Rigor & Reproducibility - Translational Genomics Lab

The Translational Genomics Lab (TGL) adheres to a standard of excellence in experimental methodology to ensure accurate and reliable results. In close collaboration with the UNC Lineberger Office of Genomics Research (OGR) and the UNC Lineberger Bioinformatics Core (LBC), TGL has implemented several operational procedures that highlight our commitment to the rigor and reproducibility of cancer research. Here are some of those practices.

Project Consultation

Before samples are submitted, a project consultation with the investigator(s), TGL, OGR, and any other relevant group (e.g., LBC) is scheduled to discuss all aspects of a project to ensure the methodology is appropriate for the research questions being address. We find that meeting to discuss a project is much better than attempting to work out the details of a project over email.
To transition what is discussed during the consultation into a research plan, an informal Statement of Work (SOW) and estimate are provided to the investigator(s), the TGL Director, and OGR Director for signatures to make sure everyone agrees with how the samples will be processed, the estimate costs of those services, and the estimated turnaround time.

Systems

On a project level, TGL relies on a custom-built project management system, Manifest Annotation Database (MAND), that was designed by OGR and built and maintained by LBC. MAND keeps track of all project details, such as where the samples originated, the processing details (e.g., the library prep method and sequencing parameters), and any communication that occurs between all relative groups. This system is backed up on LBC servers, so the information for all historic projects can be easily accessed by TGL, OGR, and LBC team members. This centralized database is especially important when there is staff turnover.
On a sample level, TGL uses a custom-built Laboratory Information Management System (LIMS) that was designed, built, and maintained by LBC. In LIMS, TGL keeps track of all the metadata associated with each container generated during sample processing using automated upload features that utilize output files directly from our equipment to reduce the risk of manual entry errors. Through MAND, automated project reports are generated for the end user at the end of sample processing without the need for error prone manual curation. These reports contain all information provided by the investigator at the start of the project, all metadata associated with sample processing, and the location of the final deliverable (e.g., fastq file path).
On a data level, TGL collaborates with LBC to securely store and backup all the data we generate on Lineberger servers. The stored raw data is not directly accessible to anyone except LBC to eliminate the risk of files being invertedly changed or deleted.

Lab Operations

To reduce the risk of contamination, TGL lab space is divided into two sections that are physically located on separate wings of Marsico Hall: a pre-amplification area and a post-amplification area. All protocol steps that occur prior to PCR take place in the pre-amplification area, while all steps that occur after PCR take place in the post-amplification area of the lab. Foot traffic between the two areas is kept to a minimum, but when it must occur, PPE is changed between spaces.
To maintain sample and reagent integrity, TGL refrigerator and freezer temperatures are continuously tracked using Minus80 Monitoring.
TGL workflows are designed around automated instrumentation (e.g., KingFisher Flex for nucleic acid extractions, Hamilton NGS STAR automated liquid handlers for library prep and pooling) to improve reproducibility and reliability.
TGL instruments are covered by vendor service contracts and undergo regular preventative maintenance to proactively avoid unexpected equipment failures.
TGL sources reagents and consumables from reputable vendors, such as Illumina, Twist Bioscience, New England Bioscience, Hamilton and Beckman Coulter, and uses them within their stated expiration dates.
TGL team members are trained to strictly follow the standard operating procedure (SOP) for each service offering to ensure reproducible results and minimize the risk of mistakes.

Data Quality Control

Assay performance is proactively monitored with positive and negative controls so issues can be identified and addressed as early as possible.
With the help of LBC, identity checks are performed for projects with multiple samples from the same individual to ensure there were no swaps that occurred during sample processing.