DIY Clinical Lab or medical device requiring regulation?

[John Griessen]

On 03/16/2016 12:57 PM, Gordana Ostojic wrote:
> I meant capillary electrophoresis not gel one. The company that I mentioned does that and calibration is basically by
> comparing the different peaks. Regarding the lancet, I see your point but my goal would be mass data (bunch of biomarkers
> measured repeatedly).

So, how do you reuse a capillary? Let it stay with electrophoresis voltage across it until all goes enough toward one end, then
start another sample going through the same capillary gel contents? Would you then soak it in a salt/buffer solution to remove
most of the "leftover junk" at that end? If you wanted to reuse them very many times, would you reverse direction after so many
times? Would you start to see a background fog of "leftover junk" from previous runs?

With reuse, some gradual change would come, so something to compare against that is not
used/contaminated/diffused-to-different-concentrations would be needed. Is there a DNA ladder component or two that you can run
with your desired-to-test range of DNA?

What would it take to cost reduce calibrated DNA ladder components so citizen scientists or health condition trackers could use
some of it every few days? How would the steps be automated to the point of:

1. get a drop of blood safely
2. start capillary EP run
3. scan to recognize when DNA ladder component A is at 80% of the capillary length
4. scan to gather the full results along the capillary length
5. recycle/purge the capillary and put in storage/waiting mode

Hmm... sounds like a medical device again...

Hmm... Theranos is doing something secret like it right now with lots of attention and money and patents...


On 03/16/2016 12:57 PM, Gordana Ostojic wrote:
> I firmly believe that we do need army of citizens digging and finding patterns in their own body to be able to advance health.
> They just need a convenient and reliable device like fitness bracelets now(and start with urine or saliva in the beginning).
> You don't even have to be rigorous with it, if you will measure every day, just want to know abnormalities (new peak, sudden
> increase).

I'm not up on just when the regulators tell you to stop. From the trouble Theranos is having -- related to blood draws -- this
all may be OK with them as long as step 1. is separate and has no medical device involved other than a lancet.

anyone else know more about FDA rules?

[Gordana Ostojic]

So, how do you reuse a capillary? Let it stay with electrophoresis voltage across it until all goes enough toward one end, then
start another sample going through the same capillary gel contents? Would you then soak it in a salt/buffer solution to remove
most of the "leftover junk" at that end? If you wanted to reuse them very many times, would you reverse direction after so many
times? Would you start to see a background fog of "leftover junk" from previous runs?

> One can use voltage or positive or negative pressure in capillaries or microfluidics to reverse the flow. I was thinking to have 2 membranes to reduce capillary contamination. One to keep the big junk out and easily cleanable with water, and the one that was act as separation gate (opening this would start data sampling). At the end I would get water into other end and reverse the flow and then push air through. If you see the company their channel width is of the order of 100s of microns so no it is less tricky than high resolution CE done in research. 

 

With reuse, some gradual change would come, so something to compare against that is not
used/contaminated/diffused-to-different-concentrations would be needed. Is there a DNA ladder component or two that you can run
with your desired-to-test range of DNA?

> There is a paper for CE done on DNA but you dont need to run ladder if you know for example that the 2 largest peaks (like sodium potassium or.. ) you can figure out rate of flow compared to them (well this can be tricky for very different stuff, but you get the gist, compare unknown to the standard things taht must be in blood). 

What would it take to cost reduce calibrated DNA ladder components so citizen scientists or health condition trackers could use
some of it every few days? How would the steps be automated to the point of:

1. get a drop of blood safely
2. start capillary EP run
3. scan to recognize when DNA ladder component A is at 80% of the capillary length
4. scan to gather the full results along the capillary length
5. recycle/purge the capillary and put in storage/waiting mode

Yup something like that. 

Hmm... sounds like a medical device again...

Nope if you do not do interpretation of peaks (23&me got into trouble for interpretation). Let people have a website when they can discuss the peaks and what they mean themselves. Gov cannot forbid that (hopefully). The overall goal is to: a) have individuals discuss patterns b) people themselves know when they felt bad/good and if they can correlate that with some peak (even without knowing what it is).

 

Hmm... Theranos is doing something secret like it right now with lots of attention and money and patents... 

> They don't reveal what they do but my guess is that they are just doing simple fluidics or other known tricks to reduce the amount of blood. My goal is to avoid money that would reduce my ytrouble. Doing great for now for that part ;)