How to Scale a Data Driven Solution?

In some ways, it's a lot easier to push electrons about on the computational level than cells about on the wet lab level, right? So anything that you can do to help streamline and improve that analysis is often interesting enough to explore. I'd like to shift gears a little and go behind the scenes and learn about your approach. When you're faced with a new biological question, what is your typical strategy for defining and then improving upon a data driven solution? Sure. "I myself have always been on the machine learning side and the statistics side"