Research Document - 2016/038

Report on the Performance Evaluation of the Fluidigm BioMark Platform for High-Throughput Microbe Monitoring in Salmon

By Miller, K.M., Gardner, I.A., Vanderstichel, R., Burnley, T., Schulze, A.D., Li, S., Tabata, A., Kaukinen, K.H., Ming, T.J., and Ginther, N.G.

Abstract

A formal evaluation of the Fluidigm BioMarkTM high throughput microfluidics platform for use in microbe surveillance research was undertaken.  Forty-seven assays to 46 microbes suspected or known to cause disease in salmon worldwide were developed or derived from the literature, and appropriate controls obtained.  The evaluation completed stage-1 of the OIE pathway, ascertaining the analytical sensitivity, analytical specificity, and repeatability and consistency between platforms within a laboratory for all 47 assays.  Note, however, that our repeatability analyses took place over less than two months, so we may not have adequately factored in time.  The microfluidics platform can assess duplicates of all 47 microbes at once across 96 samples, but to do so, due to the very small size of the reaction chambers (7nl), requires an enrichment step that includes a multiplex with all assay primers.  Therefore, we developed a unique approach to additionally assess the role of the serial targeted enrichment on the specificity and relative quantitation of the assays.

The full undertaking of this study entailed 349,440 qPCR reactions over 37 dynamic arrays on the BioMark platform and 11-384 well plates on the ABI 7900.  By all measures, the assays assessed on the platform performed to high standards, demonstrating a high degree of sensitivity (limit of detection under 40 copies per reaction chamber, with most assays less than 10 copies), specificity (most assays demonstrating >98% sensitivity and specificity to the targeted microbes), and repeatability (averaging 96% over all assays for samples run by two technicians and near perfect agreement for scoring alone).  Consistency of data between the two platforms was also high, with a concordance correlation coefficient (CCC) of 0.95 after correction for the 8-10 Ct difference between platforms.  If samples were only scored positive for duplicate detections and appropriate Ct cutpoints were applied, specificity and sensitivity between platforms rose to >0.99.  Finally, there was no evidence that the STA enrichment step had any great impact the analytical sensitivity or specificity of the assays.

In the process of evaluation, while we found that the platform provided high caliber, reliable data, we did identify a few assays that performed relatively more poorly than others on one or more measures.  Most were still within reasonable limits to use for research purpose to document variance in prevalence and load, but a few were targeted for removal or replacement.  Issues with amplification curve quality appeared to have the largest effect on assay performance, and affected six of the 47 assays. Optimization of primer/probe concentrations could resolve this issue for most, however.

The BioMark platform proved to provide reliable, rapid and inexpensive quantitative data on microbe presence and load.  We did not evaluate the platform for diagnostic use, which would require re-assessment under different criteria to ascertain whether diagnostic testing would be an appropriate use of this technology in salmon.  We note, however, that the platform is already being applied for human viral and bacterial diagnostics and water quality testing.

Accessibility Notice:

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.

Date modified: