prediction programs.
[0050]
In the present embodiment, a single transmembrane region-containing protein is obtained from human genome information by separate protein discrimination processing in the previous stage. The amino acid sequence data of this single transmembrane region-containing protein is obtained as data of an unknown candidate protein. In this way, an array obtained by a certain degree of sorting processing may be acquired within the scope of the present invention. However, the present invention is not limited to this, and the obtained candidate protein sequence may be any sequence obtained from human genome information.
[0051]
Then, a hydrophobic profile is created from the acquired amino acid sequence information by the same method as the score matrix creation process. This is to obtain a highly hydrophobic region near the N-terminus of the candidate protein. Here, as described above, the average hydrophobicity value at each position is obtained in the region of 100 residues near the N-terminus. Then, the most hydrophobic position and the sequence of 20 residues before and after it are determined. This is the sequence of the highly hydrophobic region that the candidate protein has near the N-terminus. However, the number of residues on the N-terminal side from the highest hydrophobic position may be 20 or less. And the discrimination | determination score of the arrangement | sequence obtained in this way is calculated with reference to a score matrix.
[0052]
FIG. 12 shows a calculation formula for the discrimination score in the present embodiment. Schematically, the discrimination score is an average of the amino acid scores around the reference position (the reference position and the front and rear 20) obtained as described above, and the amino acid score at each position is obtained from the score matrix.
[0053]
In FIG. 12, i is the type of amino acid residue, and P is the position on the sequence. As already described, the position P is represented by the number of residues from the highest hydrophobic position which is the reference position.
[0054]
Further, M is the position of the residue farthest from the reference position to the N-terminal side, and N represents the position of the residue farthest from the reference position to the opposite side. Therefore, M represents the number of residues on the N-terminal side from the reference position, and N represents the number of residues on the reverse side. N is always 20, but M depends on the candidate protein. This is because the number of residues on the N-terminal side may be 20 or less, and this point is as described above.
[0055]
As shown in FIG. 12, the discrimination score is an average of the scores of amino acid residues in the range from position M (≧ −20) to position N (= 20), and the score of each amino acid residue is calculated from the score matrix. Read. By using this calculation formula, a discrimination score for the peripheral region of the reference position, that is, the highly hydrophobic region is obtained.
[0056]
FIG. 13 shows a specific example of discriminant score calculation. In this example, M = 20, ie there are 20 residues N-terminal to the reference position.
[0057]
In the example of FIG. 13, the amino acid residue at position (-20) is K, and the corresponding score is -0.8. Similarly, a score corresponding to the amino acid residue at each position is obtained from the matrix. The average of the 41 scores thus obtained is obtained as a discrimination score.
[0058]
FIG. 14 shows the distribution of scores obtained using the score matrix. FIG. 14 shows the distribution of scores of all data included in the data set when creating the score matrix, that is, the result of the self-consistency test.
[0059]
As shown in FIG. 14, according to this embodiment, glycosyltransferase proteins can be distinguished from cell membrane localized type II membrane proteins with 100% accuracy. Therefore, even when amino acid sequence information of an unknown candidate protein is provided, it can be determined with high accuracy that the candidate protein is not a cell membrane localized type II membrane protein but a potential glycosyltransferase protein candidate.
[0060]
In FIG. 14, the discrimination score of glycosyltransferase protein is distributed in a range of −0.1 ≦ S ≦ 0.6. Therefore, in the present embodiment, the above numerical value is set as a determination threshold. Then, a discrimination score of an unknown candidate protein is calculated, and if the discrimination score is within the above numerical range, it is determined that the candidate protein is a promising candidate for a glycosyltransferase protein.
[0061]
FIG. 15 shows a series of the above-described determination process. The amino acid sequence data of the unknown candidate protein, that is, the sequence position number and amino acid residue are read (S40). In the present embodiment, as described above, a single transmembrane region-containing protein is obtained from human genome information by separate discrimination processing in the previous stage. The amino acid sequence data of this single transmembrane region-containing protein is acquired in S40 as data of an unknown candidate protein. When the sequence data is acquired in S40, the average hydrophobicity value of 23 residues is calculated in order from the N-terminus using the acquired sequence data (S42), and the 23 residues within the range of 100 residues or less from the N-terminus The point having the highest average hydrophobicity value is set as the reference point 0, and the position number is changed (S44). The score matrix is read, and the scores for each position of the appearing amino acids are added within the range of 20 residues before and after the reference point, and the average value is calculated (S46). However, as illustrated, the number of residues on the N-terminal side of the reference point may be 20 or less (M ≧ −20). If the calculated average score is −0.1 or more and 0.6 or less (S48), it is determined that the processed data is a strong candidate for glycosyltransferase (S50).
[0062]
The protein information processing method of the present embodiment has been described above. In the above embodiment, the target protein is a glycosyltransferase protein and the control protein is a cell membrane localized type II membrane protein, but the present invention is not limited to this. For example, the control protein may be a protein containing a signal peptide. Further, the target protein may be other proteins localized in cell membrane, vesicle membrane, nuclear membrane, mitochondria and the like.
[0063]
FIG. 16 shows a protein including a glycosyltransferase protein and a signal peptide. Both topologies are similar. That is, both proteins have a transmembrane region near the N-terminus. However, as shown in the figure, in the protein containing the signal peptide, the protein is cleaved and the C-terminal region is separated from the transmembrane region. On the other hand, in the glycosyltransferase protein, the C-terminal region is not separated from the transmembrane region. That is, glycosyltransferase protein is a protein containing a signal anchor.
[0064]
Discrimination of a protein including a glycosyltransferase protein and a signal peptide can also be performed in the same manner as in the above embodiment.
[0065]
17 to 19 show score matrices when the target protein is a glycosyltransferase protein and the control protein is a protein containing a signal peptide. The data set of glycosyltransferase protein used for matrix creation is the same as that in the above embodiment, and the number of entries is 126. The number of entries of the protein including the signal peptide is 416, and these are obtained from the database.
[0066]
The score matrix creation method of FIGS. 17 to 19 may be the same as that of the above-described embodiment except that the control data sets are different. The discriminant score calculation method using the score matrix is the same as that in the above embodiment.
[0067]
As a result of the self-consistency test using the above score matrix, the discrimination accuracy was about 90%. That is, although the accuracy is reduced as compared with the case where the control protein is a cell membrane type II membrane protein, it can still be discriminated with considerably high accuracy.
[0068]
Next, the combination determination of this embodiment will be described. This combination discrimination makes it possible to discriminate a target protein from a plurality of types of control proteins. In the present embodiment, the plurality of types of control proteins are proteins including the cell membrane-localized type II membrane protein and signal peptide used in the above description, and the target protein is a glycosyltransferase protein.
[0069]
FIG. 20 shows a score distribution obtained using the two score matrices described in the above embodiment.
[0070]
The horizontal axis is a discrimination score calculated using a score matrix for discriminating glycosyltransferase proteins from cell membrane localized type II membrane proteins. On the other hand, the vertical axis represents a discrimination score calculated using a score matrix for discriminating a protein containing a signal peptide from a glycosyltransferase protein.
[0071]
In FIG. 20, two discrimination scores calculated from known protein data are plotted. The known data is data used for matrix preparation in the above description, and is data of proteins including glycosyltransferase protein, cell membrane localized type II membrane protein and signal peptide.
[0072]
As shown in the figure, by using both matrices, glycosyltransferase protein can be distinguished from cell membrane localized type II membrane protein and signal peptide protein with high accuracy. From FIG. 20, the appropriate determination threshold parameter range is −0.1 ≦ S ≦ 0.6 for the discrimination score related to the cell membrane localized type II membrane protein, as described above. Moreover, regarding the score using the matrix for the protein containing the signal peptide, the determination threshold parameter range is 0 ≦ S ≦ 0.6.
[0073]
The above is the examination result of the score distribution obtained from known proteins. Discrimination for an unknown candidate protein is performed as follows. Two discriminant scores for the candidate protein are determined using the two score matrices. When these two discriminant scores satisfy the conditions of the above two determination threshold parameters, the candidate protein is not a cell membrane type II membrane protein, not a protein containing a signal peptide, and therefore may be a glycosyltransferase protein. It is determined that the nature is high.
[0074]
FIG. 21 shows a series of the above-described combination determination process. The amino acid sequence data of the unknown candidate protein, that is, the sequence position number and amino acid residue are read (60), and the 23-residue average hydrophobicity value is calculated in order from the N-terminus (S62). The point having the highest average 23-residue hydrophobicity within the range below the group is set as the reference point 0, and the position number is changed (S64). The score matrix is read, and the score for each position of the appearing amino acid is added within the range of 20 residues before and after the reference point, and the average value is calculated (S66). Discriminant score S (Glyco / Plasma) is obtained using the score matrix for cell membrane localized type II membrane protein, and discriminant score S (Glyco / SignalPeptide) is obtained using the score matrix for protein containing signal peptide. Is required. If −0.1 ≦ S (Glyco / Plasma) ≦ 0.6 and 0 ≦ S (Glyco / SignalPeptide) ≦ 0.6 (S68), the unknown candidate protein is a potential candidate for glycosyltransferase. (S70).
[0075]
The combination determination method of the present embodiment has been described above. Next, the protein information processing apparatus of the present embodiment will be described. Also in this apparatus, it is assumed that the target protein is a glycosyltransferase protein and the control protein is a cell membrane localized type II membrane protein.
[0076]
FIG. 22 shows the protein information processing apparatus of the present embodiment. The information processing apparatus 10 in FIG. 22 is configured by a computer. A general-purpose computer may be used, or a dedicated computer may be used. The information processing apparatus 10 includes a basic computer configuration such as a CPU, RAM, and ROM. Then, the information processing apparatus 10 executes a program for realizing the protein information processing of the present invention. This program is installed on the computer.
[0077]
As shown in FIG. 22, the protein information processing apparatus 10 is provided with an input device 12 and an output device 14. The input device 12 is, for example, a keyboard and a pointing device. The output device 14 is a display and a printer, for example.
[0078]
The information processing apparatus 10 in FIG. 22 is a single computer. However, the function of the information processing apparatus 10 may be distributed to a plurality of computers.
[0079]
Further, the information processing apparatus 10 in FIG. 22 may be connected to another computer through communication such as a network. In this case, the communication function of the information processing apparatus 10 can function as the input device 12 and the output device 14.
[0080]
The information processing apparatus 10 may be connected to another computer via the Internet. In this case, the WWW server can function as the input device 12 and the output device 14.
[0081]
Furthermore, the reading function can function as the input device 12 when data necessary for information processing is read from the recording medium. Further, when the processing result is written to the recording medium, the writing function can function as the output device 14.
[0082]
As shown in FIG. 22, the information processing apparatus 10 is configured to process the protein information by using a discrimination criterion information storage unit 20, a candidate sequence acquisition unit 22, a hydrophobic profile creation unit 24, and a highly hydrophobic region specification unit. 26, a score calculation unit 28, a determination unit 30, and an output processing unit 32.
[0083]
The discrimination criterion information storage unit 20 stores discrimination criterion score information. As described above, the discrimination reference score information is a score matrix in the present embodiment. Candidate sequence acquisition unit 22 acquires amino acid sequence information of a candidate protein to be discriminated. This information is input from the input device 12 and acquired by the candidate sequence acquisition unit 22. As is clear from the above description of the information processing method, the entire protein sequence may not be acquired. At least an appropriate range of sequence data in the vicinity of the N-terminal may be acquired. In the present embodiment, as described above, a single transmembrane domain-containing protein (unknown as to whether it is a glycosyltransferase, etc.) obtained separately using a protein discrimination algorithm is acquired. The invention is not limited to this, and data before performing such sorting processing may be acquired, or data after other sorting processing may be acquired.
[0084]
The hydrophobic profile creation unit 24 creates a hydrophobic profile from the sequence data of candidate proteins. As already described, in this embodiment, a profile of the average hydrophobicity value at each position is created. Then, the highly hydrophobic region specifying unit 26 specifies a highly hydrophobic region near the N-terminus of the candidate protein from the hydrophobic profile. Here, the residue at the position where the average hydrophobicity value is the highest and the 20 residues before and after the residue are specified as the highly hydrophobic region.
[0085]
The score calculation unit 28 obtains a discrimination score for the identified highly hydrophobic region. In this process, the score matrix in the discrimination criterion information storage unit 20 is referred to. A score corresponding to the amino acid at each position is obtained from the score matrix, and an average of the scores is calculated.
[0086]
The determination unit 30 refers to a predetermined determination threshold parameter to determine whether the candidate protein is a glycosyltransferase protein candidate. As already described, the discrimination threshold parameter satisfies −0.1 ≦ S ≦ 0.6 and is stored in the apparatus.
[0087]
The output processing unit 32 causes the output device 14 to output the discrimination score calculated by the score calculation unit 28 together with the discrimination result of the discrimination unit 30. Such information is displayed on a display and printed by a printer.
[0088]
Next, FIG. 23 shows another protein information processing apparatus of the present embodiment. This processing apparatus 100 supports the combination determination described above. Hereinafter, description of the same components as those in FIG. 22 will be appropriately omitted.
[0089]
In the configuration of FIG. 23, the discrimination criterion information storage unit 20 stores a score matrix similar to that of FIG. However, the discrimination reference storage unit 20 stores two types of score matrices when the control protein is a protein including a cell membrane type II membrane protein and a signal peptide.
[0090]
The score calculation unit 28 calculates two discrimination scores S (Glyco / Plasma) and S (Glyco / SignalPeptide) by using the two score matrices individually. The combination determination unit 40 uses a determination threshold parameter set in advance for each of these two determination scores. When the two discrimination scores both satisfy the setting condition, the combination discrimination unit 40 determines that the candidate protein is a glycosyltransferase protein candidate. The output processing unit 32 causes the output device 14 to output two discrimination scores together with the discrimination results.
[0091]
As described above, according to the present invention, a single transmembrane is performed using the discrimination criterion score information obtained from the result of the position-specific amino acid appearance frequency analysis performed on the highly hydrophobic region near the N-terminus. Information for discriminating the region-containing protein is obtained, thereby enabling discrimination with high accuracy.
[0092]
The information processing of the present invention uses discrimination reference score information obtained from the result of position-specific amino acid appearance frequency analysis performed with the highest hydrophobic position in the highly hydrophobic region near the N-terminal as the reference position. Accordingly, the discrimination reference score information obtained by more accurately extracting the characteristics of the highly hydrophobic region (transmembrane region) is used, and thus information for discrimination can be obtained with high accuracy.
[0093]
Further, the information processing of the present invention uses the highest hydrophobic position as a reference position when obtaining a discrimination score of a candidate protein. Thereby, information for discrimination can be obtained with high accuracy.
[0094]
The information processing of the present invention is advantageously used to discriminate proteins localized on the Golgi apparatus membrane from proteins localized on the cell membrane, as described in the above embodiment. In addition, the information processing of the present invention is advantageously used for discriminating proteins localized on the Golgi apparatus membrane from proteins including a signal peptide.
[0095]
The target and control proteins are not limited to the above, and can be changed within the scope of the present invention. In principle, mitochondrial, vesicular and nuclear membrane proteins are also applicable to the information processing of the present invention as target proteins or control proteins.
[0096]
The information processing according to the present invention performs the above-described combination determination. That is, the information processing of the present invention refers to a plurality of types of discriminant reference score information respectively corresponding to a plurality of types of control proteins, obtains a plurality of discriminant scores from candidate proteins, and discriminates based on the combination thereof. Thereby, the target protein is discriminated from a plurality of types of control proteins. The accuracy of discrimination can be improved. In the above embodiment, the score matrix is mainly used for determining whether or not the candidate protein is the target protein. However, the same score matrix can be used to determine whether a candidate protein is a control protein. This is also clear from the fact that the discrimination accuracy between glycosyltransferase protein and cell membrane localized type II membrane protein was 100%. Both determinations may be made at the same time. Such a configuration is also included in the scope of the present invention.
[0097]
In the above aspect, another embodiment of the present invention can be described as follows. In other words, the information processing apparatus according to this aspect is configured to determine a discrimination criterion score based on a position-specific amino acid appearance frequency analysis for a highly hydrophobic region near the N-terminus of the first and second types of known single-transmembrane region-containing proteins. Store information. With reference to this discrimination criterion score information, a discrimination score of the candidate protein is calculated. Whether or not the candidate protein is the first type protein may be determined, whether or not the candidate protein is the second type protein may be determined, or both of them may be performed. . For example, the first and second types of proteins are glycosyltransferase proteins and cell membrane localized type II membrane proteins. In this case, the score matrix that is the discrimination reference score information may be the same as that in the above-described embodiment. .
[0098]
Another aspect of the present invention is a method for generating the above-described suitable discrimination reference score information (score matrix in the embodiment). Thereby, highly accurate discrimination becomes possible.
[0099]
Another aspect of the present invention is a computer-readable recording medium in which advantageous discrimination reference score information obtained by the appropriate method described above is stored. Thereby, highly accurate discrimination becomes possible.
[0100]
While preferred embodiments of the invention have been described above, it will be appreciated that various modifications can be made to the embodiments and that the appended claims fall within the true spirit and scope of the invention. It is intended to cover all possible variations.
[0101]
【The invention's effect】
As described above, according to the present invention, it is possible to provide a technique that enables a single transmembrane region-containing protein such as a glycosyltransferase protein to be accurately discriminated.
[Brief description of the drawings]
FIG. 1 is a diagram showing the structure of a glycosyltransferase protein.
FIG. 2 is a diagram showing an example of a hydrophobic profile.
FIG. 3 is a diagram showing a formula for calculating an average hydrophobicity value in a hydrophobicity profile.
FIG. 4 is a diagram illustrating hydrophobic alignment.
5 is a graph showing reliability and average graphs of profile groups after the alignment shown in FIG. 4; FIG.
FIG. 6 is a diagram showing the results of a position-specific amino acid appearance frequency analysis for a highly hydrophobic region in an aligned state.
FIG. 7 is a diagram showing a position-specific score matrix obtained from a frequency analysis result.
FIG. 8 is a diagram showing a continuation of the score matrix of FIG.
FIG. 9 is a diagram showing a continuation of the score matrix of FIG.
FIG. 10 is a diagram illustrating a score matrix using mathematical formulas.
FIG. 11 is a diagram showing a series of flows for creating a score matrix.
FIG. 12 is a diagram illustrating a calculation formula for a discrimination score using a score matrix.
FIG. 13 is a diagram illustrating a specific example of discrimination score calculation processing;
FIG. 14 is a diagram showing a distribution of scores calculated from all data used for creating a score matrix.
FIG. 15 is a diagram illustrating a series of processes using a score matrix.
FIG. 16 shows a protein containing a glycosyltransferase protein and a signal peptide.
FIG. 17 is a diagram showing a score matrix when the target protein is a glycosyltransferase protein and the control protein is a protein containing a signal peptide.
18 is a diagram showing a continuation of the score matrix of FIG.
19 is a diagram showing a continuation of the score matrix of FIG.
FIG. 20 is a diagram illustrating a score distribution obtained using two score matrices in a two-dimensional format.
FIG. 21 is a diagram showing a flow of a combination determination process using two score matrices.
FIG. 22 is a block diagram showing the protein information processing apparatus of the present embodiment.
FIG. 23 is a block diagram illustrating a protein information processing apparatus according to another embodiment.
[Explanation of symbols]
10 Information processing equipment
12 Input devices
14 Output device
20 Discrimination criterion information storage unit
22 Candidate sequence acquisition unit
24 Hydrophobic profile generator
26 Highly hydrophobic region specific part
28 Score calculator
30 discriminator
32 Output processing section

Claims (28)

A protein information processing apparatus used to determine whether a candidate protein is a target single-transmembrane region-containing protein,
Discriminating to store discriminant reference score information based on position-specific amino acid appearance frequency analysis for each position of amino acid residue units in the highly hydrophobic region near the N-terminal of a target and control known transmembrane region-containing protein A reference information storage unit;
A candidate sequence acquisition unit for acquiring amino acid sequence information of a candidate protein to be distinguished;
A score calculation unit that refers to the discrimination criterion score information stored in the discrimination criterion information storage unit and calculates a discrimination score of a highly hydrophobic region near the N-terminus of the candidate protein;
Including
The discriminant criterion score information stored in the discriminant criterion information storage unit refers to a plurality of known transmembrane region-containing proteins for each of the target and the control, and when the highest hydrophobic position is set as the reference position. A protein information processing apparatus, which is made based on a frequency analysis for determining an amino acid appearance frequency for each position of an amino acid residue unit in a region around a reference position . The protein information processing apparatus according to claim 1,
Each protein amino acid residue unit position in the reference position peripheral region is determined by a distance represented by the number of amino acid residues from the reference position. The protein information processing apparatus according to claim 1 or 2,
The protein information processing apparatus, wherein the peripheral region of the reference position is a region composed of 20 residues before and after the reference position. The protein information processing apparatus according to claim 1 ,
The discrimination criterion score information stored in the discrimination criterion information storage unit is information indicating a score according to the type of amino acid at each position of amino acid residue units in the peripheral region of the reference position,
The protein information processing apparatus, wherein the score calculation unit obtains a discrimination score for a highly hydrophobic region from a score at each position in amino acid residue units of a candidate protein obtained from the discrimination reference score information. The protein information processing apparatus according to claim 4,
The protein information processing apparatus, wherein the discrimination reference score information is position-specific score matrix information in which amino acid types at each position of amino acid residue units in a peripheral region of the reference position are expressed in a matrix format . In the protein information processing apparatus according to any one of claims 1 to 5 ,
The target single transmembrane region-containing protein is a protein localized on the Golgi apparatus membrane, and the control single transmembrane region-containing protein is a protein localized on the cell membrane. Protein information processing equipment. In the protein information processing apparatus according to any one of claims 1 to 5 ,
Protein information characterized in that the target single transmembrane region-containing protein is a protein localized on the Golgi apparatus membrane, and the control single transmembrane region-containing protein is a protein containing a signal peptide. Processing equipment. In the protein information processing apparatus according to any one of claims 1 to 5 ,
The discrimination criterion information storage unit stores a plurality of types of discrimination criterion score information respectively corresponding to a plurality of types of control proteins in order to discriminate a target protein from a plurality of types of control proteins.
Furthermore, a combination discriminating means for discriminating whether or not the candidate protein is a candidate for a target protein from a combination of a plurality of discriminant scores obtained from the plurality of types of discriminant reference score information is provided. Protein information processing equipment. In the protein information processing apparatus according to any one of claims 1 to 8,
Furthermore, a known information acquisition unit that acquires amino acid sequence information of a known transmembrane region-containing protein of a target and a control,
For each of the target protein and the control protein, an appearance frequency analysis unit that analyzes a position-specific amino acid appearance frequency for each position of the amino acid residue unit of the highly hydrophobic region;
Discrimination standard score information calculation unit for obtaining a score representing the ratio of position-specific amino acid appearance frequencies of both proteins as discrimination standard score information;
A protein information processing apparatus comprising: Discriminant reference score information based on position-specific amino acid appearance frequency analysis for each position of the amino acid residue unit in the highly hydrophobic region near the N-terminal of the first and second types of known single-transmembrane region-containing proteins A discrimination criterion information storage unit for storing
A candidate sequence acquisition unit for acquiring amino acid sequence information of a candidate protein to be distinguished;
A score calculation unit that refers to the discrimination criterion score information stored in the discrimination criterion information storage unit and calculates a discrimination score of a highly hydrophobic region near the N-terminus of the candidate protein;
Including
The discriminant criterion score information stored in the discriminant criterion information storage unit refers to a plurality of single transmembrane region-containing proteins known for each of the first type and the second type, and sets the highest hydrophobic position as a reference position. A protein information processing apparatus, which is made based on a frequency analysis for determining an amino acid appearance frequency for each position of an amino acid residue unit in a region around the reference position when the reference position is set . A protein information processing method used to determine whether a candidate protein is a target single-transmembrane region-containing protein,
A step of acquiring discriminant reference score information based on position-specific amino acid appearance frequency analysis for each position of amino acid residue units in a highly hydrophobic region near the N-terminal of a known single-transmembrane region-containing protein of target and control When,
Obtaining amino acid sequence information of candidate proteins to be distinguished ;
Referring to the discrimination criterion score information acquired by the step of acquiring the discrimination criterion score information, calculating a discrimination score of a highly hydrophobic region near the N-terminal of the candidate protein ;
Including
The step of obtaining the discrimination reference score information is performed by referring to the amino acid sequences of a plurality of known single-transmembrane region-containing proteins for each of the target and the control, and surrounding the reference position when the highest hydrophobic position is set as the reference position. A protein information processing method , comprising: obtaining discrimination criterion score information created based on a frequency analysis for obtaining an amino acid appearance frequency for each position of an amino acid residue unit of a region . The protein information processing method according to claim 11,
The protein information processing method, wherein each position of the amino acid residue unit in the peripheral region of the reference position is determined by a distance represented by the number of amino acid residues from the reference position. The protein information processing method according to claim 11 or 12,
The protein information processing method, wherein the reference position peripheral region is a region composed of 20 residues before and after the reference position. The protein information processing method according to claim 11,
  The step of acquiring the discrimination reference score information is a step of acquiring information indicating a score according to the type of amino acid at each position of the amino acid residue unit in the peripheral region of the reference position,
  The step of calculating the discriminant score is a step of obtaining a discriminant score of a highly hydrophobic region from a score at each position in amino acid residue units of a candidate protein obtained from the discriminant reference score information. Method. The protein information processing method according to claim 14,
The step of acquiring the discrimination reference score information is a step of acquiring position-specific score matrix information in which the types of amino acids at each position in the amino acid residue unit in the peripheral region of the reference position are represented in a matrix format. A protein information processing method characterized by In the protein information processing method in any one of Claims 11-15,
The step of obtaining the discrimination criterion score information is a step of obtaining a plurality of types of discrimination criterion score information respectively corresponding to a plurality of types of control proteins prepared for discriminating a target protein from a plurality of types of control proteins ,
The protein further includes a step of performing a combination determination for determining whether or not the candidate protein is a candidate for a target protein from a combination of a plurality of determination scores obtained from the plurality of types of determination reference score information. Information processing method. In the protein information processing method in any one of Claims 11-16,
Furthermore, obtaining the amino acid sequence information of the known transmembrane region-containing protein of the target and control;
Analyzing the position-specific amino acid appearance frequency for each position of the amino acid residue unit of the highly hydrophobic region for each of the target protein and the control protein;
Obtaining a score representing the ratio of the position-specific amino acid appearance frequencies of both proteins as discrimination reference score information;
A protein information processing method comprising: A program that causes a computer to execute protein information processing for determining whether a candidate protein is a target single-transmembrane region-containing protein,
Reference to amino acid sequences of a plurality of known single-transmembrane region-containing proteins for each of the target and control, and a highly hydrophobic region near the N-terminus of the known single-transmembrane region-containing protein of the target and the control The discriminant reference score information created based on the position-specific amino acid appearance frequency analysis for obtaining the amino acid appearance frequency for each position in the amino acid residue unit of the region around the reference position when the highest hydrophobic position of the reference position is used as the reference position is acquired Steps,
Obtaining amino acid sequence information of candidate proteins to be distinguished ;
Referring to the discrimination criterion score information acquired by the step of acquiring the discrimination criterion score information, calculating a discrimination score of a highly hydrophobic region near the N-terminal of the candidate protein ;
A program that causes a computer to execute . The program according to claim 18, wherein
Each position of the amino acid residue unit in the peripheral region of the reference position is determined by a distance represented by the number of amino acid residues from the reference position. The program according to claim 18 or 19,
The program wherein the region around the reference position is a region consisting of 20 residues before and after the reference position. The program according to claim 18, wherein
  The step of acquiring the discrimination reference score information is a step of acquiring information indicating a score according to the type of amino acid at each position of the amino acid residue unit in the peripheral region of the reference position,
  The step of calculating the discrimination score is a step of obtaining a discrimination score of a highly hydrophobic region from a score at each position in amino acid residue units of a candidate protein obtained from the discrimination reference score information. The program according to claim 21, wherein
The step of acquiring the discrimination reference score information is a step of acquiring position-specific score matrix information in which the types of amino acids at each position in the amino acid residue unit in the peripheral region of the reference position are represented in a matrix format. A program characterized by In the program according to any one of claims 18 to 22,
The step of obtaining the discrimination criterion score information is a step of obtaining a plurality of types of discrimination criterion score information respectively corresponding to a plurality of types of control proteins prepared for discriminating a target protein from a plurality of types of control proteins ,
The program further includes a step of performing combination determination for determining whether or not the candidate protein is a target protein candidate from a combination of a plurality of determination scores obtained from the plurality of types of determination reference score information. . A computer-readable recording medium storing the program according to any one of claims 18 to 23. A discriminant reference score information creating method for creating discriminant reference score information used for discriminating whether a candidate protein is a target single transmembrane region-containing protein,
Obtaining amino acid sequence information of known single transmembrane domain-containing proteins of target and control;
Determining a highly hydrophobic region near the N-terminus of a target and control known transmembrane domain-containing protein ;
Analyzing the position-specific amino acid appearance frequency for each position of the amino acid residue unit of the highly hydrophobic region for each of the target protein and the control protein ;
Obtaining a score representing the ratio of the position-specific amino acid appearance frequencies of both proteins as discrimination reference score information;
The discriminant reference score information creation method characterized by including . A computer-readable recording medium that records discrimination criterion score information used to determine whether a candidate protein is a target single-transmembrane region-containing protein,
Finding a highly hydrophobic region near the N-terminus of the target and control known transmembrane domain-containing proteins,
For each of the target protein and the control protein , the position-specific amino acid appearance frequency for each position of the amino acid residue unit of the highly hydrophobic region is analyzed,
Find a score that represents the ratio of the frequency of position-specific amino acids in both proteins.
A recording medium storing a score for each position and amino acid for each position. The recording medium according to claim 26 , wherein
The target single transmembrane region-containing protein is a protein localized on the Golgi apparatus membrane, and the control single transmembrane region-containing protein is a protein localized on the cell membrane. recoding media. The recording medium according to claim 26 , wherein
The target single transmembrane region-containing protein is a protein localized on the Golgi apparatus membrane, and the control single transmembrane region-containing protein is a protein containing a signal peptide. .

Images