|
|
 |
 |
|
|
| PRIME is an acronym for Proteome Research Information Management Environment. This software is built on a distributed open architecture enterprise platform. It is used in information management, tracking and analysis for proteomics research. PRIME has several major components. These include a laboratory information management system (LIMS), viewers for 2D Gel images and mass spectrometry plots, a streaming engine for real time data input from state of the art spectrometers, including both MS and MS/MS data, an automated protein database search engine for both Peptide Mass Fingerprint and MS/MS peptide fragment searches, and a data discovery toolkit. PRIME stores the significant experimental data, and search result data in a relational database for use in data mining. PRIME is highly scalable, secure, and extendable due to its distributed architecture. |
|
Click above to go to the
public entry site for
PRIME alpha version 1.66.02. |
|
|
|
About PRIME
LIMS
The LIMS portion of the system enforces the secure entry of experiment related information to the PRIME database. This includes organization, project, experiment, sample and researcher data. In addition, SOPs (Standard Operating Procedures) can also be entered. Samples are tracked through various stages of analysis, by using barcodes at each physical stage of the experimental lifecycle of the sample. The full experimental lifecycle of the sample starts with the arrival of the sample, then proceeds through gel creation, well plate creation, and mass spectrometry target plate creation. All of these physical objects can be archived as required. Samples, 2D gels, well plates and target plates, as well as archive areas are all uniquely bar-coded. The stations at which the sample is handled are also monitored. This process ensures that the entire life cycle of a sample is traceable. During image analysis of gels, the images are stored in multiple formats, initially as TIFF files and JPEG's. These files, along with detected spot information, are uploaded via file interfaces to the PRIME database. Merged gel image information resulting from merging multiple gel images can also be stored in the PRIME database. This allows for statistical analysis of gels and their spots.
Sample Processing Stages
Sample Data Entry
At the sample check in station, sample data can be entered using the PRIME sample check-in module. A barcode is assigned and placed on the sample during this process. If the sample information is entered remotely using the sample entry web page, then the barcode associated with it is placed on the sample when it arrives at the check in station.
Gel Preparation Station
A portion of the sample is used to prepare a 2D Gel at this station. The PRIME gel preparation software module handles the data collected in the process. Information generated from the IEF and SDS processes is stored in the database. The 2D Gel is bar-coded, with the barcode being incorporated in the gel. Both the sample barcode and gel barcode can be linked manually or via scanning.
Image Analysis Station
Here the gel is scanned to obtain the image and the image is processed using commercial software to identify spots. These spots and the gel images are uploaded to the PRIME database. The process is handled using the PRIME image analysis management software module. The gel image embedded barcode is scanned or manually entered to associate the gel with the image.
Spot Picking Station/Target Plate Station
The PRIME spot picker module supervises the process here. At this station a robot is programmed to pick the spots and place them in a well plate. Another robot then moves the well contents to a target plate. Both the well plate and the target plate are bar-coded so they can be tracked through the rest of the process. The sample barcode and well plate are scanned to link them or manually linked, and the well plate and target plate barcodes are scanned to link them or manually linked.
Mass Spectrometry Station
Here the PRIME mass spectrometry supervisory module monitors the process. The target plate barcode is scanned to identify the material that is being analyzed. M/Z and intensity results information, peak list information, MS station information, and target plate spot location information are stored in the PRIME database.
Proteomics Analysis Module
Both Peptide Mass Fingerprint (PMF) and MS/MS Peptide Fragment data extracted from mass spectrometer runs are stored in PRIME. Using mass spectrometry peak lists as input public proteomic databases are used to identify possible proteins and peptides in the case of PMF and peptides and their fragments in the case of MS/MS. Probability information is related to the accuracy of the estimation. This information is stored in PRIME, along with the identity of software used for the identification. This data is stored in such as way as to allow for easy data mining.
At this point it can be seen that PRIME allows tracking of the sample through to the Protein ID by linking information at each stage of the process, including which equipment, researchers, and operating procedures were involved at each stage.
PDA Usage with Bar-coding System
To allow for convenience of use PDA's (Personal Digital Assistants) will be used with barcode scanner attachments to do the bar code scanning. This will allow for easy scanning at any archive area. The PDA's will be connected wirelessly using the IEEE 802.11b protocol, to the local area network. Some fixed stations will also have barcode scanner hardware attachments.
PRIME 2D Gel Viewer
The viewer portion allows the gels to be viewed and annotated in a variety of ways. Relative intensity analysis results can be viewed within a gel and differential analysis done between gels. Spot lists can be viewed and modified or new spot lists created. A user can select additional spots for evaluation. Similar viewers for 1D gels and for LC MS data have also been developed.
Mass Spectrometry Analysis and
Real Time Data Streaming
Data streaming modules allow the incorporation of mass spectrometry data to the Prime database. The data generated by state of the art mass spectrometry machines is automatically analyzed and these results are imported into the Prime database.
Data Discovery Portion
Modules for advanced analysis of proteomics data will be developed in Phase 3. These will allow the user to build dynamic queries via keyword selection or to select from standard queries to run.
Software Architecture
The software architecture is based on a structure similar to Java 2 Platform, Enterprise Edition (J2EE). The layers of the system architecture are shown in the figure below. Specific terms related to the figure are specified below it. One of the goals of PRIME is to allow for an extensible architecture. The system is designed so other development groups can plug modules into it while only having to divulge a software client interface. In this case no software teaming agreement would need to be obtained. Tabs are used on the user interface of a client. This allows another software group to have their own screen real estate through a tab. A messaging interface will be used to provide services to their tab. Since the client operates via remote interfaces to communicate with the web layer, a partnering group can interface with their own web layer and database without affecting the PRIME database architecture or contents. At the same time they can obtain access to prime information via the PRIME interface layer based on the permissions as users. In addition the software is designed to not use an application server. This was done to make it vendor independent. Hence although PRIME is currently implemented on Oracle 9i, it can easily be ported to support another relational database such as IBM DB2.
Code Generator
A code generator is used to generate Java data class files mirroring the database schema. These are utilized by the data transport mechanism implemented by the session objects. Since these objects support the same common functionalities, the servlets can support the presentation layer in a generic way. Also when new tables or views are incorporated, the supporting base code can be generated, thereby reducing development time. Specialized servlets still require hand coding. Use of the code generator compensates for not using a vendor specific Application server. More specifically it provides an object relational mapping, so that programmers working outside the session layer can use an object approach. Generating these class files during build time allows for enhanced run time performance of the web server.
Top of page |
|
PRIME
Links
LIMS
Sample Processing Stages
Sample Data Entry
Gel Preparation Station
Image Analysis Station
Spot Picking Station/Target Plate Station
Mass Spectrometry Station
Proteomics Analysis Module
PDA Usage with Bar-coding System
PRIME 2D Gel Viewer
Mass Spectrometry Analysis and Real Time Data Streaming
Data Discovery Portion
Software Architecture
Code Generator |
| |
|
|
|
|
|
|
|
