Sequence clustering method
The sequence clustering method addresses the challenge of managing virus safety in biopharmaceuticals by efficiently processing base sequences to identify potential virus contamination through lead acquisition and reference clustering, ensuring effective virus safety management in biopharmaceuticals.
Patent Information
- Application Number
- PCT/JP2024/017548
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-05-14
- Filing Date
- 2024-05-11
- Publication Date
- 2024-11-21
AI Technical Summary
Existing methods for managing virus safety in biopharmaceuticals face challenges in efficiently clustering sequences from databases due to high duplication rates and overlapping viral base sequences, leading to difficulties in identifying potential virus contamination.
A sequence clustering method that includes lead acquisition, ID set acquisition, frequent lead identification, and reference clustering steps, using a computer to process base sequences and convert high-frequency reference identification information into cluster identification, thereby reducing the number of references to be considered and facilitating quick analysis.
Enables efficient clustering of sequences, reducing the risk of false negatives and allowing for rapid analysis of potential virus contamination in biopharmaceutical samples by aggregating references into manageable clusters.
Smart Images

Figure JP2024017548_21112024_PF_FP_ABST
Abstract
Description
Sequence clustering method
[0001] The present invention relates to a method for clustering sequences.
[0002] JP 2022-176173 A describes a method for classifying biological sequences.
[0003] For example, in the field of biopharmaceutical manufacturing, where pharmaceuticals are produced using biological materials such as cells, safety control of the manufactured pharmaceuticals may involve checking that the final product is not contaminated with viruses or components derived from the manufacturing host cells. Samples derived from biopharmaceutical manufacturing intermediates or final products may be sequenced and searched against a database containing nucleotide sequences previously obtained from samples containing viruses. If the results of this search reveal that a given sample contains a nucleotide sequence derived from a virus, the risk of viral contamination in that sample can be recognized. Therefore, to implement virus safety control of biopharmaceutical manufacturing intermediates and final products, a database containing the names of organisms and nucleotide sequences previously obtained from samples containing that organism can be used.
[0004] Incidentally, when creating a database by collecting nucleotide sequences obtained from samples containing viruses, it is necessary to design the amount of nucleotide sequences to be included in the database. This is because the number of viral nucleotide sequences that have been sequenced to date is enormous, and there is a large amount of overlap between nucleotide sequences from samples obtained at different times and places for the same type of virus. While databases with many overlaps as mentioned above are highly comprehensive and have a low risk of not finding sequences that actually contain viral origins in a search, they have the disadvantage that when searching for nucleotide sequences using the database, a single read may match multiple times with the same type of viral nucleotide sequence, making it difficult to tally the results.
[0005] Japanese Patent Application Laid-Open No. 2022-176173
[0006] An object of the present invention is to provide a method for clustering sequences that can easily cluster sequences contained in a database in association with reads derived from a sample.
[0007] This invention is based on the finding that references can be clustered by measuring the similarity between references (determining their relatedness) using leads.
[0008] The sequence clustering method includes a lead acquisition step (S110), a lead ID acquisition step (S120), an ID set acquisition step (S130), a high frequency lead ID acquisition step (S140), a high frequency lead identification information acquisition step (S150), a related reference information acquisition step (S160), a reference cluster step (S170), and a replacement ID set acquisition step (S180).
[0009] The lead acquisition step (S110) is a step of acquiring a lead, which is a base sequence derived from a sample. An example of the sample is a biopharmaceutical.
[0010] The lead ID acquisition step (S120) is a step of acquiring lead identification information.
[0011] The ID set acquisition step (S130) is a step of referring to a database containing the base sequences of multiple references, storing the base sequence of each reference in association with the biological species and the identification information of each reference, comparing the read with the references contained in the database, and acquiring an ID set that is a pair of the identification information of the read and the identification information of the reference if the base sequence of the read is contained in the reference. Examples of reference base sequences contained in the database are base sequences derived from viruses, fungi, and bacteria that may be contained in biopharmaceuticals.
[0012] The high frequency read ID acquisition step (S140) is a step of calculating the number of identification information pieces of the reads included in the ID set and obtaining high frequency read identification information, which is identification information of high frequency reads, which are reads with a high occurrence rate.
[0013] The high frequency read identification information acquisition step (S150) is a step of obtaining high frequency reference identification information, which is identification information of a reference paired with high frequency read identification information included in an ID set.
[0014] The related reference information acquisition step (S160) is a step of using the high frequency reference identification information to acquire, from a database, information including the biological species of the high frequency reference sequence stored in association with the high frequency reference identification information.
[0015] The reference cluster step (S170) is a step that enables high frequency reference identification information to be replaced with a new piece of identification information, that is, first cluster identification information.
[0016] The replacement ID set acquisition step (S180) is a step of acquiring a replacement ID set in which high frequency reference identification information included in an ID set is converted into first cluster identification information.
[0017] A preferred example of the ID set acquisition step (S130) includes a first read group acquisition step (S131) and a first ID set acquisition step (S132). The first read group acquisition step (S131) is a step of comparing the read with a first host-derived reference, which is a reference derived from the host of the sample, to obtain a first read group consisting of reads excluding reads whose base sequences are included in the first host-derived reference sequence. The first ID set acquisition step (S132) is a step of comparing the first read group with a reference included in a database to obtain a first ID set. This example preferably further includes a second read group acquisition step (S133) and a second ID set acquisition step (S134). The second read group acquisition step (S133) is a step of comparing the read with a second host-derived reference, which is a reference derived from the host of the sample, to obtain a second read group consisting of reads excluding reads whose base sequences are included in the second host-derived reference sequence. The second host-derived reference sequence is a sequence different from the first host-derived reference sequence. The second ID set acquisition step (S134) is a step of comparing the second group of reads with the references contained in the database to acquire the second ID set.
[0018] A preferred example of the ID set acquisition step (S130) further includes a duplication removal step (S135). The duplication removal step (S135) is a step of, when there are multiple ID sets in which the pair of lead identification information and reference identification information is the same, leaving only one of the ID sets and deleting the other ID sets.
[0019] A preferred example of a sequence clustering method is that when a replacement ID set contains multiple pieces of identification information for a certain read, a repeating step (S190) may be performed using the replacement ID set, which includes a high-frequency read ID acquisition step, a high-frequency read identification information acquisition step, a related reference information acquisition step, and a replacement ID set acquisition step.
[0020] A computer for performing sequence clustering includes a processor and a memory coupled to the processor. The memory stores a program including instructions. When executed by the processor, the instructions cause the computer to perform a sequence clustering method including a lead acquisition step (S110), a lead ID acquisition step (S120), an ID set acquisition step (S130), a high-frequency lead ID acquisition step (S140), a high-frequency lead identification information acquisition step (S150), a related reference information acquisition step (S160), a reference cluster step (S170), and a replacement ID set acquisition step (S180). When executed by the processor, the instructions may further cause the computer to perform a repeat step (S190).
[0021] The above program is a program for causing a computer to execute a sequence clustering method including a lead acquisition step (S110), a lead ID acquisition step (S120), an ID set acquisition step (S130), a high-frequency lead ID acquisition step (S140), a high-frequency lead identification information acquisition step (S150), a related reference information acquisition step (S160), a reference cluster step (S170), and a replacement ID set acquisition step (S180). The program may further cause the computer to execute a repeat step (S190).
[0022] The non-transitory information recording medium stores the above program.
[0023] According to this invention, sequences contained in a database can be easily clustered in association with reads derived from a sample, which makes it easy to analyze, for example, whether a sample contains a virus.
[0024] FIG. 1 is a flowchart showing an example of the steps of a clustering method. FIG. 2 shows an example for explaining the clustering method of the present invention. FIG. 2A shows example sequences of reads and references and example identification information. FIG. 2B is a diagram for explaining an ID set acquisition process including a first read group acquisition process and a first ID set acquisition process. FIG. 2C is a diagram for explaining the process after an ID set has been acquired. FIG. 2D is a diagram for explaining an example of clustering processing after a conversion table has been obtained. FIG. 2E is a diagram for explaining an example of the steps up to obtaining a second conversion table in the iterative process. FIG. 2F is a diagram for explaining an example of the steps after obtaining the second conversion table in the iterative process. FIG. 3 shows an example for explaining a clustering method when a new reference is used. FIG. 3A is a diagram showing an example of obtaining an ID set using a second reference. FIG. 3B shows an example of obtaining a second recombinant ID set using the obtained ID set.
[0025] The following describes embodiments of the present invention with reference to the drawings. The present invention is not limited to the embodiments described below, but also includes appropriate modifications of the embodiments that are obvious to those skilled in the art.
[0026] FIG. 1 is a flowchart showing an example of the steps of a clustering method. As shown in FIG. 1, the sequence clustering method includes a lead acquisition step (S110), a lead ID acquisition step (S120), an ID set acquisition step (S130), a high-frequency lead ID acquisition step (S140), a high-frequency lead identification information acquisition step (S150), a related reference information acquisition step (S160), a reference clustering step (S170), and a replacement ID set acquisition step (S180). As shown in FIG. 1, this method may further include an iteration step (S190). This method is basically implemented by a computer or server. The sequence clustering method involves aggregating references into groups (clusters). Clustering can reduce the number of references to be considered, enabling faster analysis of sample-derived reads.
[0027] A computer has an input unit, an output unit, a control unit, a calculation unit, and a memory unit, and each element is connected by a bus or the like to enable the exchange of information. For example, the memory unit may store a control program or various information. When predetermined information is input from the input unit, the control unit reads the control program stored in the memory unit. The control unit then reads the information stored in the memory unit as appropriate and transmits it to the calculation unit. The control unit also transmits the input information as appropriate to the calculation unit. The calculation unit performs calculation processing using the received various information and stores it in the memory unit. The control unit reads the calculation results stored in the memory unit and outputs them from the output unit. In this way, various processes and steps are performed. Each unit or means executes these various processes. A computer may have a processor, and the processor may realize various functions and steps. A computer may be standalone. A computer may have some of its functions distributed between a server and a terminal. In this case, it is preferable that the server and the terminal can exchange information via a network such as the Internet or an intranet. The computer may include a processor and a memory coupled to the processor. The memory may store instructions that, when executed by the processor, cause the computer to perform various processes or function as various elements. The computer may be provided with various training data to construct a learning model and perform various calculations through machine learning. In this case, the computer may perform various analyses using the learning model created through machine learning and deep learning of AI (artificial intelligence).
[0028] The sequence clustering method may be implemented by a sequence clustering device. This device is implemented by a computer and includes a lead acquisition means for performing a lead acquisition step (S110), a lead ID assignment means for performing a lead ID acquisition step (S120), an ID set acquisition means for performing an ID set acquisition step (S130), a high-frequency lead ID acquisition means for performing a high-frequency lead ID acquisition step (S140), a high-frequency read-containing ID set acquisition means for performing a high-frequency lead identification information acquisition step (S150), a related reference information acquisition means for performing a related reference information acquisition step (S160), a reference cluster means for performing a reference cluster step (S170), and a replacement ID set acquisition means for performing a replacement ID set acquisition step (S180). This device may further include a repeating means for performing a repeating step (S190).
[0029] FIG. 2 shows an example for explaining the clustering method of this invention. FIG. 2A shows example sequences of reads and references, as well as example identification information. For simplicity, the example in FIG. 2A deals with reads A to S derived from a specimen and references 1 to 13. Reference 1 is a sequence derived from a host (host cell genome sequence). Reference 2 is a sequence derived from a host infected with a virus. References 3 to 13 are either recorded in a database related to viruses or stored in the database as references for viruses associated with the specimen. In this example, reads A to D are reads derived from sources other than the host. Reads E to S are reads derived from references 1 or 2. Nucleic acids derived from host cells are also included in the specimen and are read together with reads derived from sources other than the host by a sequencer or the like. Therefore, reads derived from the host are also included as reads derived from the specimen. In the example in FIG. 2A, the host cell from which reference 2 was derived was infected with a virus (reference 13), and therefore has the sequence of reference 13 as a partial sequence. References 3 to 7 relate to the same virus but obtained on different specimen acquisition dates. References 8 to 10 also relate to the same virus but obtained on different dates. References 11 to 13 are sequences derived from viruses, etc.
[0030] The example shown in Figure 2A is an example, and in the actual clustering, the length of the read sequence was approximately 50 to 200 bases. Furthermore, the reference lengths of References 1 and 2, which are host cell genomes, were approximately 2.6 billion bases, while References 3 to 13 were several hundred to several tens of thousands of bases. The base length of SARS-CoV-2 was approximately 20,000 bases. The base length of the read may be 5 to 100,000 bases, 10 to 10,000 bases, or 10 to 1,000 bases. The base length of the reference is preferably longer than the base length of the read, and is preferably 10,000 to 50 billion bases, and may be 100,000 to 10 billion bases, or may be 1 million to 10 billion bases.
[0031] The lead acquisition step (S110) is a step of acquiring a lead, which is a base sequence derived from a sample. An example of a sample is a biopharmaceutical. For example, the clustering device is connected to a sequencer. The sequencer analyzes the base sequence contained in the sample. The analyzed base sequence (read) is input into the clustering device. In this manner, the clustering device acquires the lead. Note that the clustering device may acquire the lead by dragging and dropping the base sequence of the read into software in the clustering device. The acquired base sequence of the read is stored in a storage unit as appropriate. Hereinafter, the information acquired by the clustering device in this manner may be stored in a storage unit as appropriate. In this step, for example, a base sequence derived from a sample analyzed by a sequencer may be input into the clustering device to acquire a lead (and a lead ID).
[0032] The lead ID acquisition step (S120) is a step of acquiring lead identification information. For example, when acquiring leads, the clustering device may also acquire the identification information of each lead stored together with the leads. Furthermore, the clustering device may assign identification information to each lead stored in the storage unit and store the identification information in the storage unit. In this way, the clustering device acquires lead identification information.
[0033] In the example of Figure 2A, there were 18 reads, so reads A to S were input to the clustering device along with their sequence information and stored in the memory of the clustering device. Note that the base sequences of the reads may be input, and identification information (reads A to S) may be assigned to each input base sequence of the read and stored in the memory.
[0034] The ID set acquisition step (S130) refers to a database containing the base sequences of multiple references, storing each reference's base sequence in association with the biological species and the identification information of each reference, and compares the read with the references contained in the database. If the base sequence of the reference contains the base sequence of the read, this step acquires an ID set, which is a pair of the identification information of the read and the identification information of the reference. Examples of reference base sequences contained in the database are base sequences derived from viruses, bacteria, fungi, etc. (viruses, etc.) that may be contained in biopharmaceuticals. The above database may also store references derived from viruses, etc. The comparison process is similar to the process described above.
[0035] A preferred example of the ID set acquisition step (S130) includes a first read group acquisition step (S131) and a first ID set acquisition step (S132). The clustering device preferably includes a first read group acquisition means for performing the first read group acquisition step (S131) and a first ID set acquisition means for performing the first ID set acquisition step (S132). The first read group acquisition step (S131) is a step of comparing a read with a first host-derived reference, which is a reference derived from the host of the sample, and obtaining a first read group, which is a read excluding reads whose base sequence is included in the first host-derived reference sequence. The host-derived reference of the sample may be obtained from a database or from a storage unit other than a database. The host of the sample refers to the host used to obtain various components used in the sample. For example, in the case of a pharmaceutical product for human use, human cells may be used to obtain the pharmaceutical product, in which case the human may be the host. When a component contained in the sample is cultured using fetal bovine serum (FBS), an example of the host of the sample is a cow. The first ID set acquisition step (S132) is a step of acquiring the first ID set by comparing the first read group with a reference contained in a database.
[0036] In the example of FIG. 2A, Reference 1, which is a reference for a host-derived sequence (host cell genome sequence), corresponds to the first host-derived reference sequence. FIG. 2B is a diagram for explaining an ID set acquisition process including a first read group acquisition process and a first ID set acquisition process. Reference 1 may be obtained from a database or from a storage device other than a database. As shown in FIG. 2B, the clustering device compares Reads A to S with Reference 1. Specifically, the clustering device reads the base sequence of Read A from the storage unit and performs a comparison process to determine whether it is included in the base sequence of Reference 1. The clustering device also performs this process on reads B and subsequent reads. In the example of FIG. 2, the base sequence of Reference 1 includes the base sequences of Reads E to S. Therefore, the clustering device removes Reads E to S from Reads A to S and stores Reads A to D in the storage unit as a first read group.
[0037] In the example of FIG. 2B , an ID set (first ID set) is obtained using the first read group as a read. In the example of FIG. 2B , the database stores references 3 to 13 as references related to viruses, etc. This database stores information on the biological species and information on the base sequence for each of references 3 to 13, which is reference identification information. The clustering device then compares each of reads A to D included in the first read group with references 3 to 13. Specifically, the clustering device reads the base sequence of read A from the storage unit and performs a comparison process to determine whether the base sequence of read A is included in the base sequences of references 3 to 13. This process is also performed for reads B to D. In the example of FIG. 2 , multiple ID sets are obtained, such as ID read A-ID reference 3. However, if, for example, a partial sequence of the reference indicated by ID reference 3 matches the sequence of the read indicated by ID read B, a set such as ID read B-ID reference 3 may exist. ID sets are obtained in this manner and stored in the storage unit.
[0038] A preferred example of the ID set acquisition step (S130) further includes a duplication removal step (S135). The duplication removal step (S135) is a step in which, when there are multiple ID sets in which the pair of lead identification information and reference identification information is the same, only one of the ID sets is retained and the others are deleted. For example, data of combinations of sample-derived lead IDs and database-derived sequence IDs can be aligned in a row. Then, for rows of data that completely match, one row can be retained and the rest can be deleted. For example, if there are two or more ID read A-ID reference 3 pairs (if the sequence of read A exists in two or more places in the sequence of reference 3), only one ID set can be retained and the other ID sets (pairs in which the IDs of the read and reference are the same) can be deleted. In this way, an ID set after duplication removal can be obtained. Note that if there are no duplicates in the ID sets, the ID set obtained in the previous step will be the ID set after duplication removal. For this reason, in subsequent steps, the ID set after duplication removal will also be simply referred to as an ID set. The clustering device reads out the lead identification information and reference identification information contained in each ID set, and if there are multiple ID sets that match these, it simply retains one of the ID sets and deletes the others.
[0039] In the example of Figure 2B, there are no multiple ID sets in which the pair of lead identification information and reference identification information included in the ID set is the same. Therefore, there is no need to perform the duplication removal step (S135). Although not the example of Figure 2B, for example, if there are multiple ID sets of ID read A - ID reference 3, it is sufficient to leave one of the ID sets and delete the others.
[0040] The high-frequency lead ID acquisition step (S140) is a step of calculating the number of identification information pieces of leads included in an ID set and obtaining high-frequency lead identification information, which is identification information pieces of high-frequency leads that are leads with a high occurrence rate. The clustering device reads the ID set stored in the storage unit, performs a calculation to calculate the number of identification information pieces of leads included in the ID set, and calculates the number of identification information pieces of each lead included in the ID set. The clustering device then compares the calculated numbers of identification information pieces of each lead. In this way, the clustering device can obtain high-frequency lead identification information.
[0041] 2C is a diagram for explaining the process after acquiring the ID set. In the example of FIG. 2C, the ID set includes five ID reads A, and the ID reads A are high-frequency read identification information.
[0042] The high-frequency read identification information acquisition step (S150) is a step of obtaining high-frequency reference identification information, which is identification information of a reference paired with the high-frequency read identification information included in the ID set. The clustering device reads out the high-frequency read identification information from the storage unit, and reads out identification information of the reference paired with the high-frequency read identification information in the ID set. In this way, the clustering device can obtain the high-frequency reference identification information.
[0043] In the example of Figure 2C, five of the ID sets include ID read A, and therefore the reference identification information in these five ID sets (references 3 to 7) is the high-frequency read identification information. In the example of Figure 2C, the high-frequency read identification information was grouped together to form a cluster (cluster 1).
[0044] The related reference information acquisition step (S160) is a step of using the high-frequency reference identification information to acquire, from a database, information including the biological species of the high-frequency reference sequence stored in association with the high-frequency reference identification information. The clustering device uses the high-frequency reference identification information to acquire, from a database, information including the biological species of the high-frequency reference sequence stored in association with the high-frequency reference identification information. Examples of information including the biological species of the high-frequency reference sequence may include the biological species (e.g., virus name), as well as any one or more of information about the reference acquisition date, information about the reference sequence, and information about past clusters. The clustering device may identify the biological species of the high-frequency read from the acquired information about the biological species. Furthermore, since the clustering device aggregates information about the biological species of the references related to the high-frequency read, displaying this information makes it easier to determine the biological species of the high-frequency read. By performing the related reference information acquisition step (S160), it is possible to determine whether to group the high-frequency reads into a cluster not only based on the identification information (ID) but also based on the biological species name, such as a virus. In the related reference information acquisition step (S160), if there is one type of biological species in the high frequency read, it may be determined that the high frequency read originates from that biological species. Furthermore, if there are multiple types of biological species in the high frequency read, the biological species with the highest frequency may be designated as the biological species of the high frequency read. Then, only the high frequency reads that are the high frequency biological species may be subjected to the following clustering. Furthermore, if the biological species of the high frequency read include those derived from the sample, the following clustering may be performed excluding those high frequency reads.
[0045] The related reference information acquisition step (S160) is an optional step and may be omitted. As described above, a reference related to one of the biological species obtained by the related reference information acquisition step (S160) may be used for clustering, which will be described later.
[0046] In the example of FIG. 2C , information such as the biological species of references 3 to 7 is read out in association with ID read A. The clustering device may determine, for example, that ID read A originates from a coronavirus based on the read out biological species. For example, if references 3 to 7 are references related to coronaviruses, it may be determined that the biopharmaceutical sample may have been contaminated with coronaviruses. Also, if references 3 to 6 are references related to coronaviruses and reference 7 is a reference derived from fungi or a host, it may be determined that the biopharmaceutical sample may have been contaminated with coronaviruses. In this case, references 3 to 6 may be clustered in a reference clustering step described below.
[0047] The reference cluster step (S170) is a step of replacing the high-frequency reference identification information with a new single piece of identification information, the first cluster identification information. In this step, information for replacing the high-frequency reference identification information with the new single piece of identification information, the first cluster identification information, may be stored in a storage unit. An example of such information is information (e.g., a table) that stores the high-frequency reference identification information in association with the first cluster identification information. For example, the clustering device stores the first cluster identification information in association with the high-frequency reference identification information in the storage unit. This allows the clustering device to convert the high-frequency reference identification information into the first cluster identification information.
[0048] In the example of FIG. 2C, ID references 3 to 7 are replaced with ID cluster 1. In the example of FIG. 2C, a first conversion table (conversion table 1) is obtained. Using this table, ID references 3 to 7 can be rewritten into a unified ID like ID cluster 1.
[0049] The replacement ID set acquisition step (S180) is a step of acquiring a replacement ID set by converting high-frequency reference identification information included in an ID set into first cluster identification information. Here, an example of an ID set is the ID set acquired in the ID set acquisition step (S130). The clustering device stores lead identification information and reference identification information for each ID set. Therefore, if the reference identification information for each ID set is high-frequency reference identification information, the clustering device converts the high-frequency reference identification information into cluster identification information and acquires a replacement ID set. Furthermore, if converting the high-frequency reference identification information into cluster identification information results in multiple ID sets having the same pair of lead identification information and reference identification information, the clustering device may retain only one of the ID sets and delete the other ID sets. In this way, the clustering device can acquire a replacement ID set.
[0050] FIG. 2D is a diagram for explaining an example of clustering processing after obtaining a conversion table. In the example of FIG. 2D, ID references 3 to 7 are replaced with ID cluster 1. This generates five pairs of ID read A and ID cluster 1. Therefore, the clustering device may leave one ID pair of ID read A and ID cluster 1 and delete the remaining four. In this way, a replacement ID pair can be obtained. In this way, the degree of association between references can be measured using the read. As shown in FIG. 2D, the clustering device can obtain a replacement ID pair.
[0051] In a preferred example of a sequence clustering method, when a replacement ID set contains multiple pieces of identification information for a certain read, a repeat step (S190) may be performed using the replacement ID set, which includes a high-frequency read ID acquisition step, a high-frequency read identification information acquisition step, a related reference information acquisition step, and a replacement ID set acquisition step. In the high-frequency read ID acquisition step (S140), the clustering device calculates the number of each read included in the ID set and stores it in a memory unit. The clustering device may determine whether to perform the repeat step (S190) on the read with the second highest number after the high-frequency read as a new high-frequency read. For example, the clustering device may store a threshold in the memory unit, and perform the repeat step (S190) if the number included in the ID set of the read with the second highest number after the high-frequency read is equal to or greater than the threshold. In this way, this invention enables reference clustering by measuring the similarity between references using reads.
[0052] 2E is a diagram illustrating an example of the process of obtaining the second conversion table in the repeated process. In the example of FIG. 2E, the same process as that for ID read A described above is performed on ID read B. As a result, a second cluster (cluster 2) is obtained, and a second conversion table (conversion table 2) is also obtained.
[0053] FIG. 2F is a diagram illustrating an example of the process after the second conversion table is obtained in the repetitive process. As shown in FIG. 2F, in this example, a second set of recombinant IDs with advanced clustering is obtained. The example in FIG. 2F shows the repetitive process performed once. This repetitive process may be performed multiple times. For example, if the repetitive process is performed again, reference 11 and reference 12 will become cluster 3.
[0054] A preferred example of the sequence clustering method includes a second read group acquisition step (S133) and a second ID set acquisition step (S134). The newly acquired second ID set is then used to perform the various steps described above. By doing so, it becomes possible to determine whether the host itself is infected with a virus by comparing the first ID set with the second ID set, or by comparing the first replaced ID set with the second replaced ID set. In other words, false negatives can be prevented. This is because when a host-derived reference and a read are compared, if the host itself is infected with a virus, even if the virus is contained in the sample, it may be determined to be host-derived and negative (no virus is contained).
[0055] By using the replacement ID set (nth replacement ID set) obtained as described above, it is possible to check whether a sample is contaminated with a virus or the like using a small number of ID sets. Therefore, this specification also provides a method for checking the contamination status of a sample using the above-mentioned clustering method. In other words, if a reference derived from a virus or the like is present in the replacement ID set, the computer can determine that the sample is contaminated with a virus or the like.
[0056] The second read group acquisition step (S133) is a step of comparing the reads with a second host-derived reference, which is a reference derived from the host of the sample, to obtain a second read group, which is a read group excluding reads whose base sequences are included in the second host-derived reference sequence.The second host-derived reference sequence is a sequence different from the first host-derived reference sequence.The second ID set acquisition step (S134) is a step of comparing the second read group with a reference included in the database to obtain a second ID set.
[0057] FIG. 3 shows an example of a clustering method using a new reference. FIG. 3A shows an example of obtaining an ID set using a second reference. FIG. 3B shows an example of obtaining a second recombinant ID set using the obtained ID set. The cells from which Reference 2 was derived were infected with the Reference 13 virus before the reference genome was obtained. As a result, the sequence of the Reference 13 virus was mixed into Reference 2 as a host cell-derived sequence. The presence of the Reference 13 virus in the sample was overlooked. When removing host-derived sequences from the reads by matching, false negatives can be suppressed by using multiple independently obtained reference sequences.
[0058] A computer for performing sequence clustering includes a processor and a memory coupled to the processor. The memory stores a program including instructions. When executed by the processor, the instructions cause the computer to perform a sequence clustering method including a lead acquisition step (S110), a lead ID acquisition step (S120), an ID set acquisition step (S130), a high-frequency lead ID acquisition step (S140), a high-frequency lead identification information acquisition step (S150), a related reference information acquisition step (S160), a reference cluster step (S170), and a replacement ID set acquisition step (S180). When executed by the processor, the instructions may further cause the computer to perform a repeat step (S190).
[0059] The above program is a program for causing a computer to execute a sequence clustering method including a lead acquisition step (S110), a lead ID acquisition step (S120), an ID set acquisition step (S130), a high-frequency lead ID acquisition step (S140), a high-frequency lead identification information acquisition step (S150), a related reference information acquisition step (S160), a reference cluster step (S170), and a replacement ID set acquisition step (S180). The program may further cause the computer to execute a repeat step (S190).
[0060] The non-transitory information recording medium is a medium that stores the above program. Examples of the recording medium include a hard disk, a CD, a CD-ROM, a floppy disk, a Blu-ray disk, a USB memory, and an SD card.
[0061] The present invention can be used in the pharmaceutical and reagent industries.
Claims
1. A lead acquisition step for acquiring a lead, which is a base sequence derived from a specimen; a lead ID acquisition step for acquiring identification information of the lead; an ID set acquisition step for referring to a database including base sequences of a plurality of references, in which the base sequence of each reference is stored in association with a biological species and identification information of each reference, and for matching the lead with the references included in the database, and acquiring an ID set that is a pair of identification information of the lead and identification information of the reference if the base sequence of the lead is included in the base sequence of the reference; a high-frequency lead ID acquisition step for calculating the number of identification information of the lead included in the ID set, and acquiring high-frequency lead identification information that is identification information of a high-frequency lead, which is a lead with a high occurrence rate; a high-frequency lead identification information acquisition step for acquiring high-frequency reference identification information that is identification information of a reference paired with the high-frequency lead identification information included in the ID set; and a related reference information acquisition step for using the high-frequency reference identification information to acquire information including the biological species of the high-frequency reference sequence stored in association with the high-frequency reference identification information from the database. A method for clustering sequences, comprising: a reference cluster process which is a process for replacing the high-frequency reference identification information with a new identification information, which is a first cluster identification information; and a replacement ID set acquisition process which is a process for acquiring a replacement ID set in which the high-frequency reference identification information included in the ID set is converted into the first cluster identification information.
2. A method according to claim 1, wherein the ID set acquisition process includes a first read group acquisition process which is a process of comparing the read with a first host-derived reference which is a reference derived from the host of the specimen, and obtaining a first read group which is reads excluding reads whose base sequences are included in the first host-derived reference sequence, and a first ID set acquisition process which is a process of comparing the first read group with a reference included in the database, to obtain a first ID set.
3. The method according to claim 2, wherein the ID set acquisition step further comprises: a step of comparing the read with a second host-derived reference that is a reference derived from the host of the sample, to obtain a second group of reads excluding the reads whose base sequences are included in the second host-derived reference sequence, the second group of reads being sequences different from the second host-derived reference sequence and the first host-derived reference sequence; and a second ID set acquisition step of comparing the second group of reads with a reference included in the database to obtain a second ID set.
4. The method according to claim 2 or 3, wherein the sample is a biopharmaceutical, and the reference base sequence included in the database is a base sequence derived from a virus that may be contained in the biopharmaceutical.
5. A method according to claim 4, wherein the ID set acquisition step further includes a duplicate removal step, which is a step of, when there are multiple ID sets in which the set of identification information of the lead and the identification information of the reference is the same, deleting the other ID sets and leaving only one of the ID sets.
6. A method according to claim 4, wherein, when a replacement ID set contains multiple pieces of identification information for a lead, the replacement ID set is used to perform the high frequency lead ID acquisition process, the high frequency lead identification information acquisition process, the related reference information acquisition process, and the replacement ID set acquisition process.
7. A computer for performing sequence clustering, the computer comprising a processor and a memory coupled to the processor, the memory storing a program including instructions which, when executed by the processor, cause the computer to perform the sequence clustering method of claim 1.
8. A program for causing a computer to execute the sequence clustering method according to claim 1.
9. A non-transitory information recording medium storing the program according to claim 8.
Citation Information
Patent Citations
Method, computer program and computer-implemented method for classifying biological sequences (identification of unknown genomes and closest known genomes)
JP2022176173A
Microbial contamination countermeasure selection device, system, method, and program
WO2019092847A1
