Abstract
Single-nucleotide polymorphism (SNP) genotyping is a fundamental tool in the rapidly growing area of complex diseases and pharmacogenomics. SNP patterns that correlate with disease or response to treatment, respectively, are identified using bioinformatic techniques. We present an integrated laboratory information and management system (LIMS) for our high-throughput TaqMan™-based SNP genotyping platform. Three new client tools (ProjectManager, AssayManager, OrderTool) for our LIMS improve quality control and workflow management. The programs support organizing multiple genotyping experiments as projects, managing reagents with barcodes, and automation of assay ordering. The tools are freely available at our homepage.
Introduction
At the Kiel location of the German National Genotyping Platform, we use high-throughput single-nucleotide polymorphism (SNP) genotyping based on TaqMan technology and bioinformatic processing to identify SNPs that correlate with complex phenotypes. 1 Polygenic disorders show an unprecedented large diversity regarding the genetic susceptibility factors involved. The NOD2 gene in Crohn's disease illustrates this. 2 –4 NOD2 explains about 5%-10% of all Crohn's disease cases that are homozygotes or that show compound heterozygosity. Another 20%-30% of cases show influence of NOD2 in addition to other, not yet identified disease genes. The remainder of cases probably have to be explained completely by other disease genes than NOD2.
The necessity for high-throughput SNP genotyping is large even after discovery of a first disease gene. 5 High-density, genomewide association studies, which are expected to be conducted in polygenic conditions in the near future, will have an even greater demand for large-scale SNP genotyping. Although the typical flow in a platform laboratory is between 20 and 100,000 SNP genotypes per day, this may increase exponentially within a few years to several million per day.
We have developed an integrated database with client applications to facilitate lab processes and analysis of genotyping data. 6 This laboratory information and management system has evolved since 1998 and supports data import from TaqMan files, checking for Mendelian inheritance, dataexport, management of pedigree and phenotype information, and data analysis.
Here we report new client tools for our database that improve quality control, management of laboratory workflow, and handling of complex genetic data. The tools have been written in Visual Basic 6 and were implemented on Microsoft Windows NT 4 systems. They require a Microsoft SQL Server 7 database. Although conceptually evolved in a molecular environment that uses TaqMan as a typing technology, the programs can be adapted to other molecular technologies.
ProjectManager
The database includes a plate-tracking mechanism that records the following processing steps of a TaqMan genotyping experiment: reaction setup of plates, polymerase chain reaction, TaqMan data import, and Mendel checking. A scientist can create a project by compiling a list of assays and a list of plates. ProjectManager shows scientists and technicians the processing status of each individual 96-well or 384-well plate for each assay of a project in a table (Fig. 1). The table also shows whether genotypes already exist for a variant and whether the assay has failed. Scientists may prioritize the assays in a project, and technicians may enter assay-specific comments. Projects mainly serve as work plans for technicians.
Screenshot of ProjectManager. A project consists of a list of assays and a list of plates (which we call a plate set). The table shows the color-coded processing status (P = pipetted, I = imported (TaqMan data), F = finished) for each assay on each plate. Discarded tubes are marked with a “D.” The column labeled “gt” shows whether any genotypes already exist for the variant of the selected assay.
ProjectManager also includes an analysis and review tool for multiple projects for scientists, called project overview. Allele distributions for each assay on each plate can be demonstrated in an overview. This is an important quality control feature for detecting abnormalities very early in the genotyping process. A color-coded display shows which plates have not yet been genotyped and which plates have abnormal allele distributions (Fig. 2). Scientists can create a locus map necessary for data analysis based on the assays of an overview.
Screenshot of ProjectManager. Scientists can compile multiple projects to a project overview that color-codes the allele distributions for each assay on each plate.
AssayManager
The main feature of this tool is managing barcodes of tubes containing specific assay reagents. We have developed a private seven-digit barcode consisting of two digits for the chromosome, a four-digit running number, and a digit indicating the individual tube number inside the set of identical tubes with reagents for this assay. Technicians use AssayManager to import electronic assay datasheets, print barcode labels, and validate tubes by associating a barcode with an assay (Fig. 3). When a tube is scanned, information about the tube, assay, and contained oligonucleotides is shown. Users can also search available assay tubes by the genetic locus and its variants that are targeted to see whether an assay is already in house.
Screenshot of AssayManager. Technicians use this tool to print barcode labels. After choosing an assay source and entering a chromosome, the program generates new barcodes and prints them on adhesive labels.
OrderTool
OrderTool automates the ordering process of assays. This requires the following steps: assembly of the flanking regions for an SNP, quality assurance, and preparation of a machine-readable text file. Users can provide lists of “rs,” “ss” or “tsc” numbers for existing SNPs or sequences for newly designed assays. OrderTool also crawls through public databases to extract the flanking regions for SNPs with existing assays into our database. For quality assurance, OrderTool cross-blasts targeted sequences against preordered sequences to avoid ordering assays twice. OrderTool masks sequence repeats to ensure that assay oligonucleotides for the targeted SNPs are not located in a highly repetitive region that would lead to failure of an assay. After each step, OrderTool suggests an updated list of candidate SNPs (Fig. 4). In a final step, OrderTool creates a machine-readable text file that the user can e-mail directly to the vendor to initiate design of assays and the manufacturing and synthesis process.
Screenshot of OrderTool. The program shows a list with details for each item of an SNP order.
Conclusion
Interaction among the client applications enhances quality control and workflow management: OrderTool creates new variants for ordered assays in the database, which Assay-Manager checks when importing an electronic datasheet. ProjectManager generates a list of barcodes for those assays. Users scan assay tubes before a genotyping experiment to ensure that the assays are the correct ones for the project. Because SNP genotyping with TaqMan and similar PCR-based techniques is a widely used technology, our software may be of interest to other groups doing similar research.
Availability
Executables and source code of client applications and SQL scripts are freely available at http://www.mucosa.de/neu/index_bioinformatics.html.
Acknowledgments
The authors thank Tim Lu for contributing to the database and Birthe Petersen and Tanja Wesse for critical discussion and practical testing. This work was sponsored by grants from the National Genome Research Network (NGFN) and other public sources