Skip to content

Illumina AI – ChatGPT for your genome…



Get my A.I. + Business Newsletter (free): https://natural20.com/ #genomesequencing #ai #deepmind TIMELINE 00:00 …

39 thoughts on “Illumina AI – ChatGPT for your genome…”

  1. one more really really interesting fact, it used to take us before Alpha Fold a PHD of someone to determine the structure, which would take in average around 2-3 years, right now? 15 min .. yeah let that sink in.
    There was a breakthrough around a months go where it helped us determine the reason of extinction of an animal from around 2.5 million years ago if am not mistaken, again .. let that sink in.

  2. As a student working in the field of synthetic biology. I do believe that AI has a lot of potential applications, but I have no way of knowing how close we are to some of them. I believe that disease prediction is the most realistic, but that could lead to problems with life insurance companies charging customers more if they are genetically predisposed to certain conditions. Also, the creation of GMOs has been rudimentary up until recently (IE: putting a gene in with a highly active promoter or a leaky promoter that responds to certain stimuli) and trying to create more complex systems in cells is incredibly challenging due to the unbelievable number of potential cross reactions that could occur, leaky promoters leading to unintended background expression, etc. The problem lies more in the fact that there isn’t a one solution fits all approach that could work even for some of the “simplest” of fixes. While I believe that trying to get an Ai to help design the solutions is a goal worth working towards, I do not believe we have enough data and real world examples to train the Ai on making actual solutions. Essentially, the solutions we have made so far have flaws and if the Ai trains using those flaws it could wind up doing more harm than good. You put crap in you get crap out.

  3. Many meaningful discoveries are on the horizon. More discovery and powerful tools will enhance our understand of the language nature writes software in. It's important to emphasize how successful and pervasive this software is! The programs nature evolves operate in the physical reality. Understanding that is not just a revolution in biology, but also in nanotechnology.

  4. This needs much more ethical, cautious approach, checks and balances,much more than AI, generating unknowns to ensure potential irreversible cock-ups leading to 🤷‍♂️☠️

  5. Holy shit, i was working in this for a long time now, this is just glorious that our most briliant experts finaly gave a chance to combine the power of AI and biology

  6. A team of pharmacologists, biochemists & engineers already worked out the concept of this back in 2000, now that the theory is in place, any company with machine learning capabilities can apply it.

  7. as someone who is playing with small models for fun and seeing the difference a little more power gives, i am stoked someone is putting that much juice into diabetes and such.

  8. Ok, So flesh computor? Think about it, using this for organic pc parts, I do not see why you would want to do that but it'd be like Scorn

  9. This is the most underwhelming video I’ve watched on AI, so far. Basically, Illumina included AI in a software to do GWAS studies. Duuh!

  10. Can people stop trying to make lifeforms that will eat that fact that its not biodegradable is one of its biggest strengths if they succeed then either everything will have to back to being metal or stone(ceramics usually) which are heavy and unwieldly and rare or plastic will have to be treated in a way that stops them from being biodegradable like wood is which puts us right back at square one but with extra steps and more expense and waste. Plastic eating bacteria is fucking stupid! Just stop!

  11. All good with the exception of access to the tech, which historically has been denied to those below, an or who are not part of the ‘in-group’. Eugenics related policy has become the unspoken drive for those who continue to seek to control the wealth, and thus access to the most valued commodity ,(which is information).

  12. Does the electron repulsion of a proteins molecular configuration act as a limiter to what shapes it can fold into or act as a shortcut/driver for the shape it ultimately settles into?

Comments are closed.