JP2003141101A - Analysis information management system - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide an analysis information management system capable of efficiently and easily managing even an enormous amount of analysis information. A sample registration and anonymization of contents are executed by an anonymizing means at the time of receiving an analysis sample. A barcode input IF 10 is arranged at each step from process 1 to N of the analysis flow, and sample identification information is input. Based on the input identification information, the SOP of the work process and the next process name are displayed so that the work procedure can be confirmed. After the analysis result passes through the accuracy checking means 14 on the server 3 and the quality judging unit 15 by the data transfer unit 13, the non-defective product is registered in the database 11, and a re-reaction instruction 16 is issued for the NG product. A time-series characteristic change of the quality judgment result registered in the database is monitored by the device monitoring means 17. [Effect] Large-scale analysis information can be managed efficiently.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sample tracking, data processing and information management method in a large-scale analysis process such as a gene nucleotide sequencing process, and a processing step for implementing the method, which can be processed by a computer. Automation technology.

[0002]

2. Description of the Related Art For example, a pharmaceutical manufacturer inputs information on a manufacturing process by inputting a manufacturing lot identification code (bar code or the like) at each step of the manufacturing process of a drug and collectively managing it in a server. Management is realized. That is, when manufacturing tablets and the like, at each step of the process, the work process name, worker name, lot ID, etc. are collected from the bar code input interface device, the measurement data of the chemical raw material is collected by the automatic weighing device, etc., and registered in the database. By doing so, a system that enables tracing of manufacturing history and inventory management of raw materials has been built. In recent years, in the field of biotechnology as well, for the purpose of managing the analysis process,
Having the above equivalent function, LIMS (L aborato
ry I nformation M anagement
Experimental information management system called S ystem) has come to be built. In particular, in recent years, large-scale gene base sequencing has been carried out with the human genome project as an opportunity, and a new automatic base sequencer (sequencer) using separation by capillary (capillary) has been adopted. There is a need to manage and process large amounts of information. This device has 96 channels and can perform 96 nucleotide sequencing operations at the same time. Since one operation is completed in about 2 to 4 hours, 500 to 1000 sequences can be obtained per day. It is known as a multi-capillary sequencer (hereinafter, sequencer). The DNA clones to be analyzed were also 96 wells (8 x 12
High-throughput analysis is being promoted by using microtiter plates with lanes) or 384 wells (16 x 24 rows). Therefore, in addition to the analysis history of each well (working date, reaction conditions, reagent lot, worker name, relationship between clone and primer set), the above information includes downstream information processing (creation of contig, determination of total length, target) Cooperation with search (whether a gene is known or unknown) becomes important.

Regarding the method for determining the gene nucleotide sequence,
For example, it is described in detail in "Genome-Approach to New Life Information System" (TA Brown, published by Medical Science International, 2000).

[0004]

However, in order to realize high throughput of large-scale nucleotide sequencing centering on gene analysis and improvement of reliability of analysis results, the above-mentioned process control system alone is not sufficient. is there. For example, in order to increase the throughput of the determination of the nucleotide sequence of the DNA to be analyzed, it is important to improve the yield of each plate (clones successfully read = number of wells). When the sequence reading fails or the reliability of the read sequence is judged to be low, it is necessary to identify the target clone and perform reanalysis. In particular, when performing full-length gene analysis, the correspondence between each sequence and clone, the actual position on the plate, and the gene sequencing method suitable for each clone must be managed without fail.

However, in the past, such information is generally of the nature managed by work instructions and experimental notes, and if a defect occurs in the process of determining the nucleotide sequence of a gene, its cause Which process of analysis (reaction system,
It took an enormous amount of time to find out if it was in a sequencer, cluster of arrays, etc.).

On the other hand, in a large-scale analysis of a gene, in order to secure a plurality of migration paths, a large number (96 or more) of capillaries with an inner diameter of about 0.1 mm are arrayed in the Prism 3700 multi-capillary sequencer (US Pat. No. 072177 and Japanese Patent Laid-Open No. 06-138037) are the mainstream. Fluorescence emission wavelength bands do not overlap for DNA sequencing 4
DNA is labeled with more than one type of fluorophore. The excitation light of the phosphor is an Ar ion laser (wavelength 488 nm, or 515 nm).
nm) is used, and the excited fluorescence is spectrally detected by the fluorescence detection unit of the sequencer, which makes it possible to identify each of the four types of bases (ATCG). The fluorescence intensity signal data output for each base is called chromatographic data, and is output from the sequencer together with the sequence data obtained after the base call, but the detected fluorescence intensity is greatly affected by the reaction system and the detection system. , The so-called waveform quality may vary greatly depending on the base. In other words, it is necessary to check the analysis data for abnormalities due to sample adjustment failure or migration failure. Specifically, the operator must visually judge the clarity of the peak of the waveform corresponding to each base. The nucleotide sequence length obtained from the analysis data is about 500. Considering the case of cDNA, it is usually necessary to obtain 10 or more data before the whole sequence is determined. The peak needs to be inspected. In addition to being time-consuming, it is difficult to eliminate the possibility of overlooking an unclear peak due to individual differences in the standard of peak clarity.

The above work is not only time consuming, but also requires knowledge and experience from the worker. The present invention provides a system that automates such work and can reduce the labor of the worker and improve the sequencing throughput.

[0008]

[Means for Solving the Problems] Under such a purpose,
A first invention is an analysis information management system for managing information related to work in an analysis flow of processes 1 to N, and an information input means for inputting information related to the work, and the input information. An information registration means for registering in the database for each of the analysis target samples, and automatically detecting the end of the analysis process N, and outputting the analysis data for each of the analysis target samples output in the analysis process N to a data management device or Data transfer means for automatically transferring to the server, accuracy check means for automatically checking quality and accuracy of the analysis data transferred by the data transfer means, and evaluation result of quality and accuracy of analysis data by the accuracy check means Is registered in the process information database as the attribute data of the sample to be analyzed. And means, on the basis of the accuracy check means by analyzing data quality and accuracy of the evaluation results,
Based on the quality judgment means for judging the quality level of the analysis data and the quality judgment result of the quality judgment means, if the quality of the analysis data does not satisfy a predetermined standard, an instruction for re-reaction or analysis re-issue is issued. Re-reaction instructing means and device monitor means for supporting the monitoring of characteristic changes of the analysis device used in the analysis flow based on the evaluation result of the quality and accuracy of the analysis data obtained by the accuracy check means It is characterized by being provided.

The second invention is the same as the first invention,
The information input means is characterized in that the identification information on the sample in which the identification information is recorded in advance and the operator name are read and input.

The third invention is the same as the first invention,
The information management means reads the process names of processes 1 to N on the analysis flow, the worker name, the identification information of the sample to be analyzed, and the processing contents by the input means, and transfers the information to the data management device or the server. The process is then registered in the process information database.

A fourth invention is the same as the first invention,
The data transfer means includes means for monitoring whether or not a file of transfer target data has been created at regular intervals in the device that outputs the analysis data, and a data management device (if the transfer target file has been created). Or server)
It is characterized by including a communication means for confirming permission of transfer by mutual communication with and a means for restarting the client program.

A fifth invention is the same as the first invention,
The accuracy checking means is characterized in that after the evaluation target is narrowed down by the primary screening by the accuracy evaluation of the tax, the secondary screening by the dense evaluation is executed.

A sixth invention is the same as the first invention,
The accuracy checking means is characterized by focusing on a quality evaluation value of a signal waveform forming the analysis data.

A seventh invention is the same as the above-mentioned first invention,
The accuracy checking means is characterized by evaluating a change in the amplitude of a signal waveform forming the analysis data and classifying it as non-defective product data or defective product data.

An eighth invention is the above-mentioned first invention, wherein
The above-mentioned accuracy check means is characterized in that it is possible to identify an abnormality due to generation or mixing of bubbles from the signal waveform forming the analysis data.

A ninth invention is the above-mentioned first invention, wherein
The accuracy check means pays attention to the change in the amplitude of the signal waveform forming the analysis data, and sets the reference threshold value th1 set in advance.
If it is lower than the threshold value or higher than the reference threshold value th2, it is determined to be defective, and data classification means for classifying the analysis data into the evaluation reference threshold value is provided.

In a tenth aspect based on the first aspect, the accuracy checking means includes means for counting the number of unanalyzed data in the analysis data and comparing the count with an allowable threshold. Characterize.

An eleventh aspect of the present invention is the same as the first aspect of the present invention, wherein the quality judgment section is according to the sixth to ninth aspects of the present invention.
It is characterized in that whether or not the analysis result of the analysis target sample satisfies a predetermined quality standard based on the accuracy check result.

In a twelfth aspect based on the first aspect, the quality determination section according to the fifth to tenth aspects, wherein the analysis result of the sample to be analyzed satisfies a predetermined quality standard based on the accuracy check result. It is characterized by determining whether there is.

In a thirteenth aspect based on the first aspect, the quality determination section determines whether or not the analysis result of the sample to be analyzed satisfies a predetermined quality standard based on the accuracy check result according to the sixth aspect. It is characterized by

In a fourteenth aspect based on the first aspect, the quality determining means determines whether the analysis result of the sample to be analyzed satisfies a predetermined quality standard based on the accuracy check result according to the seventh aspect. It is characterized by

In a fifteenth aspect based on the first aspect, the quality determining means determines whether the analysis result of the sample to be analyzed satisfies a predetermined quality standard based on the accuracy check result according to the eighth aspect. It is characterized by

In a sixteenth aspect based on the first aspect, the quality determination means determines, based on the accuracy check result according to the ninth aspect, whether the analysis result of the sample to be analyzed satisfies a predetermined quality standard. It is characterized by

A seventeenth aspect of the present invention is based on the analysis flow according to the first aspect, wherein the re-reaction instructing means is based on a result determined by the quality determining means according to the eleventh to sixteenth aspects of the invention. It is characterized in that any one of the processes 1 to N instruct the re-reaction of the sample to be analyzed, which is determined not to satisfy the predetermined criterion, or the re-execution of the analysis.

In an eighteenth invention based on the first invention, the re-reaction instructing means is the eighth invention or the seventeenth invention.
In the quality determination means based on the invention, when an abnormality occurs in the analysis data due to generation or mixing of bubbles, maintenance means for instructing maintenance of the analysis device, and after performing maintenance of the analysis device, The sample to be analyzed is characterized by issuing a re-reaction instruction in process N.

In a nineteenth aspect based on the first aspect, the apparatus monitoring means obtains the accuracy checking means and registers the time-series quality and accuracy of the analysis data registered in the process information database. It is characterized in that it is a device monitor means for supporting the monitoring and maintenance of the characteristic change of the analysis device by evaluating the change.

A twentieth aspect of the invention is a sample anonymization means according to the first aspect of the invention, wherein the sample to be analyzed can be analyzed after anonymizing the attributes and contents of the analysis request source and the sample. It is characterized by including.

A twenty-first invention is the above-mentioned first invention, wherein in each process from the process 1 to the process N of the analysis flow, the identification information of the sample to be analyzed acquired by the information input means, It is characterized by comprising work instruction display means for retrieving the analysis flow information registered in the process information database and the work instruction of the process, and displaying the work instruction of the current process and the next process name.

A twenty-second aspect of the present invention is, in the above-mentioned first aspect, the amount of analysis target sample to be analyzed is budget-registered in advance in the process information database, and the analysis target sample judged to be non-defective by the quality judgment means. It is characterized by comprising a process progress management means for managing the progress of the analysis process by comparison with the quantity of.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, an analysis information management system according to an embodiment of the present invention will be described with reference to the drawings, taking a step of determining a nucleotide sequence of a gene as an example. FIG. 1 is a conceptual diagram showing the overall configuration of an analysis information management system according to the present invention. The main purpose of the analysis line in this step is to determine the partial or full-length nucleotide sequence of the introduced gene (analysis target sample).

First, after receiving a DNA sample or a DNA fragment for analyzing a nucleotide sequence into the analysis line (procedure for receiving 1), the name of the customer entrusted with the analysis and information related to the sample (clone name, quantity) Etc.) is anonymized by the anonymization means 2 when the sample is registered. The work associated with anonymization is performed under the authority of the person in charge of analysis line management or the person in charge of analysis project. More specifically, the customer name is converted into a customer identification code based on a designated code table, and is registered in the sample management database 4 on the data management device or the server 3 together with the sample-related information. Information related to the contents of the sample is registered in the database area that only the administrator or the person in charge of analysis has access authority, and then based on the prescribed encryption rule,
Issue a unique identification code for each sample and manage the database. In the present embodiment, the database group is centrally managed by the server 3. However, database management by a distributed server is also possible and not limited to centralized management.

In order to increase the throughput of the sample to be analyzed, it is generally loaded in a 96-well or 384-well plate in a dispensed manner and then put into the analysis work. Therefore, an identification code for each plate is added. The identification code for each sample or well shall be dependent on the identification code for each plate. Here, the identification code for each plate is attached to the plate as a barcode label.
In this embodiment, the identification code marking means is a bar code, but a method using an IC chip, a wireless IC tag, or the like may be used.
It is not limited to barcodes only.

Further, in the sample registration / anonymization means 2, a sample of contracts from a plurality of customers or a sample of a plurality of projects is stored in the above-mentioned customer name and contents information and storage.
It has the function of registering the database with the ID and managing inventory.

A work instruction sheet 5 is issued to a worker in charge of analysis by a higher-level manager or a person in charge of the analysis project. In the work order, the assignment of the worker, specification of the work target sample and quantity, and application protocol (reaction system) are described. Note that the work instruction creation function is configured as a WWW web server application that runs on the server 3 and has the function of searching for the project name, analysis target sample, worker name, and protocol name registered in the database. , Any client machine connected to the same network as the server 3 can access it. However, when inputting the work instruction data, the user name and password are input to authenticate the editing work.

The worker assigned by the above work instruction sheet is conscious of the customer name or the contents of the sample,
Based on the designated plate identification code, it is possible to proceed with the work according to a predetermined protocol.

The well-known DNA sequencing techniques are the shotgun method and the primer-walking method. The shotgun method uses long DNA such as genomic sequences.
It is known as a method for efficiently determining a sequence. On the other hand, the primer walking method is a technique that can reliably read a relatively short DNA sequence of about several thousand bases, and is used for gap closure during cDNA sequencing or genome sequencing.

The analysis flow from "plasmid extraction" 6 (process 1) to "sequencer" 9 (process N) shown in FIG. 1 can be applied to both the shotgun method and the primer walking method. However, the pretreatment associated with the extraction of DNA is omitted, and the sample to be analyzed is premised on a plasmid library.

The bar code input interface 10 is arranged at each step from process 1 to process N. This interface operates as a client program on a networked device or a personal computer, and constitutes a client / server system and a server program operating on the server 3. With this interface, the barcode attached to identify the plate is read by the barcode scanner. FIG. 2 shows a basic configuration diagram of the client program. The bar code input interface 10 interprets the input bar code information by the reading / interpretation unit 51, and then the database search function 52.
Via the server 3, the process information database 11 is searched to confirm that the sample is a registered sample and the current process name by the process information collating function 53. If the integrity of the registration information has been obtained, the corresponding standard work instructions related to the process of (SOP, S tandard O peration P
rocedures) and the next process name from the server. Function to display the next process name along with the received SOP
54 is displayed on the monitor screen by 54, and the confirmation of the analysis procedure by the operator is used. Similarly, if consistency is not obtained,
An error is displayed on the monitor screen 55 to that effect. If the process is executed normally, the termination status of the current process is registered56 in the process information database, and the process information of the sample is updated.

By the above means, it becomes possible for the worker to proceed with the analysis work without burden even if the work has a low experience value.

In the above description, the bar code input interface has a function of collecting not only the bar code of the plate but also information related to the work. In this specification, the information related to the work in the analysis flow means the process name, the worker name, the analysis target sample, and the processing content. The process name is configured so that the processes before and after it can be confirmed from the analysis flow protocol registered in the database in advance. For the worker's name, read the bar code attached to the name tag or employee ID card.

According to the above means, it is possible to manage the sample history information at each step, and sample tracking can be realized. Here, sample tracking is a system that makes it possible to comprehend which sample, when, in what state, and where it is, by creating a database. This allows the trace of the sample to be traced even after the analysis is completed.

DNA to be sequenced is plasmid extracted 6
(Process 1), after purification, a sample was prepared with a buffer solution and a fluorescent substance used for labeling, and then PCR (Polymerase Cha
in Reaction) 7 method (Process N-2)
After purification 8 (process N-1, excluding unreacted fluorophore), sequencer 9 (process N) is applied to obtain the base sequence. Here, the sequencer is a measuring device for determining the base sequence of a gene, and means a device for performing the gel electrophoresis separation of fluorescently labeled DNA and measuring the running DNA in real time.

The data transfer unit 13 uses the time point when the determination of the base sequence by the sequencer 9 is completed as a trigger for the server.
Transfer the data file to 3. For example, for one 96-well plate, the result file group for 96 samples will be transferred.

The data transfer unit 13 is attached to the sequencer
It consists of a client program that runs on a PC and a server program that runs on a file server. The client and server programs will be described with reference to FIGS. The transfer process is performed in the following procedure. (1) The client program monitors whether the transfer target file is created in the sequencer 9 at regular intervals 1
02. (2) When the file to be transferred has been created, the client program performs socket communication with the server program and sends the file transfer permission request 103. (3) Upon receiving the file transfer permission 204, the client program collects a plurality of files to be transferred as an archive (tar format) 105 into one file. (4) The client program transfers 105 the archive to the determined work area of the file server by using FTP. The client program that completed 106 FTP transfer is
By performing Socket communication, the FTP transfer completion message is sent 107 to the server program. (5) The server program that received the message of FTP transfer completion 205 expands the archive to the original file 206
To do. The server program will
After performing system-specific processing such as moving to a storage area, a message 211 indicating normal processing end is sent to the client program. Then, the transfer file name, date and time, etc. are recorded in the management file. The client program that has received the message 108 indicating that the normal processing has ended writes the transfer completion record 109 in the management file and returns to the processing of (1).

By the way, when a network failure or a file server failure occurs, the server program does not send a transfer permission message or a normal processing end message.
Alternatively, the processing is interrupted in the state of (5) and the subsequent file transfer is not executed. Therefore, by displaying the management file on the file server or the contents of the database to the operator via a web browser, the operator can confirm which file was transferred normally. When the client program is restarted after the network or file server failure is resolved, the transfer process restarts from (1). However, since there are a plurality of sequencers and a plurality of client PCs operate at the same time, it is necessary to activate a plurality of client programs at the same time. To facilitate this start-up work,
FIG. 5 shows a mechanism for automatically restarting a file server by sending a restart message to a plurality of client programs via the network. This mechanism is realized by the following method. (1) On the client PC, start 301 the program for starting the client program. This work is done manually once at the start of operation. (2) The startup program starts the client program as an external program when itself starts. (3) When an error occurs and it is necessary to restart the client program, the file server sends a restart message to the client PC. (4) Upon receiving 302 a restart message from the server program, the startup program forcibly terminates the client program that is running and then re-executes it again.

The data transfer unit 13 is configured to transfer the analysis result from the sequencer 9 to the server 3 by the above mechanism. Conventionally, a researcher or a worker collects analysis data from each sequencer via a medium such as an MO disk, but by this means, the analysis data of the sequencer is transferred to a server and is centrally managed. Therefore, the work efficiency can be improved.

On the other hand, the data transferred from the sequencer includes the chromatographic data (fluorescence intensity signal) output for each of the four types of bases (ATCG) and the base sequence data. Regarding sequence data, it is necessary to check whether there are any abnormalities due to sample preparation failure or electrophoresis failure. This is automatically executed by the following means. First, the accuracy check unit 14
After performing a quality check focusing on the waveform amplitude (intensity) of the chromatographic data and the clarity of the waveform and evaluating the presence / absence of bubbles, the quality was judged by the plate unit as the primary screening15, and as a result, it was judged as a good product. The secondary screening was performed for each sample (clone), and the quality was determined15. When the product becomes NG, the judgment of the re-reaction instruction 16 is made. These series of processes are automatically executed in the server.
As will be described later, the quality judgment result and the re-reaction instruction can be referred to by the browser from the client PC 17 via the web server.

The processing relating to accuracy check, quality judgment and re-reaction instruction will be described below with reference to FIGS. 6 to 9. In FIG. 6, the average value of the waveform signal amplitude of the chromatographic data of each sample (clone) is examined. That is, if the average value of the intensity is between the lower limit threshold Th1 and the upper limit threshold Th2, it is determined to be a non-defective sample 403, and if not, 404 is determined to fall below the lower limit threshold Th1 and above the upper limit threshold Th2. After classification 405, it is counted as the number of NG products. FIG. 7 shows the relationship between the chromatographic data 450 corresponding to an arbitrary base and the above threshold value. In addition, Table 451 in the figure shows an example of a waveform in which a problem occurs in adjusting the sample amount. The above is repeated for the data for one plate. The quality determination unit obtains the ratio between the number of samples (number of NG products) that finally became NG and the total number of samples on the plate, and compares 407 with a preset NG allowable rate. When the ratio of the number of NG samples is equal to or more than the allowable rate, re-migration by the sequencer (process N) or re-analysis (process 1) can be displayed together with the determination result in the re-reaction instruction. These results can be confirmed with a web browser.

On the other hand, there are various means for evaluating the waveform quality. For example, Phred (http://www.phrap.com) developed at Washington University in the United States and Paracel (ht
Software such as TraceTuner from tp: //www.paracel.com/tracetuner/) is commercially available. These software not only provides nucleotide sequences but also QV accuracy information for each nucleotide.
(Quality Value) Value can be output. The QV value is based on multiple parameters such as the distance between peaks, whether or not the peaks corresponding to adjacent bases of the same type are clearly separated, the signal ratio between peaks adopted as bases and those not, and actual data. It is a value calculated based on the relationship with the error rate of the base in. As a general standard by this software, it is said that the reliability of the obtained base is sufficiently high if QV ≧ 20, but the standard for comprehensive evaluation of sequences is not constant.

In this example, the QV value for each base of each sample was obtained by the above software, and based on this result,
It is assumed that the accuracy check of array data will be performed. This will be described below with reference to FIG. A window (block) with a certain width is set on the base sequence for quality evaluation. For example, find the average QV value within a 20-base wide window and compare it with a preset threshold.
502, and if it is less than or equal to the threshold value, it is regarded as an NG block. While shifting this window by the unit of window width on the base sequence, repeat the above calculation until the end of the data
Is counted. The obtained number of NG blocks is compared with the Block threshold, and the sample is judged as non-defective or N
Judge. NG items are counted as NG item number 504. The above calculation is repeated for the data for one plate, and the quality judgment unit 505 checks the ratio of the number of NG products and the sum of the number of NG products to the total number of samples and compares 506 with the NG allowable ratio. If it is higher than the allowable rate, give instructions for re-reaction on a plate-by-plate basis. For re-reaction instruction, re-migrate by sequencer (Process N) or redo analysis (Process 1)
Can be displayed together with the above determination result. These results can also be confirmed with a web browser.

In the present embodiment, as described above, the multi-capillary sequencer manufactured by ABI, Inc. in the United States is taken as an example. However, air bubbles may sometimes be generated or mixed in the capillaries. For example, as shown in FIG. 9, on the waveform of the chromatographic data of the fluorescence intensity (intensity) detected for each base, the peak of the waveform corresponds to each base and can be clearly identified in the normal region. A peak is observed at the site where bubbles are generated when four bases are co-located. For this reason,
The base cola cannot detect a significant difference, making it impossible to identify the base. In this case, the unreadable base is expressed as N. Although there are other factors that cause N for reading, in the case of the above, due to diffuse reflection on the bubble surface and inside of the excitation light, simultaneous and equivalent level responses appear on the detector corresponding to the fluorescent label of each base. It is believed to be due. Therefore, if the bubbles can be identified,
This is extremely effective because problems can be solved by maintaining the sequencer such as removing bubbles.

In the present invention, by utilizing the above physical phenomenon, NG
The cause of the product is identified and then fed back to the process. That is, the projection (arithmetic mean) of the 4-base chromatographic data on the same coordinate (base position) is obtained. By this calculation, the place where the bubble is generated becomes visible.
A fixed threshold value or a one-dimensional noise detection filter as shown in Fig. 10 can be used to detect the position coordinates of the bubbles. Especially in the latter case, when the offset due to the DC component is locally generated on the chromatographic data, this effect can be reduced, so that the bubbles can be detected stably. That is,
With respect to the projection data, an average value in the range of ± 2 to ± n centering on the projection value at a certain point of interest is obtained. This average value is compared with the projection value of the point of interest, and if it is greater than or equal to a preset threshold value, it is determined to be a bubble. The reason why the dead zone region of ± 1 is provided in this filter is that it is assumed that the bubbles extend over a plurality of bases. The dead zone region range or the averaging range is not limited to the filter shape shown in FIG. 10, and may be adjusted according to the reaction system or phenomenon.

FIG. 11 shows a display example of the result of the processing up to the accuracy check unit 14, the quality judgment unit 15, and the re-reaction instruction 16.
This display example can be displayed via a web browser using any client PC 17 on the same network as the server.

The means for instructing re-reaction on a plate-by-plate basis have been described above, but these are positioned as the primary screening quality judging means, and their purpose is to efficiently exclude those clearly judged to be defective.
After the plate is judged as acceptable, it is effective and necessary to carry out quality control by secondary screening such as sequence read length in sample (clone) units. Hereinafter, the steps of the accuracy check 600 and the quality judgment means in sample units will be described with reference to FIG. 1) Count 602 the number of bases that cannot be read (indicated by N) in the base sequence data. The count starts from MinBase specified in advance. 2) The number of N counted in the set window (N_
Calculate the position (MaxBase) of the window when count) exceeds the set threshold (N_th). 3) Find the read length L (= MaxBase-MinBase) of the array,
It is compared with a threshold value (SeqLength) set in advance, and if it is equal to or more than the threshold value, the non-defective sample data 503 is set. Otherwise, it is determined to be NG product 504.

In the above, the accuracy check and the quality judgment are automatically executed. Regarding the result, for example, it is possible to provide a function of monitoring the waveform of the chromatographic data of the sample which has become an NG product. Is.

The accuracy check judgment value obtained for each sample is registered as attribute information of the sample corresponding to the process information database 11. This data is used by the device monitor means 17 described below.

As a basic idea, the ratio of Good base defined by the following algorithm in the readable sequence length of the base sequence of the gene is calculated and displayed in time series. The steps of the algorithm are as follows. 1) For the QV value data, the difference between the first and last position coordinates that satisfies QV ≧ R is the “read length”. Here, R is a lower limit allowable value of the preset QV value. 2) Count the number of bases where QV ≧ G, and use this as “Go
od base number ", the read length is L or more,

[0058]

[Formula 1] And calculate the average around the plate. 3) Registration in the database Register the sequencer ID used, the date and time of data transfer, the process related information such as the plate name (bar code) and the plate average ratio. In addition, maintenance records such as the date and time of replacement of the capillaries of the sequencer and the date and time of nitric acid cleaning are also registered. 4) Viewer function With the viewer function shown in Fig. 13, it is possible to obtain a guideline for determining the maintenance schedule by referring to the graph that plots the "average plate ratio" and the number of days elapsed after the replacement of the capillary. To do.

The apparatus monitoring means of this embodiment can easily and automatically obtain a standard for determining the maintenance scheduling related to the sequencer. Although the present embodiment is directed to the sequencer, it is possible to obtain a guideline for device maintenance by monitoring the change over time in the analysis result by the device on the analysis line based on the same idea as above.

Further, in the system of the present invention, when registering customer and sample information in the database, the planned analysis quantity of the analysis target sample is registered in advance in a budget, and the analysis target sample judged to be non-defective by the quality judging means. It is possible to construct a process progress management means for managing the progress of the analysis process by comparing with the quantity. With the above invention, the person in charge of analysis and the operator can automate the entire process from the end of analysis of the sequencer to the determination of re-reaction, which has required a huge amount of data processing time. In other words, when the sequencer analysis is completed, the accuracy check result is output in almost real time,
The presence or absence of a re-reaction instruction for the plate to be analyzed can be confirmed on the server.

[0061]

As described above, according to the present invention, a bar code input interface mounted on a networked device or a PC is arranged at each step of the analysis flow from process 1 to process N. so,
It is possible to manage the information on the processing history of the sample to be analyzed. Further, in each step, the work procedure can be confirmed by simply reading the identification code of the sample, and the processing flow can be surely advanced.
Further, since the analyzed data is automatically transferred to the server, the accuracy is checked, and the product is classified as a non-defective product or an NG product, it is possible to save the trouble of collecting the data from the devices scattered on the analysis line. According to the present invention, with respect to a plate that has become NG, an instruction for re-reaction (re-migration in the sequencer) or re-execution of analysis is issued, but the manager or the responsible person makes an original judgment on this result and It is also possible to instruct the operator. On the other hand, in the system according to the present invention, it is also possible to integrate the samples or clones that have been determined to be NG and integrate them into a new plate for analysis. Needless to say, it is easy to manufacture a reorganization tool for NG product data because all the information required for that purpose is registered in the database.

In the case of performing full-length analysis of genes, it becomes possible to manage the relationship between each sequence and clone, the position on the actual plate, the information on each clone, primer set, and reaction system. . Up to now, the analysis data has been visually checked to see if there are any abnormalities due to sample adjustment failure or migration failure. However, the present invention allows automation, and a certain standard that does not depend on the subjectivity of the operator. It is possible to evaluate the quality.

The present invention makes it possible to automate the analysis work and reduce the labor of the worker and improve the sequencing throughput.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of an analysis information management system to which the present invention has been applied.

FIG. 2 is a diagram showing a configuration of a barcode input interface in the analysis information management system.

FIG. 3 is a diagram illustrating a flow of a client program of a data transfer unit according to the embodiment of this invention.

FIG. 4 is a diagram illustrating a flow of a server program of a data transfer unit according to the embodiment of this invention.

FIG. 5 is a diagram illustrating a flow of a program for automatically starting / stopping a client program according to a message from a server according to the embodiment of this invention.

FIG. 6 is a diagram for explaining accuracy check, quality judgment and re-reaction instruction means focusing on the waveform amplitude in the embodiment of the present invention.

FIG. 7 is a diagram for explaining an accuracy check logic focusing on a waveform amplitude and an example of classified NG products in the embodiment of the present invention.

FIG. 8 is a diagram for explaining accuracy check, quality judgment and re-reaction instructing means focusing on the waveform quality evaluation value (QV) in the embodiment of the present invention.

FIG. 9 is a diagram illustrating a bubble detection unit that specifies an NG array caused by an abnormality such as bubbles mixed in a capillary of a sequencer according to an embodiment of the present invention.

FIG. 10 is a diagram illustrating the logic of the one-dimensional filter that identifies an abnormality caused by bubbles in the embodiment of the present invention.

FIG. 11 is a diagram illustrating a display example of a quality determination result in the embodiment of the present invention.

FIG. 12 is a diagram illustrating a flow of accuracy check for each sample in the secondary screening in the example of the present invention.

FIG. 13 is a diagram illustrating a viewer and a maintenance support method in the device monitoring means of the present invention.

[Explanation of symbols]

3 ... Data management device or server 4 ... Sample registration database 6-9 ... Flow of gene nucleotide sequence determination work 9 ... Sequencer (electrophoresis device) 11 ... Process information database 14 ... Accuracy check section 15 ... Quality judgment unit 16 ... Re-reaction indicator 17 ... Client PC 51 ... Bar code reading / interpretation unit 52 ... Database search unit 53 ... Process information collation unit 56 ... Database registration unit.

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Haruomi Kobayashi             4-6 Kanda Surugadai, Chiyoda-ku, Tokyo             Hitachi, Ltd. Life Science Promotion             Within the business unit (72) Inventor Hiroya Koshiba             4-6 Kanda Surugadai, Chiyoda-ku, Tokyo             Hitachi, Ltd. Life Science Promotion             Within the business unit (72) Inventor Kazuyuki Fukuda             5-22-1 Kamimizuhonmachi, Kodaira-shi, Tokyo Stock             Ceremony Company Hitachi Cho-LS System             Within (72) Inventor Ayako Wakabayashi             5-22-1 Kamimizuhonmachi, Kodaira-shi, Tokyo Stock             Ceremony Company Hitachi Cho-LS System             Within (72) Inventor Kazue Otsuka             5-22-1 Kamimizuhonmachi, Kodaira-shi, Tokyo Stock             Ceremony Company Hitachi Cho-LS System             Within F term (reference) 2G045 AA35 DA12 DA13 DA14 FB05                       FB13 GC15 HA12 JA01 JA04                 4B029 AA07 AA21 AA23 BB20 CC03                       FA15                 4B063 QA01 QA13 QA18 QQ02 QQ05                       QQ41 QR06 QR20 QR31 QR38                       QR41 QR55 QR59 QR80 QR82                       QS11 QS24 QS25 QS28 QS34                       QS39 QX01 QX07                 5B056 BB62 HH01                 5B075 ND20 UU19

Claims (22)

[Claims] 1. An analysis information management system for managing information related to work in an analysis flow of processes 1 to N, comprising an information input means for inputting information related to the work and the input information. Information registration means for registering in the database for each of the analysis target samples, and automatic detection of completion of the analysis process N, and analysis data for each analysis target sample output in the analysis process N is a data management device or server. A data transfer means for automatically transferring the data, an accuracy check means for automatically checking the quality and accuracy of the analysis data transferred by the data transfer means, and an evaluation result of the quality and accuracy of the analysis data by the accuracy check means. , Quality information registration to be registered in the process information database as attribute data of the sample to be analyzed The quality of the analysis data based on the evaluation result of the quality and accuracy of the analysis data by the accuracy check means, and the quality of the analysis data based on the quality judgment result of the quality judgment means. Is not satisfied with the predetermined criteria, the re-reaction instructing means for issuing an instruction for re-reaction or re-execution of the analysis and the analysis flow based on the quality and accuracy evaluation results of the analysis data obtained by the accuracy checking means. An analysis information management system, comprising: a device monitor means for supporting the change in the characteristics of the analysis device used in 1. 2. The analysis information management system according to claim 1, wherein the information input means reads and inputs the identification information on the sample in which the identification information is recorded in advance and the operator name. 3. The information management means reads the process names of processes 1 to N on the analysis flow, the worker name, the identification information of the analysis target sample, and the processing contents by the input means, and the data management device or the server. 3. The analysis information management system according to claim 1, wherein the information is transferred to and registered in a process information database. 4. The data transfer means, in the device for outputting the analysis data, means for monitoring whether or not a file of transfer target data has been created at regular intervals, and when the transfer target file has been created. 2. The analysis information management system according to claim 1, further comprising communication means for confirming permission and approval of transfer by mutual communication with a data management device (or server), and means for restarting a client program. 5. The accuracy check means is a two-step accuracy check means for executing a secondary screening by a dense evaluation after performing a narrowing down of the evaluation target in the primary screening by a precision evaluation of the tax. The analysis information management system according to claim 1. 6. The analysis information management system according to claim 1, wherein the accuracy check means is an accuracy check means focusing on a quality evaluation value of a signal waveform forming the analysis data. 7. The accuracy check means evaluates a change in amplitude of a signal waveform forming the analysis data and classifies the data as good product data or defective product data.
Analysis information management system described. 8. The analysis information management system according to claim 1, wherein the accuracy check means is capable of identifying an abnormality due to generation or mixing of bubbles from a signal waveform forming the analysis data. 9. The accuracy checking means pays attention to a change in the amplitude of a signal waveform forming the analysis data, and determines that the accuracy is lower than a preset reference threshold th1 or higher than a reference threshold th2. The analysis information management system according to claim 1, further comprising data classification means for classifying the analysis data into the evaluation reference threshold value. 10. The accuracy check means counts the number of data which cannot be analyzed among the analysis data,
The analysis information management system according to claim 1, further comprising means for comparing with an allowable threshold value. 11. The quality determination unit determines whether the analysis result of the analysis target sample satisfies a predetermined quality standard based on the accuracy check result according to any one of claims 6 to 9. The analysis information management system according to item 1. 12. The quality determination unit determines whether the analysis result of the analysis target sample satisfies a predetermined quality standard based on the accuracy check result according to any one of claims 5 to 10. The analysis information management system according to item 1. 13. The analysis according to claim 1, wherein the quality judgment unit judges whether or not the analysis result of the analysis target sample satisfies a predetermined quality standard based on the accuracy check result according to claim 6. Information management system. 14. The analysis according to claim 1, wherein the quality judgment means judges whether or not the analysis result of the sample to be analyzed satisfies a predetermined quality standard based on the accuracy check result according to claim 7. Information management system. 15. The analysis according to claim 1, wherein the quality judgment means judges whether or not the analysis result of the sample to be analyzed satisfies a predetermined quality standard based on the accuracy check result according to claim 8. Information management system. 16. The analysis according to claim 1, wherein the quality determination means determines whether the analysis result of the analysis target sample satisfies a predetermined quality standard based on the accuracy check result according to claim 9. Information management system. 17. The re-reaction instructing means determines, based on the result determined by the quality determining means, that one of the processes from process 1 to process N on the analysis flow does not satisfy a predetermined criterion. The analysis information management system according to claim 1, wherein the analysis information management system is configured to instruct a re-reaction of the sample to be analyzed or a re-execution of the analysis. 18. The re-reaction instructing means, in the quality judging means according to claim 8 or claim 17, when an abnormality occurs in the analysis data due to generation or inclusion of bubbles, 2. The analysis information management system according to claim 1, wherein a maintenance means for instructing maintenance and a re-reaction instruction in the process N for the sample to be analyzed are issued after the maintenance of the analyzer. 19. The apparatus monitor means obtains the accuracy check means and evaluates a time-series quality and accuracy change of analysis data registered in the process information database, thereby improving the accuracy of the analysis apparatus. 2. The analysis information management system according to claim 1, wherein the analysis information management system is device monitor means for supporting monitoring and maintenance of characteristic changes. 20. The sample to be analyzed is provided with a sample anonymizing means for enabling analysis after anonymizing the attributes and contents of the analysis request source and the sample. Analysis information management system described. 21. In each process from the process 1 to the process N of the analysis flow, the identification information of the sample to be analyzed acquired by the information input means and the information of the analysis flow registered in the process information database. 2. The analysis information management system according to claim 1, further comprising: a work instruction display means for retrieving the work instruction of the process and displaying the work instruction of the current process and the next process name. 22. The amount of analysis target sample to be analyzed is budgeted in advance in the process information database and compared with the quantity of the analysis target sample which is judged to be non-defective by the quality judging means, to thereby advance the analysis process 2. The process progress management means for managing
Or the analysis information management system described in 2.