5 ms·

Detection of Airborne Viruses

99 points by adunk 7y ago

bionhoward 7y ago
Looks great until they hit the ELISA (antibody test) which requires pre-existing antibodies for the target. That’s like when you search for your keys under the streetlight when you dropped them in the dark. Anything novel becomes an expensive and slow ordeal (aka, it doesn’t happen in practical use) because you need “test kits” (sound familiar?) ... just imagine your malware detector would only flag code which exactly matched exploits it had seen already. If they replace ELISA with a nanopore sequencer like Oxford’s MinION then you could detect all the DNA RNA in your breath and feed that data into algorithms like BLASTN. Suddenly you can see viruses no one ever saw, and thousands of them at once. Meta genomics is more sensitive than ELISA because you can see everything and more efficient because you can use the same reagents to test for anything. Then we detect virus everywhere as soon as the sequence is available on NCBI (we had a sequence for CoV-19 two months & four days ago) https://www.who.int/csr/don/12-january-2020-novel-coronavirus-china/en/ https://www.who.int/csr/don/12-january-2020-novel-coronaviru...
- sitzkrieg 7y ago
  note this is from 2018, so it makes drop in like MinION or other sequencers even more feasible
- 2rsf 7y ago
  how cost effective is your solution ?
  - londons_explore 7y ago
    The 'starter pack' costs $1,000, which is good for just two uses. However, the entire device is just made of plastic, silicon and metal, which are very cheap materials. The majority of the cost is R&D cost, which if one were to produce millions of this device would be amortized and you could bring the cost down to that of a $10 fake ink cartridge, which coincidentally has awfully similar technology.
    - notahacker 7y ago
      One would think the case for further public investment in such R&D is now made.
- Munky-Necan 7y ago
  Naive question, because I only knew of ELISA. but can you use MinION or something like it to detect DNA/RNA in blood? I would be excited if this was cost effective and more sensitive and specific for bacteria and viruses than what we have currently.
  - diag 7y ago
    With all things RNA related, sample prep and PCR amplification are the most important steps for good detection with any sequencing platform. Those factors alone make sequencing alone a near impossibility in rapid detection. PCR based assays are a much better candidate and could identify different strains in less than a few hours.
    - Munky-Necan 7y ago
      This is very interesting. I'm usually on the side of interpreting tests/ordering them instead of knowing the nuances behind them. Are there any major breakthroughs in testing that you're excited about?
- netjiro 7y ago
  There are other ways to build ultra sensitive assay technology that does not depend on antibodies, and which can be faster to develop.
- netjiro 7y ago
  I don't know the technical details of MinION. Coupling good capture with amplification and detection is required. When working on dna detection you use something like pcr for signal amplification. But surface detection can in some cases be more sensitive, since you generally have a lot more surface fragments available than you have dna or rna. Especially since many viral diseases form have huge ratios of surface fragments per every well formed infectious viral particle. Generally: a sequencer by itself doesn't have the sensitivity required. For example, for norovirus we're talking about detecting a few virus particles per cubic meter of air. You need insanely sensitive detection tech to be able to do that.
  - virusduck 7y ago
    > I don't know the technical details of MinION. Neither does OP.
- the_watcher 7y ago
  Your first concern is a problem in the current situation, but wouldn't this be immensely helpful in avoiding/preparing for a second wave in the fall (if weather has any impact) or next year (if it becomes pandemic and endemic)?
DoctorOetker 7y ago
In "the andromeda strain" there is a scene where they try to determine the size of the particle, by selecting different filtration sizes between an infected an uninfected mouse / rat. Is this representative of how droplet sizes are measured? Has this been measured for SARS-CoV-II?
- thatcat 7y ago
  no
  - DoctorOetker 7y ago
    which no or both no?
    - thecupisblue 7y ago
      Well if the first one is a no, then the second one is obviously a no since that method isn't used. Find more on virus measurement methods here: https://www.cambridge.org/core/journals/parasitology/article/sizes-of-viruses-and-the-methods-employed-in-their-estimation/BA71C94D73307F5CA7C63753135507F9 https://www.cambridge.org/core/journals/parasitology/article...
      - DoctorOetker 7y ago
        "this" in the second sentence refers to droplet size regardless of measurement method. I have asked for the SARS-CoV-II droplet size distribution many times before, for example here: https://news.ycombinator.com/item?id=22519323 https://news.ycombinator.com/item?id=22519323
        thatcat 7y ago
        The droplet size distribution would be for saliva, covid19 would effect this less than say cough velocity or other variables. The volume of fluid produced does seem to change with illness and influenza may increase the total volume produced in a cough with 63% of the particles in the size range of 0.35 to 10 μm as measured by a laser spectrometer in an influenza infected patient. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4676262/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4676262/
netjiro 7y ago
I worked on one of these projects. This technology has amazing potential for (near) real time detection and monitoring of the spread of airborne disease throughout society. Think shoe box sized air samplers in common spaces that send real time data to monitoring health agencies, alert when a space is contaminated, and so on. The same tech platform can also be used to inexpensively and quickly test individuals. Unfortunately the funding for continuation was lost in the shambles of the partial collapse of Getinge a few years ago. I was not able to find anyone else interested to continue funding the prototype development when the remnants of Getinge did not prioritise the project. Near real time in this case can mean as little as a few hours. We had a total funding of ca EUR 5M for the first prototyping project. If anyone has interest and money, I can pull together people and ip.

99 points by adunk 7y ago

Looks great until they hit the ELISA (antibody test) which requires pre-existing antibodies for the target. That’s like when you search for your keys under the streetlight when you dropped them in the dark. Anything novel becomes an expensive and slow ordeal (aka, it doesn’t happen in practical use) because you need “test kits” (sound familiar?) ... just imagine your malware detector would only flag code which exactly matched exploits it had seen already. If they replace ELISA with a nanopore sequencer like Oxford’s MinION then you could detect all the DNA RNA in your breath and feed that data into algorithms like BLASTN. Suddenly you can see viruses no one ever saw, and thousands of them at once. Meta genomics is more sensitive than ELISA because you can see everything and more efficient because you can use the same reagents to test for anything. Then we detect virus everywhere as soon as the sequence is available on NCBI (we had a sequence for CoV-19 two months & four days ago) https://www.who.int/csr/don/12-january-2020-novel-coronavirus-china/en/ https://www.who.int/csr/don/12-january-2020-novel-coronaviru...

sitzkrieg 7y ago
note this is from 2018, so it makes drop in like MinION or other sequencers even more feasible
2rsf 7y ago
how cost effective is your solution ?
- londons_explore 7y ago
  The 'starter pack' costs $1,000, which is good for just two uses. However, the entire device is just made of plastic, silicon and metal, which are very cheap materials. The majority of the cost is R&D cost, which if one were to produce millions of this device would be amortized and you could bring the cost down to that of a $10 fake ink cartridge, which coincidentally has awfully similar technology.
  - notahacker 7y ago
    One would think the case for further public investment in such R&D is now made.
Munky-Necan 7y ago
Naive question, because I only knew of ELISA. but can you use MinION or something like it to detect DNA/RNA in blood? I would be excited if this was cost effective and more sensitive and specific for bacteria and viruses than what we have currently.
- diag 7y ago
  With all things RNA related, sample prep and PCR amplification are the most important steps for good detection with any sequencing platform. Those factors alone make sequencing alone a near impossibility in rapid detection. PCR based assays are a much better candidate and could identify different strains in less than a few hours.
  - Munky-Necan 7y ago
    This is very interesting. I'm usually on the side of interpreting tests/ordering them instead of knowing the nuances behind them. Are there any major breakthroughs in testing that you're excited about?
netjiro 7y ago
There are other ways to build ultra sensitive assay technology that does not depend on antibodies, and which can be faster to develop.
netjiro 7y ago
I don't know the technical details of MinION. Coupling good capture with amplification and detection is required. When working on dna detection you use something like pcr for signal amplification. But surface detection can in some cases be more sensitive, since you generally have a lot more surface fragments available than you have dna or rna. Especially since many viral diseases form have huge ratios of surface fragments per every well formed infectious viral particle. Generally: a sequencer by itself doesn't have the sensitivity required. For example, for norovirus we're talking about detecting a few virus particles per cubic meter of air. You need insanely sensitive detection tech to be able to do that.
- virusduck 7y ago
  > I don't know the technical details of MinION. Neither does OP.
the_watcher 7y ago
Your first concern is a problem in the current situation, but wouldn't this be immensely helpful in avoiding/preparing for a second wave in the fall (if weather has any impact) or next year (if it becomes pandemic and endemic)?

DoctorOetker 7y ago

In "the andromeda strain" there is a scene where they try to determine the size of the particle, by selecting different filtration sizes between an infected an uninfected mouse / rat. Is this representative of how droplet sizes are measured? Has this been measured for SARS-CoV-II?

thatcat 7y ago
no
- DoctorOetker 7y ago
  which no or both no?
  - thecupisblue 7y ago
    Well if the first one is a no, then the second one is obviously a no since that method isn't used. Find more on virus measurement methods here: https://www.cambridge.org/core/journals/parasitology/article/sizes-of-viruses-and-the-methods-employed-in-their-estimation/BA71C94D73307F5CA7C63753135507F9 https://www.cambridge.org/core/journals/parasitology/article...
    - DoctorOetker 7y ago
      "this" in the second sentence refers to droplet size regardless of measurement method. I have asked for the SARS-CoV-II droplet size distribution many times before, for example here: https://news.ycombinator.com/item?id=22519323 https://news.ycombinator.com/item?id=22519323
      - thatcat 7y ago
        The droplet size distribution would be for saliva, covid19 would effect this less than say cough velocity or other variables. The volume of fluid produced does seem to change with illness and influenza may increase the total volume produced in a cough with 63% of the particles in the size range of 0.35 to 10 μm as measured by a laser spectrometer in an influenza infected patient. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4676262/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4676262/

netjiro 7y ago

I worked on one of these projects. This technology has amazing potential for (near) real time detection and monitoring of the spread of airborne disease throughout society. Think shoe box sized air samplers in common spaces that send real time data to monitoring health agencies, alert when a space is contaminated, and so on. The same tech platform can also be used to inexpensively and quickly test individuals. Unfortunately the funding for continuation was lost in the shambles of the partial collapse of Getinge a few years ago. I was not able to find anyone else interested to continue funding the prototype development when the remnants of Getinge did not prioritise the project. Near real time in this case can mean as little as a few hours. We had a total funding of ca EUR 5M for the first prototyping project. If anyone has interest and money, I can pull together people and ip.