4 ms·

This is my field (Im a PhD student who writes PPI inference models). I skimmed the article and start-up website and I'm a bit confused. PPI inference is not b

This is my field (Im a PhD student who writes PPI inference models). I skimmed the article and start-up website and I'm a bit confused. PPI inference is not binding affinity prediction is not binding site prediction, despite being related tasks. There are billions of PPI pairs in public datasets, there is much less binding affinity data, and even less binding site data. (Side-note: if you're hiring PPI / deep learning / comp. bio people, send me an email at the address in my bio.)

abhishaike 2y ago
Sorry yeah, I muddled with the definitions a bit, this is focused on binding affinity data. Afaict, the primary source for such data is https://en.wikipedia.org/wiki/PDBbind_database https://en.wikipedia.org/wiki/PDBbind_database, which is quite small
bglazer 2y ago
Genuinely curious, how is binding affinity prediction not PPI prediction? Isn’t a PPI just a binarization of affinity?
- jszymborski 2y ago
  I mean, yes, but it's similar to how image segmentation entails image classification (if you can segment the dog you can classify the image as one containing a dog). There's oodles of labelled images of dogs, but comparably much fewer datasets of dog silhouettes. Another factor is that it's far easier (and less informative) to predict that two proteins are capable of interacting with any degree of affinity than with a specific amount. You may say to yourself (as I once did): "Well surely a well calibrated PPI inference model will output interaction probabilities that correlate with binding affinity!" I've tested this and I've yet to find one written by myself or others that behaves this way. If this line of questioning is interesting to, definitely sign up for Google Scholar alerts on my name because we're publishing some very cool stuff on precisely this v. soon.
  - bglazer 2y ago
    Really interesting point about the non-correlation of affinity and PPI probabilities. Disappointing, honestly. Mechanistic systems bio models would really benefit from large scale affinity information, and it would be pretty cool if ML methods trained on binary PPI data learned a kind of latent affinity model. Maybe PPI models could be fine tuned to do that? Or maybe there’s specific neurons in the PPI models that correlate better with affinity, similar to the mechanistic interpretability stuff that people are doing on LLM’s? I only follow this area tangentially so I imagine I’m not the first person to have those ideas… Anyways I look forward to seeing what you publish!

abhishaike 2y ago

Sorry yeah, I muddled with the definitions a bit, this is focused on binding affinity data. Afaict, the primary source for such data is https://en.wikipedia.org/wiki/PDBbind_database https://en.wikipedia.org/wiki/PDBbind_database, which is quite small

bglazer 2y ago

Genuinely curious, how is binding affinity prediction not PPI prediction? Isn’t a PPI just a binarization of affinity?

jszymborski 2y ago
I mean, yes, but it's similar to how image segmentation entails image classification (if you can segment the dog you can classify the image as one containing a dog). There's oodles of labelled images of dogs, but comparably much fewer datasets of dog silhouettes. Another factor is that it's far easier (and less informative) to predict that two proteins are capable of interacting with any degree of affinity than with a specific amount. You may say to yourself (as I once did): "Well surely a well calibrated PPI inference model will output interaction probabilities that correlate with binding affinity!" I've tested this and I've yet to find one written by myself or others that behaves this way. If this line of questioning is interesting to, definitely sign up for Google Scholar alerts on my name because we're publishing some very cool stuff on precisely this v. soon.
- bglazer 2y ago
  Really interesting point about the non-correlation of affinity and PPI probabilities. Disappointing, honestly. Mechanistic systems bio models would really benefit from large scale affinity information, and it would be pretty cool if ML methods trained on binary PPI data learned a kind of latent affinity model. Maybe PPI models could be fine tuned to do that? Or maybe there’s specific neurons in the PPI models that correlate better with affinity, similar to the mechanistic interpretability stuff that people are doing on LLM’s? I only follow this area tangentially so I imagine I’m not the first person to have those ideas… Anyways I look forward to seeing what you publish!