[Neuroscience] Re: Equation that explains the behaviour of a circuit in voltage clamp

r norman via neur-sci%40net.bio.net (by r_s_norman from comcast.net)
Tue Apr 7 21:01:36 EST 2009


On Tue, 7 Apr 2009 18:50:31 -0700 (PDT), "Bill.Connelly"
<connelly.bill from gmail.com> wrote:

>On Apr 6, 6:13 pm, r norman <r_s_nor... from comcast.net> wrote:
>
>> Are you actually attempting to do voltage clamp experiments or are you
>> trying to solve homework problems?
>
>More general curiosity than anything. What I was REALLY trying to do
>was understand the behaviour of the op amp circuitry of your basic
>headstage, i.e. the current to voltage converter and then the
>differential amp that subtracts Vcmd. I thought I needed to know the
>equation for the feedback current, but I didn't. I was trying to prove
>to myself that even with a significant series resistance, the op amp
>circuitry still behaved properly (even if it doesn't clamp the cell
>properly).
>
>I've gotten quite obsessive about trying to understand every detail
>about e-phys recently. I was at a neuroscience course, where a lot of
>the other students (some of whom were post docs and had far 'fancier'
>papers than I do, doing whole-cell voltage clamp) really had no idea
>about what they were actually doing during voltage clamp. They just
>kinda assumed that the entire cell was clamped exactly where they set
>the dial on the preamp, and the reported current was exactly equal to
>the transmembrane current.
>
>It just kinda reminded me of how anyone can be taught to use a
>calculator, and 9 times out of 10, they'll get the right answer, but
>if they don't understand math, that 10th time, when they times 2 by 2,
>and they get 46873, they wont think it is odd.
>
>It was really quite interesting looking back on Hodgkin and Huxley's
>papers, when they 'invented' series resistance compensation... and the
>resistance they were compensating for was only a couple of kiloOhms
>(though the whole fact that they use opposite directions for current
>and voltage makes it a bit harder to read than it should be).
>
>But I just really like seeing those equations. I think dV/dt = I/C is
>really quite nice, and while not quite as succinct Vm = Vcmd - (Vm/Rm
>+ C dVm/dt)  * Rs is pretty cool too.

You are right in thinking that too many biologists don't understand
their equipment and how it works and the many, many potential pitfalls
in blindly assumping that the numbers they get are real.  On the other
hand, there is the story I related of KC Cole, who worried about
whether the data was rea,l and Hodking/Huxley, who interpreted the
data.  Only one of these two groups walked off with the prize. You can
go overboard on the fine details.


More information about the Neur-sci mailing list