After reading through papers and the code, I'm still not clear if AlphaFold and the likes do actually calculate the 3d structure. My current feeling they are more like information retrieval engines. They find statistical 'similarity' features between the primary structures of the query protein and a database protein (or a family of those) with a known 3d structure, and that known fold is the 'calculated' 3d structure. This would mean that if there is no 'similar' protein with a known 3d fold then the program cannot 'predict' anything. I realize this could be a stupid layman question but here it goes.

  • 1
    $\begingroup$ Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. $\endgroup$
    – Community Bot
    Nov 17 at 23:08
  • $\begingroup$ When you say "calculate", what do you mean? is it "retrieval" when you summarize across a family of similar structures following a sophisticated evolutionary search, comparison, structure-sequence mapping computation, etc.? What's the degree and sophistication of that summary that would mean "calculate" and not "retrieve"? $\endgroup$ Nov 19 at 20:08
  • $\begingroup$ Your suggested formulation of 'retrieval' is very good. Using molecular dynamics to obtain the 3d structure would be a 'calculation'. My conceptual understanding of AF magic is that it 'calculates' the similarity between the primary structures of the query and a DB target with a known 3d structure ( (this calculation is non-trivial, and seems actually like the crux of the matter). Because similar sequences have similar folds, the 3d structure of the DB target is the answer. It is 'retrieval' because the answer is already known but getting it from the database is hard. $\endgroup$
    – biocodz
    Nov 20 at 15:28
  • $\begingroup$ '''the 3d structure of the DB target is the answer''' - this appears to be a fallacy $\endgroup$
    – biocodz
    Nov 20 at 21:59

1 Answer 1


Alphafold calculates the "folds" throughout a protein, thus its protein structural prediction. It uses convolution-based calculation, which is a modification of co-variance. The reason it doesn't use co-variance directly is complicated, but can result in artefacts.

Thus it looks for "simultaneous" amino acid mutations in protein alignments and believes those represent close spatial proximity - based on previous supervised learning of proteins, comprising an alignment and a known structure.

It has won two CASP competitions (world competition for protein folding prediction). The training is likely deeper both in data sets and ANN/CNN/RNN "layers" (deep learning jargon) than previous protein structural prediction models and hence its success. "Deepmind" the company behind it will be Tensorflow experts, rather than structural biologists.

I don't use it but have seen its structural accuracy first-hand. The accuracy statistic RMSD was impressive.

It is important to note it is built on the back of a long term effort by this community as to calculating protein structural prediction. So it's not "radical technology", they've generally increased the accuracy of established methods. Again they've likely done this with more extensive training.

  • $\begingroup$ '''AF calculates the "folds" throughout a protein''' - if this means finding 'piecewise/partial folds' with subsequent 'concatenation' into the final 3d structure, then it would be a 'calculation' indeed. $\endgroup$
    – biocodz
    Nov 20 at 16:47
  • $\begingroup$ Thats the gist @biocodz $\endgroup$
    – M__
    Nov 20 at 19:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.