LAMP vs PCR testing for SARS-COV-2

[jimbob]

https://twitter.com/V_Saggiomo/status/1 ... 9816314880

I have no idea as to the practicality of this,

Finally is out, our super cheap, easy to produce in millions, almost universal, and with low waste impact nucleic acid detection (LAMP test for #COVID19) non-instrumental device. The answer is: a Nespresso capsule. The question is: why we did this? 1/n

https://chemrxiv.org/articles/preprint/ ... n/14224481

Any chemists care to comment - *is* it being oversold in this series of tweets?

[Turdly]

I've interacted with Vittorio on and off over the years. He's a really nice guy but I'd say that he can be a little naive at times.

I can't say I've looked into LAMP recently, but my OH (a virologist-turned toxicologist) spent a year working on LAMP-based diagnostics and thinks it is shite. That was 10 years ago though.

Some general issues with these sorts of things:
- colorimetric assays need unbiased readers (i.e. not a person) particularly when going from one colour to another otherwise it is easy to convince yourself it is not really positive/negative. This is why lateral flow assays have a control stripe to compare to (and to ensure the assay is working).
- anyone working in healthcare diagnostics can tell you that the cost of the instrument is not the problem it is the cost per assay. If a company can tie you into buying a lot of consumables for the next 10 years then they'll pretty much give you the instrument (ironically, the same model that Nespresso work on!)
- the idea that because Nespresso have a plant for filling little aluminium pods with ground coffee means that they can also fill pods with a phase-change material, then melt and mould it, is laughable. Someone needs to watch Inside the Factory to understand how specialist these types of facilities are.
- the obsession with 3D printing. Here, it is useful for demonstrating a prototype device but it reality if you needed to make millions of plastic tube holders it'd be infinitely cheaper and quicker to make them by injection moulding.

[jimbob]

I've interacted with Vittorio on and off over the years. He's a really nice guy but I'd say that he can be a little naive at times.

I can't say I've looked into LAMP recently, but my OH (a virologist-turned toxicologist) spent a year working on LAMP-based diagnostics and thinks it is shite. That was 10 years ago though.

Some general issues with these sorts of things:
- colorimetric assays need unbiased readers (i.e. not a person) particularly when going from one colour to another otherwise it is easy to convince yourself it is not really positive/negative. This is why lateral flow assays have a control stripe to compare to (and to ensure the assay is working).
- anyone working in healthcare diagnostics can tell you that the cost of the instrument is not the problem it is the cost per assay. If a company can tie you into buying a lot of consumables for the next 10 years then they'll pretty much give you the instrument (ironically, the same model that Nespresso work on!)
- the idea that because Nespresso have a plant for filling little aluminium pods with ground coffee means that they can also fill pods with a phase-change material, then melt and mould it, is laughable. Someone needs to watch Inside the Factory to understand how specialist these types of facilities are.
- the obsession with 3D printing. Here, it is useful for demonstrating a prototype device but it reality if you needed to make millions of plastic tube holders it'd be infinitely cheaper and quicker to make them by injection moulding.

Thanks. I suspected that it was an interesting proof of concept, but little else. And as you say, 3D printing is great for prototypes or short runs.

The one in red seems to be a pretty fundamental issue. The rest look to be things that could get altered by setting up an appropriate supply chain, and with a reasonable cheapness for at least certain low-utilisation systems.

[shpalman]

As we saw from the kid who built his own PCR basically all you need for this sort of thing, once you have the necessary reagents, is the right sort of temperature profile.

[nekomatic]

I've interacted with Vittorio on and off over the years. He's a really nice guy but I'd say that he can be a little naive at times.

He doesn’t seem great at recognising that something is not a compliment:

F362D197-A9D2-4A8F-B9A2-46647E097A68.jpeg

[Turdly]

As we saw from the kid who built his own PCR basically all you need for this sort of thing, once you have the necessary reagents, is the right sort of temperature profile.

Yes, and cheap basic PCR machines have been known for years e.g. this one from 2007.

The whole article is a bit "emperor's new clothes" and their bibliography helpfully shows it. The concept of the device is not new. They've taken a set of LAMP reagents, and primers designed by someone else. They've taken a commercial phase-change material that matches the temperature of previous PCM-based devices. The "inventive step" is putting the PCM in an empty Nespresso capsule and using a bowl of hot water to heat the PCM - this is a nice idea but not exactly ground breaking.

Although not in the paper, Vittorio thinks that this can be easily scaled

Screenshot 2021-03-26 at 06.10.16.png

but is still selling it on the being waste free and that 3D printing is important

Screenshot 2021-03-26 at 06.11.35.png

which is BS as it requires significant changes to manufacturing, and you'd be laughed at if you tried to make millions/billions of holders with 3D printing.

[jimbob]

I've interacted with Vittorio on and off over the years. He's a really nice guy but I'd say that he can be a little naive at times.

He doesn’t seem great at recognising that something is not a compliment:
F362D197-A9D2-4A8F-B9A2-46647E097A68.jpeg

To be fair, phase change materials* are pretty cool in certain applications.


*to me PCM means "process control modules" which are specific test patterns one creates in semiconductor processing in order to electrically measure various parameters that depend on the process - for example diffusion lengths of implants, and how they affect the location of PN junctions,

[shpalman]

That "phase change material" is "just" there to provide temperature stability?

Neat though to find something which melts at the temperature you want.

PCM to me means pulse-code modulation.

[jimbob]

That "phase change material" is "just" there to provide temperature stability?

Neat though to find something which melts at the temperature you want.

That's how I understand it. I guess thermodynamics would say that any temperature-induced phase change will have a latent heat associated with it, which would work... not just melting or boiling.