How an early-warning radar could prevent future pandemics
- Thread starter JournalBot
- Start date
It's how AAV2 has been identified as a virus of interest, and possible causative pathogen, in the cohort of child hepatitis cases in the UK and elsewhere. Adenovirus serotypes and HSV were picked up in some samples, but AAV2 was the only virus which was consistently there (and not always present in control cases)
Upvote
39
(39
/
0)
This is a nice article- using untargeted metagenomics is a great way to find new viruses. However, implementing it as an early warning system is tough. A few reasons, we have both DNA and RNA viruses which need different sequencing, where do you sample? Depending on the virus there are different reservoirs in the body, depending on the sampling and isolation you can easily bias MG which can affect the results. And of course you can get a lot of human in there for either DNA or RNA because the virus nucleic acids are only s small part. When do you use MG? When someone seems to have an unidentifiable virus? This is like 2 out of 3 times I take my kid to the doctor! Or when someone is severely sick in a hospital for extended length of time? It would be great if the article suggested a potential way to implement this.
There is an example of a company which has been doing this for a while trying to get a new viruses (https://en.m.wikipedia.org/wiki/Metabiota). They collect bush meat and samples from potential infected people in countries with wildlife interface zones.
Also PS, sequencing is not cheaper in China because they have companies that run more machines. It is cheaper in China because the western world has patents and intellectual property that allows Illumina to own the market (until recently?) and China has a heavily government subsidized “industry” that produced sequencing instruments and uses them. It is somewhat more complicated than this- but a totally weird thing to say in a mostly accurate scientific article.
There is an example of a company which has been doing this for a while trying to get a new viruses (https://en.m.wikipedia.org/wiki/Metabiota). They collect bush meat and samples from potential infected people in countries with wildlife interface zones.
Also PS, sequencing is not cheaper in China because they have companies that run more machines. It is cheaper in China because the western world has patents and intellectual property that allows Illumina to own the market (until recently?) and China has a heavily government subsidized “industry” that produced sequencing instruments and uses them. It is somewhat more complicated than this- but a totally weird thing to say in a mostly accurate scientific article.
Upvote
51
(52
/
-1)
Lots of noise to sift through to find a signal of interest.
If the case can be made that the signals are important to national security, then maybe political backing for this research is possible. The national security complex looks for signals amid lots of noise on a daily basis.
On the other hand, if the answers are deemed as only interesting and not existential in nature, then the data search and analysis will fall onto the amateur community, along the lines of SETI or asteroid/comet detection.
If the case can be made that the signals are important to national security, then maybe political backing for this research is possible. The national security complex looks for signals amid lots of noise on a daily basis.
On the other hand, if the answers are deemed as only interesting and not existential in nature, then the data search and analysis will fall onto the amateur community, along the lines of SETI or asteroid/comet detection.
Upvote
9
(9
/
0)
There are plenty of public health agencies that could use this as a tool and not have to rely on national security or amateur sleuths (which would be a horrible idea). The problem is that public health is not sexy and prone to budget cuts during times when nothing much is happening.
Upvote
25
(26
/
-1)
And of course one of issue is the handling of samples for unknown pathogens - this is one of those things that makes this work dangerous for the front line people and some for the downstream of the viruses are not properly deactivated. And finally - we do have a system for doing this in place already called the CDC, where samples go from the health department across the US for them to sequence and identify novel viruses.
Upvote
12
(12
/
0)
A way this metagenomic sequencing technique could become more wide spread is if the large universities would use it as a teaching and research tool. There will be a need for scientist who know how to use this technique and the machines that does these metagenomic sequencing and the medical industry will need people capability of running the machines. This could done through support from the state and/or federal and/or industry.
Upvote
7
(8
/
-1)
I'm grateful for the author's dual focus on both the successful technology and the failed response. We probably have just about every technology we need, including metagenomics, for early detection of emerging pandemics and rapid response to limit their spread. Alas, this isn't a technology problem, it's a failure of will. Pandemics are easier for the wealthy to avoid than for the poor to dodge. Hence it's economically advantageous for the powerful to deny the problem in order to keep the poors coming in to work, and to be replaced as they drop, while the rich Zoom into meetings from their yachts and their compounds. This is why we as a society are now quite comfortable with COVID settling in as the third leading cause of death in America, steadily killing a 9-11 worth of innocent Americans every week - we have been taught to ignore it. Public health is broken because any semblance of equity in our economy is broken.
Upvote
26
(31
/
-5)
Yet post new pandemic-discovery, one thing will still stand in the way of effective containment and the eventual tapering off of the threat. We just experienced the issue with Covid. Finding it is one thing. Effectively doing something about it in one coordinated effort without resistance is another.
Upvote
6
(8
/
-2)
Another point that the article makes is that the private majority payers in our multi-payer health system have no incentive to fund such tests, unlike the single-payer health care system that civilized countries have. Once again - why we can't have nice things.
Upvote
12
(15
/
-3)
The use of metagenomics sequencing to identify potential disease-causing organisms is a fascinating development in genomics. This approach could revolutionize our ability to detect and respond to emerging diseases before they become full-blown epidemics. However, as the article points out, there are challenges to using this technology, including the high cost and the need for specialized expertise to interpret the results. Additionally, the sheer volume of data produced by metagenomics sequencing can make it difficult to determine which organisms are causing illness. Despite these challenges, the benefits of metagenomics sequencing are clear, and it will be interesting to see how this technology is further developed and applied in the coming years.
Upvote
-10
(0
/
-10)
While I love articles like this, it just underlines and circles in red the fact that you can have all the cool tech, and detection techniques in the world but the stupidity of humans will negate most of that. 1995's Outbreak with Dustin Hoffman was an unrealistic movie but it got one thing right: concern about a person's job or getting it wrong is more important than looking into something early and getting the word out.
Upvote
7
(7
/
0)
If the Last of Us has tought me anything, it's that we should all learn a new practical skill, like horseback riding, foraging for edible berries, the many ways to cook decade old canned pasta; and how to live in peace with our mushroom zombie overlords.
Upvote
4
(5
/
-1)
I thought we'd heard threats of the next big pandemic since bird flu. But nobody wants to pay for prevention. And that ought to be the story.
Everyone from the American Medical Association to big pharma to our elected politicians appear to be opposed to public health. It's not because of the expense. It's because that's not deemed profitable enough. An ailing population can be milked for cash and every middleman loves that!
Everyone from the American Medical Association to big pharma to our elected politicians appear to be opposed to public health. It's not because of the expense. It's because that's not deemed profitable enough. An ailing population can be milked for cash and every middleman loves that!
Upvote
1
(5
/
-4)
The problem with such an "early warning system" such as this is that it relies on people seeing reason. And if there is one thing Covid has shown us, it is that you cannot reason someone out of a position they did not reason themselves into.
Or, put simply, a large percentage of the population is too stupid to live without warning labels. And they are going to kill the rest of us in the end.
Or, put simply, a large percentage of the population is too stupid to live without warning labels. And they are going to kill the rest of us in the end.
Upvote
6
(11
/
-5)
[digression]
A comment on the English language:
"is not sexy and prone to budget cuts" means "neither sexy nor prone to budget cuts"
"is not sexy and is prone to budget cuts" means what the poster intended
[/digresssion]
Upvote
-7
(1
/
-8)
Excellent book! Particularly interesting reading about early pandemic modeling techniques.
Upvote
0
(0
/
0)
If you want something much closer to reality but still scary as shit - watch the national geographic miniseries (season1) based somewhat on Richard Preston’s hot zone - which should be required reading….
https://en.m.wikipedia.org/wiki/The_Hot_Zone_(American_TV_series)
Season 2 is good also as it tells the anthrax story really well. Don’t open those envelopes!
Upvote
3
(3
/
0)
First off, while the sample source/tissue issue is a concern, it's also an issue faced by the traditional diagnostic techniques used, so there's nothing particularly worrying about this method.
In terms of DNA and RNA viruses, this isn't as big of a problem as you might think.
Regardless of the type of nucleotides that make up the viral genome, they all get converted to DNA before the sequencing libraries are prepped. Depending on the kits used, parallel DNA and RNA extractions are performed using paramagnetic beads, silica resins, or a combination of both.
Note: integrated DNA/RNA protocols are out there, but for medical diagnostics, seperate extractions are more common.
Any RNA molecules are converted to cDNA using reverse transcriptase, and that's the norm for most NGS platforms.
As for the Chinese sequencing costs, don't forget the much reduced labor costs, particularly if someone outsources the bioinformatics.
You get what you pay for in some cases, but particularly in an academic setting, the cost savings can be substantial
Upvote
6
(6
/
0)
While one does convert RNA to double stranded DNA with RT-PCR, one really can’t then just sequence the sample and also sequence the DNA you left in the sample. You need to do separate DNA and RNA extractions for metagenomics and then two different protocols for sequencing libraries are used. One is total RNA if you want to capture the viral RNA fraction, the other might be a TN-seq for low quantities of DNA or an amplified or unamplifed whole genome library If there is enough sample. Traditional diagnostic techniques are looking for known viruses, not unknowns. There really is no comparison. Most diagnostics use a antibody that binds to a fragment and has a reporter attached to it. If it’s a genomic test then you use specific primers in the PCR to amplify the target.
If one were to simply use blood or nasal discharge for DNA or RNA metagenomics the sample would be nearly all human- When we do viral sequencing we amplify with primers specific to the virus.
Yes labor is cheaper in China for lab work- but the lab work is actually a fairly small cost of doing something like untargeted metagenomics - the cost is almost all in the library kit and sequencing flow cell from Illumina.
Last edited:
Upvote
3
(3
/
0)
Then we need to stop voting for candidates running on cutting budgets. Who are some politicians/parties that have done this?
Upvote
0
(2
/
-2)
There was funding during covid and a large influx of funding into public health departments through the CDC to in part improve surveillance:
https://www.cdc.gov/media/releases/2022/p1129-cdc-infrastructure.html
Upvote
3
(3
/
0)
totally off topic, has anyone read rats, bats and vats? (or whatever the order was)
anyway... if it's a case of test any and everything with the new gizmo, then you can bypass the insurance angle altogether by testing samples from the sewer. probably totally useless unless it found something similar to known pathogens, because it doesn't tell you anything about symptoms, progression of the illness in the patient. could be useful on the macro scale to figure out how bad things get once you know what to look for.
anyway... if it's a case of test any and everything with the new gizmo, then you can bypass the insurance angle altogether by testing samples from the sewer. probably totally useless unless it found something similar to known pathogens, because it doesn't tell you anything about symptoms, progression of the illness in the patient. could be useful on the macro scale to figure out how bad things get once you know what to look for.
Upvote
-4
(0
/
-4)
Highlight by me. You will get all the funding you need when its happening but that is the proverbial barn door being closed after the horses have bolted. We need sustained funding before this shit happens to have the infrastructure in place before hand. And correct me if I'm wrong but didn't the Orange Shit Stain remove resources in China that would have helped detect this?
Upvote
4
(5
/
-1)
Yea this was post covid 2022 and into 2023 during the non orange president phase- designed to support enhanced surveillance and bolster state health departments in the US. I don’t know how successful this has been US so far. Covid made it apparent that health departments were understaffed across the US to respond to more than a listeria or stomach flu outbreak. I was noting that there has been additional investment here in the US.
But yes there was a reduction during the orange presidents term in foreign surveillance in China:
https://www.reuters.com/article/us-...a-prior-to-coronavirus-outbreak-idUSKBN21C3N5
Upvote
5
(6
/
-1)
YetAnotherAnonymousAppellation
Ars Praefectus
Maybe someone could invent a cryptocurrency where mining is based on doing actual work like this. There are other compute-intensive tasks that could also be used as "proof of work" -- weather forecasting for example.
edit: { =/= [
Last edited:
Upvote
3
(5
/
-2)
Indeed. Western countries had plenty of time to prepare for an outbreak. It took the virus about a month to leave China. First outbreaks in other countries could be jailed and eliminated. Detecting a virus three days earlier doesn't improve much.
Instead of instantaneous virus sequencing we need a pill against this ardent longing to fly criss-cross around the globe while a pandemic starts to develop and against the burning desire to avoiding protective masks like a plague.
Upvote
1
(4
/
-3)
Maybe someone could invent a cryptocurrency where mining is based on doing actual work like this. There are other compute-intensive tasks that could also be used as "proof of work" -- weather forecasting for example.
[/QUOTE]
You mean like 'Folding@home'?
https://foldingathome.org/?lng=en
Upvote
1
(1
/
0)
Indeed. This also happened:
https://www.politico.com/news/2020/04/14/trump-world-health-organization-funding-186786
Upvote
5
(5
/
0)
I worked with someone who built a fully automated sequencing pipeline, in China. Meanwhile, in the US, businesses are slow to automate, because post-docs are cheaper in the short term.
Upvote
0
(0
/
0)
YetAnotherAnonymousAppellation
Ars Praefectus
No, because the many and varied versions of "@home" all require the people involved to donate their computer time.
My point is that if cryptocurrency is going to exist (I wish it didn't, but it does), it makes a lot more sense for the "proof of work" to be actually useful work than for it to be essentially meaningless thumb-twiddling by bazillions of GPUs.
Upvote
2
(2
/
0)
Public health, especially CDC, is part of the national security complex. As is DHS and Department of State and some other federal agencies you may not know.
Anyway, my implied point about political support is that there is no funding to do this work in the federal executive branch, even though you might expect that it would naturally fit within CC's mandate.
Two factors drive what work the bureaucracy does: budget and organization. Both are subject to political whim. If you want to see a task performed by government, fix the political situation first.
Upvote
0
(0
/
0)
VividVerism
Ars Tribunus Angusticlavius
Part of the problem is (IIUC) the "work" that must be proven needs to be difficult to compute but trivial to verify. I could be wrong but it doesn't sound to this layman like "sequence all DNA/RNA sequences in this sample" would qualify.
Upvote
0
(0
/
0)
This sounds great, but is extremely difficult unless we get buy in from countries all around the world. China consistently blocked efforts for the US government to learn about COVID, even preventing the publishing of the genome. How do we expect to get a country like china on board if they refuse to cooperate?
Upvote
0
(0
/
0)
While the idea is interesting, it is completely unrealistic due to that enormous noise.
Very doubtful that it could have been used to prevent the 15 pediatric flu deaths reported so far this season. Just look at the influenza A virus because there have been more than 130 influenza A subtype combinations have been identified in nature, primarily from wild birds, there are potentially many more influenza A subtype combinations given the propensity for virus “reassortment.”
The CDC tested 80,331 flu samples for week 7 (2,441,426 in total since Oct 2, 2022) with 618 positive samples and 53 influenza A viruses detected. Influenza B seems smaller proportion of samples. The good news for the influenza B side is that B/Yamagata has still not been detected.
Now scale that to bacteria and is worse because we still have no idea of the species around and which are nasty, harmless, or mostly harmless.
Upvote
1
(1
/
0)
A lot of DC projects validate by having 2 users run the same work and comparing results. If the algorithm to combine all the snippets isn't deterministic that could make things harder, but if the work units included the random seed to use it should still be doable unless floating point fuzziness is enough to actually scramble results across different architectures instead of just slightly smearing the details. But if that's the case I don't think the base science would be that useful either so it's probably ok as long as the data file sizes are reasonable.
Last edited:
Upvote
0
(0
/
0)