JP4638726B2 - Sample set manufacturing method, gene alignment program, and sample set - Google Patents

Description

The present invention relates to a sample set for producing a sample set in which object groups are arranged, in particular, gene groups are arranged and sample groups corresponding to the gene groups are stored in a sample container having a plurality of storage units. manufacturing method relates to a gene alignment programs and sample set.

  SiRNA library, siRNA expression vector library, shRNA expression vector library, siRNA expression DNA fragment library, shRNA expression DNA fragment library, cDNA clone library, PCR for gene detection prepared for all genes of specific organisms such as humans and mice When a large-scale sample group such as a primer / probe library is stored or manipulated, a multi-well microplate such as 96-well or 384-well is usually used. For example, when a 96-well microplate is used, about 30,000 types of siRNA targeting all human genes can be stored in about 300 microplates.

Meanwhile, as a concept for classifying and organizing genes, there is a gene ontology (Gene Ontology (GO)) (see, for example, Non-Patent Document 1).
The gene ontology defines gene attributes such as “signal transducer activity (GO: 0004871)” and “receptor activity (GO: 0004872)”, and the attribute “receptor activity” is called signal transducer activity. An inheritance relationship between attributes such as “inherit attributes” is defined. The definition of attributes and inheritance relationships between attributes are available from the Gene Ontology Consortium (http://www.geneontology.org/). The correspondence between human and mouse genes and Gene Ontology attributes is based on various databases such as the Cancer Genome Anatomy Project (http://cgap.nci.nih.gov/). It is available. From the correspondence data with Gene Ontology of the gene, for example, the human ZYX gene (NM_003461) is found to have a receptor activity, and further, using the inheritance relationship between attributes, it can be derived that the ZYX gene has a signal transducer activity at the same time. it can.

In gene ontology, the definition of each attribute and the inheritance relationship between attributes are defined for gene attributes. The inheritance relationship between attributes in this gene ontology forms an acyclic directed graph (DAG) as shown in FIG. In gene ontology, as shown in Fig. 8, genes are classified and organized by "molecular function", "biological process", and "cellular component". Yes. In each classification method, an inheritance relationship between attributes is defined, and FIG. 8 shows a part of the inheritance relationship between attributes. In FIG. 8, for each attribute, the attribute name and the attribute ID in the gene ontology are shown.
Gene Ontology Consortium, "Gene Ontology Consortium Home Page", [online], 1999, Gene Ontology Consortium, [October 25, 2004 search], Internet <URL: http://www.geneontology.org/>

Consider a case where siRNA samples targeting all human genes as described above are stored in a microplate. When samples are randomly placed on the microplate, even if only samples related to a specific function (for example, transcription factors (about 1,000 samples)) are taken out and experiments are performed, almost all microplates are removed from the freezer. Need to work out. This means that almost all microplates need to be removed from the freezer, thawed, peeled off the seal, pipetted to remove the sample while checking each sample's position, sealing the plate again, and freezing. . For this reason, if a large-scale sample group is arranged at random, the efficiency of sample management and experimental operation is reduced.

  In addition to the gene ontology, the group group to which the object belongs may constitute an acyclic directed graph. In such a case, even when the objects are grouped for each group, if the group arrangement is random, the objects belonging to the group and the objects corresponding thereto cannot be handled efficiently.

The present invention has been made in order to solve the above-described problems, and the object thereof is to efficiently handle an object, and in particular, management of a sample stored in a sample container having a plurality of storage units and an experimental operation. sample set production method for improving the efficiency, is to provide a gene alignment program and the sample set.

The present invention is directed to a sample group corresponding to a gene group, and produces a sample set in which a sample corresponding to each gene is stored in a sample container having a plurality of storage units, and the gene attribute in the gene ontology For each node, a node size is calculated by summing the number of all genes belonging to this node and all nodes below this node, and the acyclic directed graph is calculated according to the node size. The branches are identified, the nodes of the branches are identified in the order of connection , the genes are aligned by sequentially aligning the genes belonging to the identified nodes, and the sample corresponding to the aligned genes is A sample set is manufactured by storing in each storage part of the sample container. According to this, it is possible to store a sample corresponding to a gene with the same attribute or a sample corresponding to a gene with an attribute inheriting the same attribute in a storage part of a sample container having a plurality of storage parts. Become. For this reason, there is a high possibility that a sample corresponding to a gene having the same attribute or a sample corresponding to a gene having an attribute inheriting the same attribute can be stored in the same sample container. Therefore, the sample set to be used can be limited, and the efficiency of sample management and experiment operation can be improved. Also, in the same sample set, samples corresponding to genes with the same attribute and samples corresponding to genes with the same attribute can be placed at close positions, for example, operation for each column is possible, The efficiency of the experimental operation can be further improved. In addition, if a specific attribute is a root node, genes of attributes corresponding to the root node and lower nodes of the root node are aligned, and a sample set in which samples corresponding to these genes are stored in a sample container is manufactured. it can.

Further, according to the present invention, when the specific node includes an unprocessed child node in which genes are not aligned, a child node for specifying a next child node to be processed according to the node size is selected from the unprocessed child nodes. Node search processing, gene alignment processing for sequentially aligning genes belonging to this specific node when there is no unprocessed child node in which the gene is not aligned in the specific node, and next processing for the parent node of this specific node It is possible to align genes by repeating the node size-corresponding depth priority search process including the parent node specifying process specified as a target, starting with the root node as a specified node and performing the gene alignment process for all nodes. .

Furthermore, according to the present invention, duplication of the same gene can be eliminated in the alignment process. According to this, it is possible to prevent samples corresponding to the same gene from overlapping, and it is possible to align overlapping genes, for example, in a larger node size. Therefore, for example, genes can be arranged together in a larger node size.

  Furthermore, according to the present invention, the sample group includes siRNA library, siRNA expression vector library, shRNA expression vector library, siRNA expression DNA fragment library, shRNA expression DNA fragment library, cDNA clone library, PCR primer / probe library for gene detection. Any one of them can be used.

The present invention arranges genes using a computer, and the computer uses a non-circular directed graph with the attributes of genes in the gene ontology as nodes, and each node is below this node and below this node. Calculate the node size that is the sum of all the genes belonging to all nodes, identify the branch of the acyclic directed graph according to the node size, identify the nodes of this branch in the order of connection, and select the gene belonging to the identified node The genes are aligned by sequential alignment. According to this, genes having the same attribute or genes having the same attribute can be continuously arranged. Further, by setting a specific attribute as a root node, genes having attributes corresponding to the root node and lower nodes of the root node can be arranged. For this reason, when an experiment is performed on a gene with a specific attribute, it is possible to arrange a sample corresponding to the gene to be experimented so that the efficiency of the experiment operation can be improved.

  ADVANTAGE OF THE INVENTION According to this invention, the efficiency of management of the sample stored in the sample container provided with the some storage part and experiment operation can be improved.

  Hereinafter, an embodiment embodying the present invention will be described with reference to FIGS. In the present embodiment, a sample plate (sample set) is manufactured by aligning genes and storing siRNA samples corresponding to the aligned genes in each well (storage part) of a microplate as a sample container. A sample set manufacturing method, a gene alignment program, and a sample set will be described. In this specification, a microplate in which a sample is stored in each well is referred to as a “sample plate”. In this embodiment, a sample plate is manufactured by storing siRNA samples prepared for all genes of a human organism in each well of a 96-well microplate. Here, a processing procedure will be described first using a simplified example, and then a case will be described in which a sample plate of an siRNA library prepared for all human genes is manufactured.

As shown in FIG. 1, the gene alignment apparatus 20 includes a control computer 21, an input unit 22, and an output unit 23.
The control computer 21 includes control means (CPU) and storage means (RAM, ROM, etc.) not shown, and performs processing described later. By executing the gene alignment program for that purpose, the control computer 21 functions as a node size calculation unit, an alignment unit, an arrangement unit, and the like described in the claims.

  For each node, the node size calculating means calculates a node size that is the sum of the numbers of all genes belonging to this node and all nodes below this node. The alignment means starts a node size-corresponding depth priority search process including a child node search process, a gene alignment process, and a parent node specification process by specifying the root node as a parent node, and gene alignment for all nodes. Repeat until processing. In the child node search process, when there is a child node whose gene alignment process has not been processed (unvisited) for the specified parent node, the node size is the largest for the child node that has not been subjected to the gene alignment process for this parent node. A child node is selected, and the selected child node is specified as a parent node. In the gene alignment process, when there is no child node whose gene alignment process has not been processed (unvisited) for the identified parent node, genes belonging to the attribute of this node are sequentially aligned for this parent node. The parent node specifying process specifies the parent node of the processing target node of the gene alignment process when the gene alignment process is performed. Genes are aligned by the node size corresponding depth priority search process including the child node search process, the gene alignment process, and the parent node specifying process. The arrangement means arranges the genes aligned by the gene alignment process in correspondence with each well of the microplate.

  The RAM in the control computer 21 includes a stack for storing data to be described later. Further, the gene ontology data 250 read from the gene ontology data storage unit 25 is recorded in the RAM, and when a node corresponding to each attribute is visited (when a gene alignment process is performed), the RAM is stored in the node. The visit flag is recorded in association.

The input unit 22 includes a keyboard, a mouse, and the like, and is used when a gene alignment request is made by the method of the present embodiment.
The output unit 23 is connected to another device such as a display device. In the present embodiment, the gene name, plate number, well number (vertical), (horizontal) of the aligned genes are output to the display device via the output unit 23, and the gene names of the aligned genes are displayed on the display device. , Plate number, well number (vertical), (horizontal) are displayed.

  In addition, the gene alignment device 20 includes a storage device such as a hard disk including a gene ontology data storage unit 25, a gene attribute data storage unit 26, an alignment data storage unit 27, and a sample designation data storage unit (not shown).

  In the gene ontology data storage unit 25, gene ontology data 250 is recorded as shown in FIG. The gene ontology data 250 is data relating to the definition of attributes in gene ontology (GO) and the definition of inheritance relationships between attributes. In the present embodiment, the gene ontology data 250 can be downloaded from the Gene Ontology Consortium (http://www.geneontology.org/). In the present embodiment, gene ontology data 250 obtained in advance is recorded in the gene ontology data storage unit 25. In the gene ontology data 250, inheritance relationships between attributes are hierarchically shown.

  Here, in order to simplify the description of the processing procedure, a non-circular directed graph 251 showing the inheritance relationship between attributes is used in FIG. In this acyclic directed graph 251, nodes A to I corresponding to attributes in the gene ontology are configured. In addition, a line (branch) connecting each node indicates an inheritance relationship between attributes.

  As shown in FIG. 3, gene attribute data 260 indicating the correspondence between genes and attributes in gene ontology is recorded in the gene attribute data storage unit 26 for each gene. The gene attribute data 260 includes, for each gene, data related to the gene name and one or more attributes.

  Data relating to gene names is recorded in the gene name data area. The accession number in the public gene database may be recorded together with the gene name or instead of the gene name.

In the attribute data area, data relating to the attribute of this gene is recorded. When a gene belongs to a plurality of attributes, a plurality of attributes are recorded.
The gene attribute data 260 can be obtained from various databases such as the Cancer Genome Anatomy Project (http://cgap.nci.nih.gov/). In the present embodiment, gene attribute data 260 obtained in advance is recorded in the gene attribute data storage unit 26.

  As shown in FIG. 4, the alignment data storage unit 27 records alignment data 270 for each gene corresponding to the sample stored in each well of the microplate. This alignment data 270 is recorded when the genes are aligned according to the processing described later. The alignment data 270 includes data relating to order, plate number, well number (vertical), well number (horizontal), gene name, and sample name.

In the order data area, data relating to the order of the aligned genes is recorded.
In the plate number data area, data relating to the identification number of the plate storing the sample corresponding to this gene is recorded.

  In the well number (vertical) data area, data for specifying the position (vertical) of a well storing a sample corresponding to this gene is recorded. Specifically, one of 12 columns (01 to 12) is recorded.

  In the well number (horizontal) data area, data for specifying the position (horizontal) of a well storing a sample corresponding to this gene is recorded. Specifically, any of 8 columns (A to H) is recorded.

  In the gene name data area, data relating to the gene names of the aligned genes is recorded. The accession number may be recorded together with the gene name or instead of the gene name.

Data relating to the sample name of this gene is recorded in the sample name data area.
In the sample designation data storage unit, sample designation data relating to the sample name of the sample produced corresponding to each gene is recorded. The sample designation data includes data related to the gene name and the sample name. There may be a plurality of samples corresponding to genes, but in this embodiment, only one type of sample specified in advance is used for each gene. Therefore, for each gene, which sample is used is designated in advance, and sample designation data in which the gene name and the sample name are associated is recorded.

  The control computer 21 performs processing described later, records the alignment data 270 in which the genes are aligned corresponding to each well of the microplate, and outputs the alignment data 270. The siRNA corresponding to the gene specified by the gene name in the alignment data 270 is stored in each well 31 of the 96-well microplate 30.

(Processing procedure)
In the system configured as described above, genes are aligned by performing a depth-first search (node size-corresponding depth-first search process) in descending order of the node size, which will be described later, using a gene ontology. A processing procedure in the case of arranging on a microplate will be described with reference to FIG. Here, the case where there is an inheritance relationship between attributes shown in the acyclic directed graph 251 will be described with reference to FIG. In FIG. 6, an acyclic directed graph (DAG) in which an attribute in gene ontology is a node 41 and an inheritance relationship is represented by a branch 42 is shown. Here, the node 41 for the attribute A is the root node. Each node 41 also has a balloon 43 that displays the gene name associated with the attribute corresponding to this node. The gene name associated with each attribute can be acquired from the gene attribute data 260. Furthermore, a node size 44 is shown above each node. In the present embodiment, the node size 44 is defined as “the number of all genes belonging to the node and all nodes below it”. For example, for node B, the total number of genes belonging to nodes B, C, D, and E is the node size, and the node size is “6”.

The processing procedure will be described below with reference to FIG. 5 using the acyclic directed graph of FIG.
First, the control computer 21 records the root node on the stack (step S1). Here, the root node is node A, and node A is recorded on the stack.

  Next, the control computer 21 determines whether or not the stack is empty (step S2). Here, since node A is recorded in the stack and is not empty (NO in step S2), it is determined whether there is an unvisited child node (step S3). Here, the child nodes for node A are node B, node F, and node G, and no visit flag is recorded.

  Since there are unvisited child nodes in this way (because YES in step S3), the node size (node size) is calculated for each unvisited child node (step S4). Here, the “node size (node size)” is defined as “the number of all genes belonging to the node and all the nodes below it”. Specifically, first, for each unvisited child node, all lower nodes are specified based on the attribute inheritance relationship in the gene ontology data 250. Then, for each unvisited child node, the number of genes belonging to the attribute of each node and the number of genes attribute data 260 for each attribute are obtained for that node and all nodes below it. To do. Then, for each unvisited child node, the node size is calculated by summing the number of genes counted for each node and all the nodes below it.

  In FIG. 6, the size of node B is the total number of genes belonging to nodes B, C, D and E (number of genes: 6), and the size of node F is the number of genes belonging to node F (number of genes: 2). The size of the node G is the total number of genes belonging to the nodes G, H, and I (the number of genes: 4). The node size calculated in this way is shown as a node size 44 on each node shown in FIG.

Then, the control computer 21 compares the node sizes of the unvisited child nodes, selects the child node having the largest node size, and records it on the stack (step S5). Thereby, the selected child node is specified as the parent node of the next process. When there are a plurality of child nodes having the largest node size, one of them is selected. Here, since node B is the largest, node B is recorded on the stack. That is, a branch including the node B branched from the node A is specified. Then, the process returns to step S2.

  Here, since the nodes A and B are recorded in the stack and the stack is not empty (NO in step S2), there is an unvisited child node for the node B that is recorded last in the stack. Is determined (step S3). Here, nodes C and E exist as unvisited child nodes. For this reason (because it is YES in step S3), the size is calculated for an unvisited child node (step S4). Then, the node sizes of node C (number of genes: 2) and node E (number of genes: 1) are compared, and node C, which is the largest child node, is recorded on the stack (step S5). That is, a branch including the node C branched from the node B is specified. Then, the process returns to step S2.

  Here, since nodes A, B, and C are recorded in the stack and the stack is not empty (NO in step S2), an unvisited child node is the last node C recorded in the stack. It is determined whether it exists (step S3). Here, node D exists as an unvisited child node. For this reason (because it is YES in step S3), the size of the unvisited child node is calculated (step S4), and the largest child node, node D, is recorded on the stack (step S5). In this case, when there is one unvisited child node, the unvisited child node may be recorded on the stack without calculating the node size. Then, the process returns to step S2.

  Here, since nodes A, B, C, and D are recorded in the stack and the stack is not empty (NO in step S2), an unvisited child of node D recorded in the stack at the end is recorded. It is determined whether a node exists (step S3). Since node D is a leaf node and there is no unvisited child node (because it is NO in step S3), the node recorded last is taken out from the stack (step S6). Here, node D is taken out. That is, in the branch including the node C branched from the node B, the node D that is the deepest node is specified. Then, the gene name corresponding to the extracted node is extracted from the gene attribute data storage unit 26 (step S7). Here, a006 and a007 are extracted as gene names. Then, the order, plate number, well number (vertical), (horizontal), gene name, and sample name are recorded in the alignment data storage unit 27 for the gene names that have not been recorded (step S8). Specifically, first, the alignment data 270 is searched with the extracted gene name. Here, when the alignment data 270 having the same gene name as the extracted gene name is recorded, the alignment data 270 is not recorded because the extracted gene name is already recorded. If not recorded, the control computer 21 numbers the order, plate number, well number (vertical), (horizontal). The plate number is numbered in order from “1”, and when the previously recorded alignment data 270 is the well number (vertical) and (horizontal) is “12, H”, the number obtained by adding 1 is numbered. To do. Well numbers (vertical) are numbered sequentially from “01” to “12”. When the well number (horizontal) of the previously recorded alignment data 270 is “H”, the number obtained by adding 1 is numbered. To do. If the well numbers (vertical) and (horizontal) of the previously recorded alignment data 270 are “12” and “H”, the number is returned to “01”. Well numbers (horizontal) are numbered sequentially from “A” to “H”, and when the well number (horizontal) of the previously recorded alignment data 270 is “H”, the number is returned to “A” and numbered. To do. Here, the order data “1”, the plate number “1”, the well number (vertical) “01”, the well number (horizontal) “A”, and the gene name “a006” are set, and the order “2”. , The alignment data 270 in which the plate number “1”, well number (vertical) “01”, well number (horizontal) “B”, and gene name “a007” are set is recorded. Further, the control computer 21 extracts the sample name from the sample name data storage unit based on the gene name, and sets it in the alignment data 270. Note that the order, gene name, and inheritance relationship of attributes in the gene ontology are shown in the alignment display 51. Then, a visit flag is recorded in association with the processed node. Here, a visit flag is recorded for node D. Then, the control computer 21 returns to the process of step S2.

  In this way, the control computer 21 continues the process. Here, nodes A, B, and C are recorded in the stack, and the node C recorded last in the stack is specified as the parent node. Since there is no unvisited child node for this node C, node C is taken out of the stack. That is, in the branch including the node C branched from the node B, the node C connected to the previously specified node D is specified. The gene name “a005” corresponding to the node C is “3” in order.

  Here, nodes A and B are recorded in the stack, and since there is an unvisited child node (node E) for node B, node E is recorded in the stack. Since there is no unvisited child node for this node E, the node E is taken out of the stack. The gene name “a008” corresponding to the node E becomes “4” in order.

  Here, since nodes A and B are recorded in the stack and there is no unvisited child node for node B, node B is taken out of the stack. The gene names “a003” and “a004” corresponding to the node B are in the order “5” and “6”.

  Here, node A is recorded in the stack, there are unvisited child nodes (nodes F and G) for node A, and node F (number of genes: 2) and node G (number of genes: 4). Compare node sizes. Then, node G, which is the largest child node, is recorded on the stack. There are unvisited child nodes (nodes H and I) for this node G, and the node sizes of node H (number of genes: 2) and node I (number of genes: 1) are compared. Then, node H, which is the largest child node, is recorded on the stack. Since there is no unvisited child node for this node H, the node H is taken out of the stack. The gene names “a012” and “a013” corresponding to the node H become “7” and “8” in order.

  Here, nodes A and G are recorded in the stack, and since there is an unvisited child node (node I) in node G, node I is recorded in the stack. Since there is no unvisited child node in this node I, node I is taken out of the stack. Here, since the gene name “a008” corresponding to the node I also corresponds to the node E, the alignment data 270 of the gene name “a008” has already been recorded. Therefore, the alignment data 270 for the gene name “a008” is not recorded at this stage.

  Here, since the nodes A and G are recorded in the stack and there are no unvisited child nodes in the node G, the node G is taken out of the stack. The gene name “a011” corresponding to the node G becomes “9” in order.

  Here, since the node A is recorded in the stack and there is an unvisited child node (node F) in the node A, the node F is recorded in the stack. Since there is no unvisited child node in this node F, the node F is taken out of the stack. The gene names “a009” and “a010” corresponding to the node F become the order “10” and “11”.

  Here, since node A is recorded in the stack and there is no unvisited child node in node A, node A is taken out of the stack. The gene names “a001” and “a002” corresponding to the node A are “12” and “13” in this order.

Here, since the stack is empty (YES in step S2), the process is terminated.
For the genes arranged in this way, the order, gene name, and inheritance relationship of attributes in the gene ontology are shown in the alignment display 51. Here, the alignment data 270 corresponding to the alignment display 51 is recorded in the alignment data storage unit 27. The gene alignment apparatus 20 outputs this alignment data 270. According to the order, plate number, well number (vertical), (horizontal), gene name, and sample name in this alignment data 270, siRNA corresponding to each gene is stored in each well of each microplate. FIG. 7 shows the inheritance relationship of attributes in the corresponding gene name and gene ontology for the siRNA stored in each well of the microplate in the above example.

  In the alignment data 270, as described above, the gene ontology is used to perform alignment by performing a depth-first search in descending order of the node size, and the plate number, well number (vertical), and (horizontal) are sequentially displayed. It is numbered. In the above example, as shown in FIG. 7, for the “01” column, the aligned genes correspond in order from “A”, and for the “02” column, the aligned genes are “A”. Corresponding in order.

  Then, siRNA corresponding to each gene is stored in each well of each microplate according to the order, plate number, well number (vertical), (horizontal), and gene name in the alignment data 270. For example, an apparatus for injecting the prepared sample into each well of the microplate is used. On the other hand, based on the alignment data 270, the plate number, well number (vertical), (horizontal), gene name, and sample name are displayed on a display device connected to the gene alignment apparatus 20. Then, according to the plate number, well number (vertical), (horizontal), gene name and sample name displayed on the display device, siRNA corresponding to each gene is assigned to the microplate and well indicated by the display on the display device. Set to the above device according to the position. Then, using this apparatus, siRNA corresponding to the aligned genes is injected into each well of the microplate. In this way, genes are aligned by performing depth-first search in descending order of node size using gene ontology, and siRNAs corresponding to the aligned genes are respectively stored in the wells of the microplate. Thus, a sample plate is manufactured.

As described above, according to the present embodiment, the following effects can be obtained.
In the above embodiment, using the acyclic directed graph with the gene attribute in the gene ontology as the node 41, the node size 44 of the arbitrary node 41 is set to all the genes belonging to this node and all nodes below this node. The node size is calculated for each node 41. Then, starting from the root node (node A), a depth-first search is performed in descending order of the node size, and after the genes (gene name display 43) belonging to the attributes of the child nodes are aligned, they belong to the attributes of the parent node. The genes are aligned by aligning the genes (gene name display 43). Then, the aligned genes are sequentially arranged corresponding to the wells of the microplate. As a result, samples corresponding to genes having the same attribute and samples corresponding to genes having the same attribute can be successively stored in the wells of the microplate. For this reason, there is a high possibility that a sample corresponding to a gene with the same attribute or a sample corresponding to a gene with an attribute inheriting the same attribute can be stored in the same microplate. Therefore, since the microplate to be used can be limited and the microplate that needs to be taken in and out of the freezer can be limited, the efficiency of sample management and experimental operation can be improved. Also, in the same microplate, samples corresponding to genes with the same attribute and samples corresponding to genes with the same attribute can be placed at close positions, so for example, operations can be performed for each column, The efficiency of the experimental operation can be further improved. Specifically, since the aligned samples are arranged vertically (8 lanes) in a 96-well microplate, the efficiency of the experimental operation can be increased using an 8-channel pipette.

In the above embodiment, the alignment data 270 is not recorded for recorded genes in step S8. That is, the aligned genes are not aligned in duplicate. As a result, samples corresponding to the same gene can be prevented from overlapping, and the genes are aligned by performing depth-first search in the order of the largest node size. Can be aligned. Therefore, siRNA samples can be placed together on the microplate for attributes that are expected to be used more frequently or have higher importance, and attributes that are expected to be used more frequently. The efficiency of sample management and experiment operation can be improved.

(SiRNA library sample plate)
In the above embodiment, a simplified example is used for the explanation of the processing procedure. However, when each sample of the siRNA library prepared for all human genes is stored in each well of the microplate, the same applies. It can be carried out. For example, about 30,000 siRNAs targeting all human genes can be stored in about 300 microplates. In this case as well, depth-first search is performed in descending order of node size using gene ontology. To align the genes.

  For example, only siRNAs (about 1,000 samples) related to transcription factors (inheritance of attributes in gene ontology: “Gene Ontology-molecular function-binding-nucleic acid binding-DNA binding-transcription factor activity”) are extracted and experimented. Assume the case to do. When siRNA samples are randomly arranged on about 300 microplates, almost all microplates need to be removed from the freezer and subjected to an experimental operation. On the other hand, as described above, siRNAs corresponding to genes having the same attribute as transcription factors are grouped by performing a depth-first search in order of increasing node size using gene ontology. Since it is arranged, about 10 sheets may be used for the experiment. Therefore, the microplate taken out from the freezer can be limited, and the efficiency of sample management and experiments can be improved. Moreover, when purchasing such a microplate, since only what is necessary can be purchased, economic efficiency improves.

  In addition, since genes having the same attribute and genes inheriting the same attribute are arranged together on the same microplate, when using siRNA corresponding to the same attribute gene, for example, using an 8-channel pipette By performing the operation, the efficiency of the experiment can be improved.

  As for an actual human gene, there are many cases where the same gene belongs to a plurality of annotations. As described above, in order to eliminate duplication of gene names when recording the alignment data 270, the same gene Alignment data 270 is not recorded in duplicate. For this reason, siRNA corresponding to the same gene is not redundantly stored in the microplate.

  In addition, which attribute is handled as a gene associated with a plurality of attributes affects the positions of microplates and wells that store siRNA. However, by performing depth-first search in descending order of node size, It will be treated as the attribute with the larger size. Therefore, when a gene is associated with a plurality of attributes, the siRNA corresponding to this gene is included in the set of siRNAs corresponding to the larger node size and arranged on the microplate.

An attribute having a large node size generally corresponds to an attribute to which a gene whose function or the like has been examined in detail belongs. For this reason, there is a high possibility that an experiment will be performed on genes associated with a plurality of attributes together with a gene group having an attribute with a larger node size. Therefore, by treating the gene attributes associated with a plurality of attributes as attributes having a larger node size, siRNAs that are likely to be used simultaneously can be assembled and placed on the microplate. As a result, siRNA that is likely to be used simultaneously can be prevented from being dispersed in many microplates, and the efficiency of sample management and experiments can be improved.

In addition, for actual genes, the functions and the like are not completely elucidated for all genes. In the association between the attribute and the gene in the gene ontology, when the more detailed function is unknown, the gene is associated with the attribute of the higher-level obvious function. For example, for a certain gene, “DNA binding”, which has the inheritance relationship of “Gene Ontology-molecular function-binding-nucleic acid binding-DNA binding”, has been clarified. Is associated with an attribute called “DNA binding”. In this way, even if a more detailed function is unknown, since the gene is associated with the attribute for the obvious function, the attribute inheriting this attribute by the depth-first search in the descending order of the node size. Next to the gene, this gene is aligned. For this reason, when a gene whose detailed function is unknown and a gene with an attribute that inherits the attribute associated with this gene is the subject of the experiment, each sample is placed at a close position on the same microplate. Therefore, the efficiency of sample management and experiments can be improved.

Furthermore, as described above, since the functions and the like of all genes are not completely elucidated as described above, the attribute to which each gene belongs may be changed. For example, as described above, it has been clarified up to "DNA binding" that has the inheritance relationship of "Gene Ontology-molecular function-binding-nucleic acid binding-DNA binding" for a certain gene. If not, this gene is "DNA binding
Is associated with the attribute "." Here, when the function of "transcription activity"("Gene Ontology-molecular function-binding-nucleic acid binding-DNA binding-transcription factor activity") that inherits the attribute "DNA binding" is revealed for this gene This gene is associated with the attribute “transcription factor activity”. In this way, since the function of the gene associated with the higher attribute because the more detailed function is unknown is elucidated in more detail, this gene can be associated with the lower attribute. May be changed.

  When aligning genes by the method of the present invention, if a gene is associated with a higher-order attribute because its detailed function is unknown, it is aligned at a position corresponding to this higher-order attribute. And if the function of this gene is elucidated in more detail so that this gene is modified to be associated with a lower attribute, this gene is aligned at a position corresponding to the lower attribute. It is done.

  The node corresponding to the attribute with which this gene is newly associated becomes a lower node of the node corresponding to the attribute with which the gene was originally associated. In the method of the present invention, a depth-first search is performed, and the gene of the attribute corresponding to a certain node and the gene of the attribute corresponding to the lower node are collectively aligned. However, this gene will be aligned in this cluster. For this reason, even if the function of this gene becomes clear in more detail, it is included in the group related to the attribute that was originally associated, so the sample corresponding to this gene is stored on the sample plate or sample plate. The arrangement does not change significantly. In addition, for genes for which more detailed functions have not been clarified, in order to align the genes with the same attributes as long as the functions have been clarified, the functions of the samples corresponding to these genes will be clarified. Within a certain range, sample plates arranged together with samples corresponding to genes having the same attribute can be manufactured.

  When a gene is associated with a lower attribute, the genes are rearranged using the correspondence relationship between the changed gene and the attribute. Specifically, the gene attribute data 260 is updated with data relating to the correspondence relationship between the changed gene and attribute. Then, by performing the above-described processing using the gene attribute data 260, the genes are rearranged. And siRNA corresponding to the rearranged gene is stored in each corresponding well of the corresponding microplate. Thereby, when the function of a gene etc. becomes newly clear, the sample plate which arranged the siRNA reflecting that information can be manufactured.

In addition, you may change the said embodiment into the following aspects.
In the above embodiment, the siRNA library targeting all human genes is assumed as a sample group. However, the present invention is not limited to this, and a large-scale sample prepared for all genes of a specific organism. The present invention can be used for sample groups. For example, siRNA library, siRNA expression vector library, shRNA expression vector library, siRNA expression DNA fragment library, shRNA expression DNA fragment library, cDNA clone library, gene detection prepared for all genes of specific organisms such as humans and mice For large-scale sample groups such as PCR primers and probe libraries, samples corresponding to genes arranged in the same manner may be placed on the microplate.

  In the above embodiment, the case where a sample is stored in a 96-well microplate has been described, but the number of holes in the microplate is not limited thereto. For example, a sample may be stored in the same manner for a 384-hole microplate. Also in this case, the samples corresponding to the aligned genes are aligned on the microplate one column at a time by performing a depth-first search in the descending order of the node size.

  In the above embodiment, the aligned samples are arranged vertically (8 lanes) for the 96-well microplate, but the aligned samples may be arranged horizontally (12 lanes). Thereby, an experimental operation can be performed efficiently using a 12-channel pipette.

  In the above embodiment, the depth-first search is performed by recording the searched nodes on the stack. However, the search algorithm is not limited as long as the depth-first search is performed in the order of the node size. For example, a depth-first search may be performed in order of increasing node size by moving to the next node using recursive calls.

  In the above embodiment, gene alignment was performed for all human genes, and a sample plate storing samples corresponding to the aligned genes was manufactured. Instead of this, a sample plate storing samples corresponding to the aligned genes is arranged for genes having specific attributes and genes that inherit these attributes among genes of specific organisms such as humans. It may be manufactured. In this case, a depth-first search is performed on the acyclic directed graph having the specific attribute as a root node in order of increasing node size. This makes it possible to manufacture a sample plate that can improve the efficiency of sample management and experimental operation even when a gene with a specific attribute is targeted. Also, the attribute of the root node may be specified by the system user. The user of this system may be either a sample plate user or a sample plate provider.

  In addition, for the attribute specified by the user, this attribute and the gene of the attribute that inherits this attribute may be aligned to provide this alignment information. This makes it possible to place a sample for a specific attribute based on the gene alignment information of this attribute and the attribute that inherits this attribute.

  ○ In the above embodiment, when comparing the node size of unvisited child nodes, the node size was calculated each time. However, the node size of each node is calculated and recorded first, and this is used to determine the node size. The node sizes of the child nodes of the visit may be compared.

  In the above embodiment, when the genes are aligned, the arrangement on the microplate is determined each time. However, the arrangement on the microplate may be determined after all the genes are aligned.

  In the above embodiment, when the genes are already aligned, the genes are not overlapped. Alternatively, the aligned genes may be overlapped and aligned. As a result, for any attribute, samples corresponding to the target attribute and the gene of the attribute inheriting this attribute can be collectively placed on the sample plate.

  Further, when the node size is equal to or larger than a certain value, the already arranged genes may be overlapped. As a result, for attributes that have a node size of a certain value or more and are expected to be used to a certain degree of high frequency, or have a high degree of importance, collect samples corresponding to this attribute and genes of attributes that inherit this attribute. Can be placed.

  In the above embodiment, assuming that siRNA samples targeting all human genes are stored in a microplate, all attributes in the gene ontology are targeted for processing. Instead, the target attribute may be limited. In this case, for example, data related to the attribute specified by the user of this system (attribute specifying data) is recorded. In step S8, the alignment data 270 is recorded only when the attribute to be processed is an attribute designated by this attribute designation data or an attribute inheriting this attribute. In this case, a depth-first search depending on the node size is performed for all genes, and the genes are aligned, and the alignment data 270 is recorded only for genes specified by the user or inherited from the attributes. it can.

  Further, the number of genes may be handled as “0” for an attribute specified by the user of this system and an attribute that is neither an attribute corresponding to a lower node of the attribute corresponding to this attribute. In this case, the node size is calculated only for the attribute designated by the user and the number of genes of the attribute inheriting this attribute. Accordingly, the genes can be aligned by performing a depth-first search depending on the node size for the gene specified by the user and the gene having the attribute that inherits the attribute.

  By doing in this way, only the sample corresponding to the gene of the specific attribute designated by the user can be aligned and stored in the sample plate. Therefore, a sample plate group in which only samples corresponding to particularly important genes among all genes are arranged and stored can be manufactured.

  In the above embodiment, the alignment data 270 is recorded when the genes are aligned. However, in addition to this, the data regarding attributes may be recorded and used to search the manufactured sample plate. In this case, a system (sample organizing system) including an alignment process for aligning genes for manufacturing a sample plate and a search process for searching for a sample stored in the manufactured sample plate may be constructed.

In the alignment step, the genes are aligned as in the above embodiment. And the sample plate which stored the sample corresponding to the arranged gene is manufactured. At this time, when the genes are aligned, data on order, plate number, well number (vertical), (horizontal), gene name, sample name, and attributes (sample arrangement specifying data) are recorded. Here, for the attribute, for example, the attribute of this gene and the attribute of the upper node of the node corresponding to this attribute are recorded.

  In the search process, the sample arrangement specifying data is specified based on the gene name or attribute input by the user of this system, and the plate number and well number (vertical), (horizontal) are extracted, Identify the number and placement on the sample plate. Here, when the search is performed based on the attribute, the sample arrangement specifying data in which the attribute to be searched is recorded in the attribute data area is specified. Then, by extracting the plate number and well number (vertical) and (horizontal) of the sample arrangement specifying data specified in this way, the plate number and the arrangement on the sample plate are specified.

  In the alignment step, only the attribute of the node corresponding to the gene to be aligned may be recorded. In this case, when performing a search based on an attribute in the search step, first, the gene ontology data 250 is used to specify a lower node of the node of the search target attribute input by the user. Then, the sample arrangement specifying data is specified for the search target attribute and the attribute of the lower node of the node of this attribute.

  As a usage form of such a sample arrangement system, for example, a case is assumed in which a user of a sample plate orders a sample plate by specifying a gene name and an attribute, and sells the sample plate accordingly. In this case, the sample arrangement system specifies the sample arrangement specifying data based on the gene name and attribute designated by the user, and extracts the plate number. The provider of the sample plate provides the user with the sample plate specified by the plate number.

  In addition, the sample organization system may be configured such that a client terminal (user terminal) used by a user of the sample plate and a server (sample organization server) can be connected via the Internet. In this case, using the user terminal, on the web page, the user inputs a gene name or attribute and makes a search request, and the plate number and well number (vertical) and (horizontal) of the sample plate are obtained from the sample organizing server. You may get it. In this case, based on the gene name or attribute transmitted from the user terminal, the sample organizing server searches the sample arrangement specifying data, extracts the plate number and the like, and transmits a web page displaying the plate number to the user terminal. The user places an order for the sample plate using this plate number. In addition, when the user keeps the purchased sample plate, this sample organization system uses the gene name and attribute for the necessary sample, and the plate number of the plate where the necessary sample is stored, The arrangement position on the plate may be made searchable.

  Thereby, the provider and user of the sample plate can easily know the plate number of the sample plate and the arrangement on the sample plate by designating the gene name or attribute. In addition, if an attribute is specified, a sample plate storing a node corresponding to the attribute and a gene of the attribute corresponding to the lower node can be specified. The sample plate corresponding to the gene group) can be identified.

  In the above embodiment, gene duplication is eliminated in step S8 by not recording the alignment data 270 for the recorded genes. Alternatively, gene duplication may be eliminated after the genes corresponding to all nodes are aligned.

In the above embodiment, the genes corresponding to each node of the entire acyclic directed graph are sequentially arranged. Alternatively, the acyclic directed graph may be divided, the genes may be aligned for each of the divided acyclic directed graphs, and these may be integrated. Specifically, first, the target acyclic directed graph is divided into a plurality of segments, and genes are aligned for each of the divided acyclic directed graphs. Then, the depth-first search is performed on the entire acyclic directed graph in the order of the node size of the highest node, and the previously arranged genes are arranged together in the order of the searched highest node. Then, gene duplication may be eliminated after the genes corresponding to all nodes are aligned.

  ○ In the above embodiment, by specifying the branches of the acyclic directed graph in descending order of the node size, specifying the nodes in order of connection from the deepest node with respect to this branch, and sequentially arranging the genes of the attributes of the specified nodes The genes were aligned. The gene alignment is limited to this if the branch of the acyclic directed graph is specified according to the node size, the nodes of this branch are specified in the order of connection, and the genes of the attribute of the specified node are sequentially aligned. I can't. For example, the genes may be aligned as follows.

  ・ Specify the branches of the acyclic directed graph in ascending order of the node size, specify the nodes in order of connection from the shallowest node in this order, and align the genes of the attribute of the specified node in order, thereby aligning the genes. May be. This processing procedure will be described with reference to FIG. 6 used in the description of the above embodiment.

  First, the attribute genes of the node A that is the root node are aligned. Then, the node F having the smallest node size among the child nodes of the node A is specified, and the gene having the attribute of the node F is aligned. Since this node F has no child nodes, the node G is returned to the node A, and the node G having the smallest node size among the child nodes that have not been subjected to the gene alignment process is specified. Then, the gene having the attribute of the node G is aligned. Next, the node I having the smallest node size among the child nodes of the node G is specified, and genes having the attribute of the node I are aligned. Then, returning to the node G, the genes having the attribute of the node H that is not subjected to the gene alignment process among the child nodes of the node G are aligned. Then, returning to the node G, the node G does not have any child nodes that have not been subjected to the gene alignment process, and therefore returns to the node A. Then, a node B that is not subjected to the gene alignment process among the child nodes of the node A is specified, and the gene having the attribute of the node B is aligned. Thereafter, processing is performed in the same manner.

  That is, for a specific node that has not been subjected to gene alignment processing, the gene alignment processing for sequentially aligning the genes of the attribute of this specific node, and if this specific node has a child node that has not yet been subjected to gene alignment processing, of these nodes A search process including a process for specifying a child node having the smallest size as a next processing target and a process for returning to the parent node of the specific node when there is no child node for which gene alignment processing has not been performed in the specific node. The genes are aligned by starting with the root node as a specific node and repeating until all nodes have been subjected to gene alignment processing.

  Even in this case, similarly, it is possible to align genes so that samples corresponding to genes with the same attribute or samples corresponding to genes with the same attribute can be continuously arranged on the sample plate. . For the same attribute, the order is reversed, but as a result, they are arranged from samples of genes whose functions and the like have been examined in more detail. For this reason, the same convenience as when the genes are arranged in the order according to the above embodiment can be obtained.

In this case, gene duplication may also be eliminated. For example, when genes belonging to the attribute of a node to be processed are already aligned in gene alignment processing, gene duplication is eliminated by deleting already aligned genes. In the case of the above example, when the gene “a008” having the attribute of the node E is aligned, the gene “a008” is already aligned as the gene having the attribute of the node I. Here, the data of the gene “a008” already arranged as the gene of the attribute of the node I is deleted, and the data of the gene “a008” is newly recorded as the gene of the attribute of the node E.

  As a result, gene duplication can be eliminated, and samples corresponding to each gene can be placed together on the sample plate for attributes that are predicted to be used more frequently or have higher importance.

  -Specify the branches of the acyclic directed graph in descending order of the node size, specify the nodes in order of connection from the shallowest node in this order, and align the genes of the specified node attributes in order, thereby aligning the genes. May be. That is, for a specific node that has not been subjected to gene alignment processing, the gene alignment processing for sequentially aligning the genes of the attribute of this specific node, and if this specific node has a child node that has not yet been subjected to gene alignment processing, of these nodes A search process including a process for specifying the child node having the largest size as the next processing target and a process for returning to the parent node of the specific node when there is no child node for which gene alignment processing has not been performed in the specific node. The genes are aligned by starting with the root node as a specific node and repeating until all nodes have been subjected to gene alignment processing.

  Even in this case, similarly, it is possible to align genes so that samples corresponding to genes with the same attribute or samples corresponding to genes with the same attribute can be continuously arranged on the sample plate. .

  In this case, gene duplication may also be eliminated. For example, with respect to already arranged genes, gene duplication is eliminated by not arranging the same genes in duplicate. As a result, gene duplication can be eliminated, and samples corresponding to each gene can be placed together on the sample plate for attributes that are predicted to be used more frequently or have higher importance.

  ・ Specify the branches of the acyclic directed graph in ascending order of the node size, specify the nodes in order of connection from the deepest node to this branch, and arrange the genes of the attribute of the specified node in order, thereby aligning the genes. May be. In other words, when there are child nodes that have not yet been subjected to gene alignment processing in a specific node, the processing for aligning genes is performed by specifying the child node having the smallest node size as the next processing target and the gene in the specific node. Node size support that includes gene alignment processing that sequentially aligns the genes of the attributes of this specific node and processing that specifies the parent node of this specific node as the next processing target when there are no unprocessed child nodes The depth-first search process is performed by starting with the root node as a specific node and repeating the gene alignment process for all nodes.

  Even in this case, similarly, it is possible to align genes so that samples corresponding to genes with the same attribute or samples corresponding to genes with the same attribute can be continuously arranged on the sample plate. .

  In this case, gene duplication may also be eliminated. For example, when genes belonging to the attribute of a node to be processed are already aligned in gene alignment processing, gene duplication is eliminated by deleting already aligned genes. As a result, gene duplication can be eliminated, and samples corresponding to each gene can be placed together on the sample plate for attributes that are predicted to be used more frequently or have higher importance.

In the above embodiment, the case where samples are stored in each well of the microplate has been described. However, the present invention is not limited to this as long as a sample set storing a plurality of samples is manufactured. . For example, the sample set may be a microarray (DNA chip) in which DNA or the like is bound to each storage part of a substrate such as a slide glass as a sample container.

  In the above embodiment, genes are aligned using an acyclic directed graph formed by inheritance relationships between attributes in gene ontology. This alignment method can be used not only when aligning genes but also when aligning objects belonging to each group of groups that form a directed acyclic graph. That is, it is possible to align objects using the same alignment method as in the above embodiment with the genes in the above embodiment as objects and the attributes as groups.

  That is, the computer calculates, for each node, a node size obtained by summing up the number of all objects belonging to this node and all nodes below this node, using an acyclic directed graph with the group to which the object belongs as a node. According to the node size calculating means, the branch of the acyclic directed graph is identified according to the node size, the nodes of the branch are identified in the order of connection, and the objects belonging to the group of the identified nodes are sequentially aligned to align the objects. By functioning as the aligning means to be aligned, the objects are aligned using a computer.

  Here, the alignment means, for example, identifies the branches of the acyclic directed graph in order from the largest node size, identifies the nodes in order of connection from the deepest node with respect to this branch, and identifies objects belonging to the identified group of nodes. The objects are aligned by sequentially aligning them. Thereby, it is possible to align the objects for each group in order from the one that is expected to be more important. Moreover, you may exclude duplication of a target object similarly to said case.

  For example, the objects (files) are arranged in this manner, with the object as a file and the group as a folder (or directory). Further, in the address book for registering the individual name and the organization to which the individual belongs, the object (individual name) is aligned by this method with the object as the individual name and the group as the organization. When an individual belongs to a plurality of organizations, duplication may be eliminated in the same manner as described above.

1 is a schematic diagram of a system according to an embodiment of the present invention. Explanatory drawing of the data memorize | stored in the gene ontology data storage part. Explanatory drawing of the data memorize | stored in the gene attribute data storage part. Explanatory drawing of the data memorize | stored in the alignment data storage part. Explanatory drawing of the process sequence of one Embodiment of this invention. Explanatory drawing of the alignment by the depth priority search process corresponding to a node size. Explanatory drawing of arrangement | positioning of the sample on a microplate. Explanatory drawing of the inheritance relationship between attributes in gene ontology.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 20 ... Gene alignment apparatus, 21 ... Control computer, 25 ... Gene ontology data storage part, 26 ... Gene attribute data storage part, 30 ... Microplate as a sample container, 31 ... Well as storage part.

Claims (12)

A method for producing a sample set in which a sample group corresponding to a gene group is targeted and a sample corresponding to each gene is stored in a sample container having a plurality of storage units,
Using an acyclic directed graph with gene attributes as nodes in gene ontology,
For each node, calculate the node size by summing the number of all genes belonging to this node and all nodes below this node,
Identify the branch of the acyclic directed graph according to the node size, identify the nodes of this branch in the order of connection, perform an alignment process to align the genes by sequentially aligning the genes belonging to the identified nodes ,
A sample set manufacturing method, wherein a sample set is manufactured by storing a sample corresponding to the aligned gene in each storage part of the sample container. The alignment process includes:
When there is an unprocessed child node in which genes are not arranged in the specific node, a child node search process for specifying the next child node to be processed according to the node size from among these,
If there is no unprocessed child node in which a gene is not arranged in a specific node, a gene alignment process for sequentially arranging genes belonging to this specific node, and a parent that specifies the parent node of this specific node as the next processing target 2. The gene alignment is performed by repeating a node size-corresponding depth priority search process including a node identification process, starting with a root node as a specific node and performing a gene alignment process for all nodes. The sample set manufacturing method as described in 2.   The sample set manufacturing method according to claim 1, wherein duplication of the same gene is excluded in the alignment process.   The sample group is any one of siRNA library, siRNA expression vector library, shRNA expression vector library, siRNA expression DNA fragment library, shRNA expression DNA fragment library, cDNA clone library, and PCR primer / probe library for gene detection. The sample set manufacturing method according to any one of claims 1 to 3. A gene alignment program for aligning genes using a computer,
The computer,
Using an acyclic directed graph with gene attributes as nodes in gene ontology,
For each node, a node size calculating means for calculating a node size that is the sum of the number of all genes belonging to this node and all nodes under this node;
Gene alignment for specifying the branch of the acyclic directed graph according to the node size, specifying the nodes of the branch in the order of connection, and sequentially arranging the genes belonging to the specified node to function as an alignment means for aligning the genes program. A gene alignment program that uses a computer to align genes in order to produce a sample set in which a sample group corresponding to a gene group is stored and a sample corresponding to each gene is stored in a sample container having a plurality of storage units. There,
The computer,
Using an acyclic directed graph with gene attributes as nodes in gene ontology,
For each node, a node size calculating means for calculating a node size that is the sum of the number of all genes belonging to this node and all nodes under this node;
An alignment means for specifying a branch of the acyclic directed graph according to the node size, specifying a node in order from a deep node with respect to the branch, and aligning genes by sequentially aligning genes belonging to the specified node ;
The gene alignment program for functioning as an arrangement | positioning means to arrange | position the said aligned gene corresponding to each storage part of the said sample container. The alignment means;
When there is an unprocessed child node in which genes are not arranged in the specific node, a child node search process for specifying the next child node to be processed according to the node size from among these,
If there is no unprocessed child node in which a gene is not arranged in a specific node, a gene alignment process for sequentially arranging genes belonging to this specific node, and a parent that specifies the parent node of this specific node as the next processing target A node size-corresponding depth-first search process including a node specifying process is started as a specified node, and is repeated until a gene aligning process is performed for all nodes, thereby functioning as a means for aligning genes. The gene alignment program according to claim 5 or 6.   The gene alignment program according to any one of claims 5 to 7, wherein the alignment means further functions as means for eliminating duplication of the same gene. A sample set for a sample group corresponding to a gene group, in which a sample corresponding to each gene is stored in a sample container having a plurality of storage units,
Using an acyclic directed graph with gene attributes as nodes in gene ontology,
For each node, calculate the node size by summing the number of all genes belonging to this node and all nodes below this node,
Identify the branch of the acyclic directed graph according to the node size, identify the nodes of this branch in the order of connection, perform an alignment process to align the genes by sequentially aligning the genes belonging to the identified nodes ,
A sample set manufactured by storing a sample corresponding to the aligned gene in each storage part of the sample container to manufacture a sample set. The alignment process includes:
When there is an unprocessed child node in which genes are not arranged in the specific node, a child node search process for specifying the next child node to be processed according to the node size from among these,
If there is no unprocessed child node in which a gene is not arranged in a specific node, a gene alignment process for sequentially arranging genes belonging to this specific node, and a parent that specifies the parent node of this specific node as the next processing target 10. The node size-corresponding depth-first search process including a node specifying process is started with a root node as a specified node, and the genes are aligned by repeating until the gene alignment process is performed for all nodes. Sample set as described in.   The sample set according to claim 9 or 10, wherein duplication of the same gene is excluded in the alignment process. The sample group is any one of siRNA library, siRNA expression vector library, shRNA expression vector library, siRNA expression DNA fragment library, shRNA expression DNA fragment library, cDNA clone library, and PCR primer / probe library for gene detection. The sample set according to any one of claims 9 to 11 .

Images