Now that the
#alphafold
hype has completely died down (ha!), I've written a new blog post on the AF2 method paper: . This is a technical deep-dive into aspects of AF2 that I find most surprising/innovative and of relevance to broader biomolecular modeling.
@MoAlQuraishi
Thank you for this! Re: your comment "From that humble {s_i}, the entire structure is projected onto 3D space, without any kind of explicit 3D reasoning" - I'm not sure I follow, since the {z_{ij}}'s, i.e., the refined pair representation also feeds into the structure prediction
@MoAlQuraishi
Beautiful write-up, and very profound thoughts on the details of the architecture. Looking forward to see your review paper, and excited to see what the community as a whole will do with all these novel ideas.
@MoAlQuraishi
Amazing read, thank you! One thing that is still unclear to me is what is the advantage to use structure crops over some kind of biological definition of domains!