ePACE Pump Mapping

Experiments
1

Hi all,

Thanks for the excellent documentation and the amazing device We’re setting up ePACE on a newly acquired eVOLVER unit. Our configuration: 16 low-flow (~1 mL/min) and 32 high-flow (~45 mL/min) pumps. We plan to run 8 lagoons, using 8 vials as lagoons and 8 as reservoirs.

From the Experiment Setup page ( Experiment Setup | eVOLVER ), it looks like you run two low-flow lines per lagoon for inducers (e.g., arabinose + drift), which would consume all 16 of our low-flow channels. That would leave everything else (media→reservoir, reservoir→lagoon, and efflux) on the two high-flow arrays. I am a little unsure if we can run the vial to vial connections (resrvoir→lagoon) on the high-flow pumps, given the low turnover of the lagoons for some selection circuits. To make sure I understand your setup correctly, I came up with some questions:

Pump mapping assumptions

In your template and schematic setup (shown here, Experiment Setup | eVOLVER ) do you assume pumps 0–31 = high-flow and 32–47 = low-flow? If not, what’s the recommended way to declare a custom map (from the epace-template/pump_cal.json file I unfortunately could not get the settings, as flow rates are 0.75ml/s for all 48)?

Can high-flow pumps handle low hourly rates?
We’d like lagoon dilution rates ≤ 1 V/h (e.g., 5 mL/h into a 10 mL lagoon). A 45 mL/min pump dispenses ~0.5-0.75 mL per 1 s pulse, so to hit 5 mL/h we’d need ~1 s every 9 min (0.75 mL × 6.7 pulses/h ≈ 5 mL/h). Is that duty cycle reliable in practice, or do you strongly recommend reservoir→lagoon on low-flow only? Do you have a minimum reliable pulse width and/or max recommended bolus volume (as % of lagoon volume) for high-flow pumps to avoid oscillatory OD / mixing artifacts?

Additional documentation
If you’ve already built the dual-manifold setup for ePACE with two inducers per lagoon, could you share a picture or two of your tubing/pump routing and any epace-template code snippets and config files associated with it?

If the strong recommendation is to keep reservoir→lagoon on low-flow, we’ll either (a) add a small external low-flow manifold for inducers and free the built-in low-flows for reservoir feed, (b) reduce to a single inducer per lagoon, or (c) purchase another low-flow pump rack from you.

Many thanks and best regards!
—Jay

1 Like
2

@Nate_Borders would probably be helpful here.

3

Hi Jay,

Thanks for the thorough questions!

Yes

Use the default map, but you need to do a pump calibration, where you will declare which pumps are slow and which are fast.

Your math is correct, however, we define the minimum bolus size to be 0.5mL. Regardless, I haven’t done significant characterization on this exact question. You could get low flow rate pumps for reservoir lagoon flow. I do know that 1) this setup seems to work well enough at 0.5V/h 2) cells will likely be growing within the lagoon at such a low flow rate 3) you should move on to higher flow rates as soon as you can

You don’t need a manifold because there is one pump per via. Instead you can use tubing splitters. epace-template code is on the github under big-e branch currently.

Best of luck! I’ll be gone for a couple of weeks by the way.

1 Like
4

Hi @Nate_Borders

many thanks for your helpful reply. I understand the configuration much better and after investing some time into a deepdive in all of your code, everything is coming together.

As soon as the evolver is up and running, we will start evolutions for several projects and we will get a lot of experience/data from it. Happy to let you know (and potentially collaborate on, how the platform could be even further optimized for ePACE and if we could make an easily accessible DIY approach for the IPPs as they did in ePACE.)

Anyhow, I hope you enjoy your time off and many thanks again.

Best regards, Jay Zoellin

2 Likes