Calculating 3-D protein structures

16 Replies, 651 Views

(2023-08-15, 01:59 PM)David001 Wrote: ...
AIs don't spit out analytic expressions (anyway, that issue doesn't seem relevant here) an AI simply prints out what it 'thinks' is a most probable answer based on information - including information on the internet.

3)      Unless you start with a set of novel proteins for which 3-D structure have not been determined, and which is then subject to 3-D structure determination after the AI has printed out its results, there is every chance the AI obtains information off the internet - which basically means that it cheats. Remember that AI's were supposed to be super-competent drivers, until they were tested on the roads and a number of pedestrians and others had paid a very heavy price.

David

I understand your skepticism, but AlphaFold was actually trained on a dataset of around 170,000 known amino acid sequences and their corresponding folded 3D structures, taking into account factors such as pressure, temperature, and pH conditions. The notion that it somehow "Googles" information when presented with a new amino acid sequence is indeed far-fetched, to put it mildly. In fact, its high accuracy rate (around 99%) suggests that there may be underlying principles governing protein folding, which is essentially what AI aims to capture: approximating complex, unknown multivariate functions.
(This post was last modified: 2023-08-15, 02:47 PM by sbu. Edited 1 time in total.)
(2023-08-15, 02:46 PM)sbu Wrote: suggests that there may be underlying principles governing protein folding, which is essentially what AI aims to capture: approximating complex, unknown multivariate functions.

i.e., 'intelligence'?  Wink
[-] The following 2 users Like Silence's post:
  • David001, Valmar
(2023-08-15, 08:58 PM)Silence Wrote: i.e., 'intelligence'?  Wink

No Silence 😄, any function an AI (current AI) can emulate can be described mathmatically even though the formula isn’t known. That’s the whole point. I think most “there is something more between heaven and earth” proponents on this forum would argue that intelligence involves aspects beyond mere mathmatical computation.
(This post was last modified: 2023-08-16, 07:32 AM by sbu. Edited 3 times in total.)
(2023-08-16, 07:27 AM)sbu Wrote: No Silence 😄, any function an AI (current AI) can emulate can be described mathmatically even though the formula isn’t known. That’s the whole point. I think most “there is something more between heaven and earth” proponents on this forum would argue that intelligence involves aspects beyond mere mathmatical computation.

My point is that if you give information about phenomena such as NDE's, high quality examples of precognition, genius level performance of various kinds, the best effects of mediums, etc. to an AI, it may regurgitate those results - seemingly absorbing them into the ordinary, but in reality it will simply hide the anomalous nature of what it has been given.

David
[-] The following 1 user Likes David001's post:
  • Valmar
This post has been deleted.
@sbu - I noticed your last post was deleted while I was moving the posts to this new, more appropriate thread and sub-forum (someone can ask to change the title if there's a better one!) I just wanted to make sure that you deleted it yourself, rather than me accidentally. If the latter is so, I can reverse that!
(2023-08-17, 12:14 AM)Ninshub Wrote: @sbu - I noticed your last post was deleted while I was moving the posts to this new, more appropriate thread and sub-forum (someone can ask to change the title if there's a better one!) I just wanted to make sure that you deleted it yourself, rather than me accidentally. If the latter is so, I can reverse that!

Thank you for taking notice. I deleted the post myself having decided I have nothing more to add regarding the computability of protein folding.
[-] The following 1 user Likes sbu's post:
  • Ninshub

  • View a Printable Version
Forum Jump:


Users browsing this thread: 1 Guest(s)