5 ms·
There are two really interesting parts of Russian medicine which the West hasn't really pursued which it should. Phages is one and they really are an incredible
by PaulKeeble 4y ago
There are two really interesting parts of Russian medicine which the West hasn't really pursued which it should. Phages is one and they really are an incredible solution to antibiotic immune bacteria, the issue is always finding the right one, there are so many different phages and bank of all of them would be enormous, 10s probably 100s of thousands.
The second is auto vaccines and Lysates which present dead bacteria to the body so it can create an appropriate immune response. They are used in Russia for things like Urinary tract Infections which in the West are often untreatable with antibiotics but they do seem to work. Get the immune system to recognise the baddy and work out how to kill it and then in the harder to reach places it can be effective. Works fairly well for Chronic infections but the issue is identifying the bacteria precisely to match and existing mix or collecting the bacteria and cultivating and killing it mechanically or with heat so it can be injected back into the patient safely in other places in the body.
Both techniques were outrun by antibiotics in the beginning, antibiotics are a lot easier to administer as you don't have to identify the pathogen precisely and its easy to make in bulk. But both these more natural solutions are a lot harder for bacteria to become immune to and have undergone the evolutionary process alongside us and them. I think both are worth serious efforts and development.
- anewpersonality 4y agoWhy don't we use machine learning to find phages?
- Fomite 4y agoWhat makes you think we don't?
- pyuser583 4y agoMachine learning isn’t great at everything. Is “finding phages” one of the tasks it good at. I remember when Google was claiming they could track the Flu based on searches, it was going to revolutionize tracking pandemics.
- Fomite 4y agoThere's a decent amount of work being done, broadly, in using machine learning to identify resistance mechanisms and work in antibiotic discovery. There's likely applications there for phage as well. But the real problems holding phage back is not finding them (they're trivial to find in the wild) but operationalizing them. For me, the most foreseeable application of ML to phage work is predictive models of antibiotic treatment failure.
- chasil 4y agoIdeally, the bacteria can't become immune. The virus evolves and adapts with the bacteria, modifying its receptors and modes of action as the host survivors mutate, which is beyond the capability of a chemical antibiotic. This is why it is so important to pre-treat the phage in a concentrated solution of the target pathogen prior to injection into a patient, which allows the virus to maximize infectivity.
- elcritch 4y agoIt seems that’d help make semi “broad spectrum” possible as well. Manufacture a serum loaded with hundreds or thousands of phages known to target superbugs and let them decimate the bacteria. Choose the survivors. If you could automate the last step it could be something any old doctors office could do.
- chasil 4y agoIt is important to carefully match the phage with the target pathogen. A mismatched phage will not be effective. This tool can't be used like antibiotics. It requires expertise in classification.
- saiya-jin 4y agoThat would explain why urinotherapy may work. Expose your immune system to pathogens that are in place unreachable to it, to mount response (multiply antigens that work on them) and voila. Just that whole idea is rather hard to swallow...
- cassepipe 4y agoWait, isn't urine sterile?