Reproducibility, the ability to carry out and reproduce the results of a single experiment, is an important aspect of scientific disciplines. However, in the life sciences, the concept of reproducibility has become a large problem. A vast number of experiments throughout the various disciplines of the life sciences are seen to lack a reproducible nature, ultimately costing and inadvertently wasting large sums of money. Synthetic biology is no exception to the troubles of irreproducibility, with inaccurate part characterisation impacting the the ability to use Bio-Design Automation (BDA) to build fully functioning, novel synthetic gene circuits.

iGEM devised the inter-lab study – an annual, large-scale study carried out by institutions around the world by researchers of varying experience levels – to determine the reproducibility of individual synthetic biology protocols and to assess and address the limiting factors of reproducibility. The first and second studies assessed reproducibility of expression of green fluorescent protein (GFP) in E. coli, using promoters described as weak or strong in the IGEM repository. The level of precision was reported to be strongly related to fluorescent strength, with greater precision achievable between strong promoters than weak promoters, independent of host strains. Issues were found however with the reporting of results; a lack of standardised units limited analysis to comparison of relative expression levels, with results being reported as the ratio of expression level between strong and weak promoters. Similarly, variation in reported expression levels was suggested to be the results of variation between instruments and method of analysis as opposed to variation within the biological systems in which they were studied; results displayed a tight replicate-to-replicate distribution, whereas results between laboratories showed a wider distribution. As such, the third interlab study included independent calibrants in an attempt to provide directly comparable units and minimise variation between instruments. Results produced by this method allowed quantitative filtering of unreliable results and outliers by comparison to positive and negative control values, increasing the 95% confidence interval normally accepted as biologically significant by four orders of magnitude compared to arbitrary units and up to 600-fold compared to normalised units. The fourth study attempted do address variation between different models of plate readers by producing a step by step protocol for measurement and analysis, allowing the production of directly comparable fluorescence measurements. In addition, ribosome binding site (RBS) sequences designed to increase precision of expression were included in the devices to be transformed into the host cells. The results of this study are still under preparation.

A weakness in the measurement of fluorescence relative to OD600, as with previous IGEM interlab protocols, is the potential discrepancy between optical density and actual cell concentration. This year the IGEM study aims to reduce lab-to-lab variability further by measuring GFP fluorescence relative to absolute cell counts or colony forming units. Normalisation of fluorescence to colony forming units goes further by allowing measurement of fluorescence relative only to viable cells, and thus a more accurate measurement of promoter strength, whereas OD600 and absolute cell count measures cannot differentiate between viable and non-viable cells.