Catching (Real) Viruses With Silicon

Roland Piquepaille writes "Researchers in Iowa have used nanotechnology to develop a very small silicon chip to catch and help identify viruses, according to Technology Research News. The device, dubbed the ViriChip, is used in conjunction with an atomic force microscope (AFM). The prototype is already able to identify several viruses and should be in labs in less than two years. A particular application could help save lives by enabling doctors to check a donor heart for potential infections before transplanting it to a patient. This overview contains more details. It also includes references to other articles about the ViriChip and images showing how it looks and a virus it detected."

The chip will be ready in a year, but...

[Anonymous Coward]

you need an atomic force microscope to analyze the results. Them AFMs ain't cheap or common.

Still need antibodies

[RobertB-DC]

As a casual fan of science, this looks way cool for detecting viruses* that we already know about, by using the antibodies that attach to the virus protein as the detection agent. But aren't we limited to viruses for whom we've already isolated antibodies?

I wonder how hard it will be to expand this into a more general virus detection/identification tool? It seems like you could break up your suite of antibody-derived proteins into smaller, more generic chunks that would be more likely to bind to the virus. But I'm getting beyond my depth -- would like to hear from someone who knows what they're talking about!

* I've heard virii is now passe' -- any confirmation?

Re:Still need antibodies

[Goldsmith]

This technique is actually worse than you think.

It can only be used on viruses which have a well known surface behavior, which we have a known antibody for and whose antibody can be attached easily to a SAM (protein monolayer).

On the other hand, this technique requires rinsing away all the viruses which are NOT attached to the antibody. It is very easy (and is done all the time) to just deposit a collection of viruses, or whatever you want, on a piece of silicon and take an AFM of it. Viruses tend to all look similar though.

They're trying to be ultra-specific, for commercial reasons. If you want to do research with this, it's being done now, and has been done for the last 10 years.

Re:Still need antibodies

[Warhaven]

I wonder if it can also be used identify damaged proteins, such as those which cause Mad Cow Disease. Nasty buggers, 'cuz they're not living, so it can be transferred from patient to patient even after going thru sterilization.

OT: Viruses or just Virus

[Valdrax]

* I've heard virii is now passe' -- any confirmation?

It's not just passe, it's grammatically nonsensical. [perl.com] Pluralize it as if it was English. "Virii" would require the nonexistant "virius" to be correct, and "viri" is the plural of "vir" already. Some scholars think that virus had no plural form and should be treated like "fructus" and just use the same word for singular and plural.

Re:Still need antibodies

[almightyjustin]

* I've heard virii is now passe' -- any confirmation?

*virii is not, and has not ever been, correct. For that matter, neither is *viri, as used in this product title. Basically, virus was a numberless noun in Latin, so it had no Latin plural, and viruses is the only valid English plural.

This page [perl.com] has more information than you probably ever wanted to know on the subject.

Re:Still need antibodies

[SB9876]

I'm a bit dubious about this tech, myself. I'm trying to figure out how this does anything that an ELISA sandwich (mmmmmmmm....sandwich) assay couldn't do. An antibody-HRP conjugate detection system can ustilize fairly simple and cheap detection methods with high linearity and very good sensitivity.

The AFM detection might have a lower detection limit but I'm concerned about how well it can handle real-world samples. Having done biological AFM, you get used to seeing random schmutz all over the place fro

Oh I can see it now ...

[Sonic McTails]

"New Silicon-based Virus Scanner Infected With Virus"
from the got-virus-scanner dept.

A related approach from 25 years ago

[Giant peach]

In the '70s, a Nobel laureate in physics, Ivar Giaever, an expert in thin films, took a sabbatical in San Diego to investigate immunology out of deep curiosity. Among the things he came up with was a carefully designed film of indium which could be deposited onto a glass microscope slide. You could then dip half the slide into a solution containing the antibody, rotate the slide 90 degrees and dip into a solution containing the antigen. If there were enough antigen present in the second solution, the quadrant of the slide on which antigen-antibody complexes were stuck was visibly different from the other quadrants in ordinary light, even though the antibody and antigen layers were one molecule thick. I don't know whether anything came of this later. He was thinking about a cheap and easily administered clinical test at the time.

I wonder...

[ion_]

Will the chip's specs be released to public? This chip would be very useful in mail servers that do virus scanning. It would only be a matter of time until e.g. Clam AntiVirus [clamav.net] supports this chip.

PS. Does anyone know whether it supports Ogg Vorbis?

Is this a justifiable venture?

[StateOfTheUnion]

The method identifies a virus using an entire virus particle rather than viral components, can detect viruses in liquids, and does not need to destroy viruses during the identification process. This allows intact, identified viruses to be further analyzed, according to the researchers.

Can we not detect viruses now based on the "entire virus particle" if we have the proper antibody (which is also required for the ViriChip)?

Can we not detect viruses in a solution now? (Granted the antibody would have to b