About: LIMS

The Laboratory Information Management System was custom-developed over a period of several years and has been designed around high-throughput chromatin structure analysis. An overview diagram is provided in Fig. R2. We can easily clone and modify this system to incorporate the requirements of any high-throughput epigenetic profiling assays including DNA methylation.

Schematics of LIMS

The system was developed in-house using a modular design so that additional functionalities and other assays can be added on-demand in a seamless manner. Samples enter the LIMS at any stage of the experimentation and are incorporated to track the samples and their associated data.

Figure 1.
  1. The properties of each constituent portion of the laboratory samples are entered into the LIMS by laboratory personnel and subsequently archived in a database.
  2. Sample QC data including QPCR and nCounter from NanoString Technology produce quality data for each sample passing through the pipeline. The metadata appropriate to a collection of reaction plates as well as a map of information contained in the wells on each plate, and associated levels of fluorescence is stored in database.
  3. Sequencing platforms including Illumina HiSeq 2000, and GAIIx; and ABI 5500xl SOLiD are used for generating the sequence data that are analyzed for quality by computing SPOT values for each sequenced sample. Data types generated by specific assays, for example,
  4. DNase I mapping;
  5. RNA-seq; and
  6. ChIP seq. Details regarding a specific sequencing run are archived in the central database.
  7. The database and visualization engine (browser) provide controlled storage and access to the intermediate and final output produced by each particular experiment, for use in quality control, ad-hoc analysis, and publication.

The core of the LIMS system is a programmatic map of laboratory workflow. Essentially, this means that various phases of the cell/tissue sample preparation pipeline, quality control procedures, and information regarding other laboratory processes are stored in a database with a user interaction layer that enables report generation and sample tracking on a monthly or daily basis.

At any given time, laboratory personnel can query the LIMS to see the state of a given experiment, and analyze results of prior experiments. Laboratory scientists have access to the LIMS via dedicated benchtop PCs, and also via any desktop or instrument PC in the lab. The LIMS has a graphical user interface to facilitate user input. Data is entered at each step in a pipeline process. Some data entry is manual, other data entry utilizes a barcode scanner. The system is workflow-driven, such that entry of data at a particular step requires that data has already been entered for previous steps in the process.

An additional component of the LIMS is a system to track materials. All raw materials, such as primers, arrays, and plates, are uniquely labeled and archived. A user can quickly audit and forecast material consumption through reporting and querying modules.

Two copies of the LIMS exist: A production copy currently used in the lab; and a development copy used by the informatics team. The development platform provides a protected environment for informatics team to prepare and debug new features without impacting any laboratory processes.

The LIMS system was built along the lines of a model-view-controller (MVC) architecture. The benefit of this common design strategy is a clean separation of the code necessary to power the database, visual components, and workflow logic. Separation of these components enables a more agile software development process by improving the ability to quickly swap in new modules and algorithms. Additionally, a clean separation of the major system components grants the ability to quickly isolate and debug errors.

The laboratory keeps extensive documentation regarding the usage and design of LIMS in an internal wiki (a user-procured web site). The wiki provides a convenient mechanism for keeping documentation up-to-date and serves as an additional forum for comments and questions.

Sample Tracking

The figures below illustrate a tracking-sequencing run in a high-throughput process. Figure 4 provides an overview of a flowcell run. This report shows samples loaded in each lane of the flow cell, the sequence yield of raw as well as uniquely-mapped tags, and the data quality of each sequenced sample. Figure 2 shows a sample summary view and figure 5 shows the sample, visualized in a genome browser.

One can use the LIMS to find all the samples of a given tissue or cell type that have been processed (figure 3), and present them in terms of the assay types, total mapping tags and the data quality of each sample processed through the sequencing pipeline, which is the most operationally complex in terms of primary and derivative samples and assays performed. Custom reports can be generated for specific cell or tissue types or an assay type performed on given set of tissues. The LIMS thus provides researchers with centralized data storage, analysis tools, and backup and archival services.

Figure 2: Sample view. Shows overview of an individual sample including sequencing, SPOT scores and tag counts.
Figure 3: Cell type view. Shows cell information and samples made from cell type
Figure 4: Overview of a flowcell run, including sample information and quality metrics.
Figure 5: Sample visualized in genome browser showing three component sequencing lanes, the pilot lane, and the final distributed, data release track.

Laboratory Workflow

The LIMS system provides a resource for laboratory personnel to manage, record and archive various components of workflow through the LabData interface. Protocols can be created, versioned and queried and inspected via the LIMS LabData interface. Cell types (tissue/lines/primary) can be registered and managed and the progress of a cell type's samples can be monitored. Essentially all aspects of the lab have an interface with the LIMS, including reagent lot tracking, vendor contact management, collaborations, etc. See Figures 6, 7, 8, and 9.

Figure 6: Entry point for managing workflow objects.
Figure 7: Cell type manager
Figure 8: Protocol manager
Figure 9: Organism manager