5 ms·
I love AI. In comparision to a lot of other technologies, we actually have jumps in quality left and right, great demos, new things which are really helpful.
by Meniceses 2y ago
I love AI.
In comparision to a lot of other technologies, we actually have jumps in quality left and right, great demos, new things which are really helpful.
Its fun to watch the AI news because there is something relevant new happening.
I'm worried regarding the impact of AI but this is a billion times better than the last 10 years which was basically just cryptobros, nfts, blockchain shit which is basically just fraud.
Its not just some GenAI stuff, we talk about blind people getting better help through image analysis, we talk about alpha fold, LLMs being impressive as hell, the research currently happening.
And yes i also already see benefits in my job and in my startup.
- bamboozled 2y agoI’m truly asking in good faith here because I don’t know but what has alpha fold actually helped us achieve ?
- Meniceses 2y agoIt allows us to speed up medical research.
- bamboozled 2y agoIn what field specifically and how ?
- scotty79 2y agoAre you asking what field of science or what industry is interested in predicting how proteins fold? Biotechnology and medicine probably. Pipeline from science to application sometimes takes decades, but I'm sure you can find news of some advancements enabled by finding out short, easy to synthesize proteins that fit particular receptor to block it or that process some simplified enzymes that still process some chemicals of interest more efficiently than natural ones. Finding them would be way harde without ability to predict how a sequence of amino-acids will fold. You'd need to actually try to manufacture them then look at them closely. First thing that came to my mind as a possible application is designing monoclonal antibodies. Here's some paper about something relating to alpha fold and antibodies: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10349958/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10349958/
- RivieraKid 2y agoI guess he's asking for specific examples of AlphaFold leading to some tangible real-world benefit.
- scotty79 2y agoWait a decade then look around.
- bamboozled 2y agoParent is right, I’ve heard mixed things about alpha folds practical usefulness. As brilliant as what it does is…
- Meniceses 2y agoAre you phishing for something or are you not sure how this actually works? Everyone who is looking for proteins (vacines, medication) need to find the right proteins for different cases. For attaching to something (antibody design), for delivering something (like another protein) or for understanding a disease (why is this protein an issue?). Covid research benefitted from this for example. You can go through papers which reference the alphafold paper to see what it does: https://consensus.app/papers/highly-protein-structure-prediction-alphafold-jumper/dbb40b7fb49c5f0f8ce0ba47ebc58972/ https://consensus.app/papers/highly-protein-structure-predic...
- bux93 2y agoNo such thing as a stupid question. It's a question that this paper in Proteomics (which appears to be a legit journal) attempts to answer, at least. https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/full/10.1002/pmic.202200128 https://analyticalsciencejournals.onlinelibrary.wiley.com/do...
- Meniceses 2y agoI didn't say stupid but sometimes people asking in a way which might not feel legimate / honest.
- bamboozled 2y agoYou were making the claim, so I thought you’d have an answer on hand. I’m wondering if it might be you who was commenting in bad faith ?