In order to further explore how we can create a Virtuous Cycle to better integrate data teams into a biotech organization, I want to go through another case study, this time looking at how we can get analysis from your data team back to the bench scientists. Again, I’m going to split things up into multiple emails, to keep them a readable length (and maybe build a sense of anticipation).
We’ll start with the scenario, which is completely made up, and maybe a bit far fetched, but should illustrate the point:
When exploring the impacts of different compounds on cells, the concentration(s) that you use in an experiment can make a big difference. Too low and it does nothing. Too high and it becomes toxic, skewing the results. You want to find compounds that work at low concentration and have big range before they become toxic. But to get there, you have to start from compounds with a narrower range, then let the chemists optimize them to get the range you want. So finding the right range of concentrations for those initial compounds is important, or you won’t have a place to start.
Your data team has come up with a model that predicts a range of concentrations that are most likely to be effective for any given compound. This can be used to choose concentrations for experiments, and preliminary testing suggests the model performs better than the lab team’s current approach. So you want them to begin trying out the model, both to validate it and to improving the data from future experiments.
Your team has created a UI that scientists can use to calculate concentrations, but no one has tried it yet. When you ask the scientists about it, they always seem to be between experiments, too late in the planning process to change the concentrations, or the experiment is too important to risk a new approach.
So how can the digital team get past these barriers and start driving adoption? Stay tuned next time to find out.
Considering the frequency of cell line contamination, your example hardly seems farfetched.
e.g. https://www.nature.com/articles/2402610
https://en.wikipedia.org/wiki/List_of_contaminated_cell_lines