Right First-Time Scale-up

Isolating strain performance from process factors in strain selection

This video illustrates how the Riffyn Scientific Development Environment enabled a customer to select the right strain for scale-up the first time, saving months of wasted time and effort and accelerating product development cycles.

In this case, a customer was engineering yeast strains for maximum succinate production, then testing yield in small scale fermentation experiments, and then using these data to choose strains for scale-up. Top strains were being identified by manual inspection plots of yield vs. time, and initially all observed changes in yield were assumed to be due to the changes in strain from sample to sample. However, the selected top performing strains were not repeating their performance in downstream runs, creating frustration for the team on both sides of the selection decision.

The customer recognized that there may be other variables contributing to succinate yield, and wanted to identify those variables, but the data were fragmented; strain engineering data were in lab notebooks, fermentation data were stored in on-board instrument databases, and analysis data were stored in Excel spreadsheets. Datasets were hard to find, even harder to interpret, and deciphering connections between datasets was almost impossible.

In this example, Riffyn SDE was used to pull all these data together into a globally accessible system, where data were contextualized and connected automatically into a single comprehensive data table ready for analysis. Using this data table, the customer was able to find an unexpected correlation between temperature and succinate yield, which changed the interpretation of their data. When the results were corrected for temperature variations, the true strain effect was revealed. The data analysis enabled by Riffyn SDE allowed the customer to select reliably improved strains for production, saving the customer months of wasted time and effort.

Process and experiment design were created, and data were captured and prepared with Riffyn. Plots were created with JMP® Software. Copyright 2018, SAS Institute Inc., Cary, NC, USA. All Rights Reserved. Reproduced with permission of SAS Institute Inc., Cary, NC.

"What Riffyn's allowed you to do is look at everything visually, collect all the data, tag it, annotate it; collect all the instrument data, all the strain data, all the metadata and associated it in one place. You wouldn't have been able to do that before."

— John Cumbers, CEO Synbiobeta


Highlighted in the Video

Process design

Sample/strain tracking and data collection

Data integration and export

Statistical analysis with JMP