ampdepolymerase

Congratulations to the team! A production ready lipid micelle with reliable human cellular uptake is non trivials. Kudos from the biotech community :)

Funnily enough the reason why virology therapy (something the Soviets did a lot of research in cuz they didn't have good antibiotics and biotech/pharma) hasn't seen much success at scale is precisely because the body's immune system adapts too quickly. Maybe we will see more success now with AI but it will be a cold day in hell before the powers that be let any garage biohacker have access to LLMs trained on this sort of information.

Why would you modify E coli for this? Use existing gut friendly bacteria and start from those. There's a wealth of material on the genetics of probiotics.

Can you add support for touch?

The phospholipid micelles are non-trivial (trade secret) to make and it's the major reason why African nations and other countries have not been able to successfully create mRNA vaccines at scale.

A cell is a bundle of proteins wrapped in a membrane that's sort of an oil drop (or as another comment said, a fat bubble). In biology it's called a phospholipid bilayer. Fun fact you can actually "merge" cells together with the help of certain viruses. Drug delivery usually involves moving molecules though this phospholipid bilayer which involves all sorts of tricks. There are pores and receptors on the membrane that can selectively bind to different biochemical molecules and proteins. A good chunk of research in bioinformatics, chemoinformatics, quantum computing is focusing on simulating protein binding dynamics and protein-protein interactions on various levels so we can design drugs that can bind to the receptors we want. (Alphafold made this a lot easier to figure out how to go from a sequence of genetic material to a specific protein shape) A RNA vaccine is kinda like a virus in that it has to be taken into a so the cellular machinery (ribosomes) can build the protein that it codes for. So having a micelle (or nanoparticle, whatever you want to call it) that can get absorbed and merged into the cell that you are targeting specifically is a Big Deal.

> getting infected with the zika virus (probably the best thing to do IMO

Virology based methods don't last very long because the immune system adapts quickly. If you want to go down that route, make sure you have experts on hand.

You should try to convince the others on the Facebook groups to undergo full genome DNA sequencing (around $1K USD, usually not covered by insurance, I would pay for the other patients if I were you, their data is more than worth it given it’s your life on the line) and submit the data to patient networks and orphan drug groups. There are lots of bioinformatics methods now (thanks to mostly advances in ML among other things) that can derive insights into the problem, without any physical assays or laboratory tests.

> the mitochondria is the powerhouse of the cell

The explanation doesn't get much better at higher levels. You have the Krebs cycle which biology people religiously memorize but it doesn't really explain much either. The actual interesting part is usually handwaved away as "magical enzyme/protein" catalysis. Understanding how the mitochondrial proteins/enzyme catalysts function would usually require a graduate degree, and maybe a background in biochemistry and biophysics.

Also "off target effects".

Work/volunteer as a minimum wage software developer lab rat (or whatever your day job speciality is). There are plenty of labs that are in need of free labor when it comes to software/engineering support in general, just ask around.

Most of the pop sci books are useless for practical use cases, and similarly the Feynman lectures self select for the physics/mathematically inclined.

Biology is a leaky abstraction, it's very hard to do anything with rigor without having a strong foundation in the fundamentals. You see the same discussion on hacker news when it comes to music, people are more interested in mapping programming concepts to music notation and complaining about western music presentation than the music itself. For biology, you need need to have a firm understanding of the central dogma and biochemistry if you want to do anything beyond surface level empirical trial and error. Most people, especially the "hacker types", only have a vague understanding of the former i.e. DNA translation and transcription and that's about the limit. You absolutely have to gain an intuition for biochemistry if you want to do things with rigor, otherwise you will just be the biotech equivalent of a bootcamp web developer, fit for washing test tubes and not much else.

No, no, no. Explanation is inaccurate. The central dogma is the central dogma because information flow is primarily one way (yes there are exceptions i.e. epigenetics but primary information flow is from DNA to protein) and this is the fundamental tenet underpinning our current understanding of biology.

> A multicellular organism gametes divide only a handfull of times per generation.

That's what recombination is for during sexual reproduction.

> It does not contain informtion how to assemble the molecular machines from the proteins, or how "to be alive" in general.

Unless you subscribe to the creationist or Larmarckian schools of thought, this is flat out wrong. DNA polymerase and similar analogues like reverse transcriptase do not exist in a vacuum. There are entire branches of evolutionary biology dedicated to studying their formation. The main transcription proteins, helicase, polymerase, and ribosomes can all be assembled from the basic proteins they themselves transcribe. (Incidentally figuring out the ultimate structure that a protein chain assembles into is what AlphaFold does, bioinformatics and ML's crowning jewel)

> There is heritable information outside of DNA, that is epigenetic in its nature.

When biologists refer to epigenetics, they mean information carriage that's not strictly tied to the nucleotides. This doesn't meant DNA isn't involved. Most epigenetics I can think of off the top of my head all involve the DNA transcription mechanism in some way.

> This is still relevant to his concern, but from the other end. They might be making the labor artificially scarce to increase pay.

This is very much true. I find that a lot of people in tech seem to put healthcare on a pedestal and believe that the professionalisation and gatekeeping of the industry create a better outcome than other engineering fields. This is very much untrue, the healthcare field is in need of massive disruption and lobbying to increase labor supply. You are being downvoted because a lot of tech people here hate to imagine that healthcare at the highest level is still subject to market forces like everything else. Medical training is being severely gatekept and hindered via the current apprenticeship/residency system. After all, we call the worst medical student, doctor. If you want to improve healthcare, tie medical school admission to the MCAT score, and only the MCAT score. You are not going to get better doctors just because candidates spend their summers building houses in some impoverished third world country.

Note FDA rules require them to retain the data for a few years.

There is a bigger iceberg lurking around. Look up the people who control medical school accreditation and funding. They have more blood on hands than many other industries put together. Medical innovation has been slowed down by magnitudes because of their gatekeeping.

More specifically pseudo science not guided by empiricism. The line between science and pseudo science is very thin. Scientific theories are often incorrect, especially in life science, but we refine upon them based in empirical and observed data.

Voit looks gimmicky. Uri Alon's treatment (An Introduction to Systems Biology) is much better.

There is a bottleneck in the number of MD positions. Without the MD title, your typical life science researcher cannot easily carry out medical research. Medical education and training has a scale problem.

I won't put too much stock in unglazed clay cooking ware.

The best C and C++ package manager I have used is Xmake/Xrepo. They support all the major package repositories out of the box including Vcpkg and Conan.

https://github.com/xmake-io/xmake#supported-package-reposito...

Human trials for drugs is the expensive part. If the situation is urgent, look into compassionate use and right to try. If you have time, the fastest way to get it to production is to have the trials done in a lower cost country. At the end of the day if you control the experiment and manufacturing pipeline, then you can make a develop a drug for a much lower price. You absolutely cannot cut corners in the experimental and engineering stage even if you do it in a more compliant regulatory regime.

I haven't looked into the Moderna vaccine but in general there are many ways to get rid of a virus. Using the typical B cell + viral glycoprotein is the most obvious way as viral activity is hard to detect when replicating inside a cell but there are many stages in the viral replication process which can be inhibited. Inhibiting protease cleavage of the viral polyprotein or preventing polymerase formation are all possible solutions. They don't really count as vaccines though.

Fwiw, rarely can lab designed viruses i.e. made through splicing have features that are more "optimal" (or efficient) than those that are evolved. Directed evolution is truly marvelous when done properly (it won a Nobel!).

Just because the bacteria is dead doesn't mean the toxins it produced have been decomposed by the heat. Leaving food out is very bad idea.

The math sounds nonsensical. The risk adjusted exposure from a reactor should definitely be higher as you would be closer to the threshold of DNA bond breaking. Being stabbed once a year isn't necessary better than getting hundred paper cuts. Any actuaries here care to comment?

You don't, not without either resequencing it with another sequencing system or benchmarking the sequencer with a known sequence.

GC rich regions as in hairpin loops? How would the sequencer deal with those?

I recommend reading this review

https://genomebiology.biomedcentral.com/articles/10.1186/s13...

I guess there are limits to ensemble methods if the underlying accuracy doesn't increase. I don't work on gene sequencing algorithms but from what I understand of ML ensemble techniques, there are certain assumptions regarding the underlying independence of the errors. The errors for nanopore should be uniform but I am not sure. Any molecular biologist here care to comment?

The one I linked to is a decade out of date and OEM discontinued.

A used laboratory grade NGS system can be had for less than 10K

https://www.ebay.com/itm/265148387179

Nanopore is still not quite ready yet for precise and high accuracy sequencing. Give it another five years.