Watch someone who has pipetted a difficult liquid for years and you will see a set of small, deliberate gestures that they can no longer fully explain. They angle the tip against the tube wall on dispense. They hold the plunger at the second stop a beat longer than seems necessary. They keep the tip just under the surface and draw slowly, then pause before lifting. None of this is written down anywhere, and much of it the person does without conscious thought. It is real knowledge, hard-won and specific to the liquid, and it is exactly the knowledge that evaporates when a protocol moves from the bench to a deck, because a liquid handler does not learn by watching. The good news is that every one of those gestures corresponds to a parameter, and translating them one by one is how you carry the skill across rather than leaving it behind on the bench.
The gestures and what they become
A manual technique is a physical encoding of a decision about the liquid. The deck has a numeric encoding of the same decision. Translation is a matter of matching them.
- Drawing slowly: the technician slows the draw because they can feel a thick liquid failing to keep up, or hear a thin one hiss air. That is the aspiration flow rate. On the deck you lower it until the column comes up clean, the same judgment expressed as a number.
- Holding at the second stop, then dispensing only to the first: this is reverse pipetting. It becomes an over-aspiration volume plus a partial dispense mode that leaves the reserve in the tip, with the excess routed to waste, exactly the reserve-for-the-film logic the thumb was performing.
- Pausing before lifting the tip: the beat they hold after aspirating, and again after dispensing, is a settling delay. It gives the column time to equalize before the tip moves, and it becomes an explicit post-aspiration and post-dispense delay in the class.
- Keeping the tip just under the surface: shallow, submerged immersion avoids gulping air on the way in and drooling liquid down the outside of the tip. It becomes the submerge depth and the liquid-level following behavior.
- Angling against the wall on dispense: touching the tip to the well wall or the existing liquid lets a viscous slug leave by contact rather than trying to jet off cleanly. It becomes a surface or wet dispense mode with a touch-off, rather than a free dispense from height.
- Pre-wetting by drawing and expelling once first: the technician who cycles the liquid once before the real transfer is stabilizing the first dispense. It becomes an explicit pre-wet step in the class.
Laid out like this, the mapping is almost mechanical, and that is the point. The technique was never mysterious. It was a set of specific responses to specific liquid behaviors, and the deck has a slot for each response.
What is easy to lose in translation
Two things do not survive a naive transcription. The first is the why. A technician slows down for glycerol and speeds up for a volatile, and the reason lives in their head, not in the gesture. If you copy only the number and not the reason, the next person who meets a slightly different liquid has nothing to reason from and starts over. The second is the calibration to reality. A skilled hand adjusts continuously to feedback, wetting, temperature, how the last transfer looked, and a fixed parameter cannot. So a faithful translation records not just the setting but the intent behind it and the conditions it was judged under, and it accepts that the class will need gravimetric verification precisely because the deck cannot feel its way the way the hand did.
Why the translation is worth the trouble
The instinct is to treat the bench technique as a crutch to be discarded once the method is automated, but that gets it backwards. The technique is the distilled result of many runs of trial and correction, and it is the fastest possible starting point for a class. Someone who already handles a liquid well by hand can hand you a validated class in far less time than someone tuning from a water default, provided you take their gestures seriously enough to translate them rather than asking them to guess at numbers they never think in. The translation is also how the knowledge stops being a single person's asset. Once the angle, the pause, and the reserve are written as parameters with reasons, they belong to the lab, and they survive the technician moving on, which is the whole reason to bother.
A skilled pair of hands is a liquid class that has not been written down yet. Translate the gestures into parameters and the reasons into notes, and the skill outlives the shift that learned it.