Protein structure and design: the papers compared side by side

A real example: what you get

5 papers compared on 5 columns.

Columns: Question, Method, Data, Main results, Limitations
19 of the 19 cells with their passage of the paper
4 points where they agree
3 points where they disagree
6 questions none of them addresses
Cells you correct by hand, CSV export, a chat with the papers

Where the papers disagree

The design studies differ in the specificity of their reported validation. One describes characterization of hundreds of designs and a cryo-EM match without a numerical structural deviation; the other reports 310 characterized proteins and two crystal structures with backbone RMSD around 1.0 Å. These results do not establish which framework is more accurate.

Its source We demonstrate the power and generality of the method, called RoseTTAFold diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of designed symmetric assemblies, metal-binding proteins and protein binders. — De novo design of protein structure and function with RFdiffusion · Data

The first results

Highly accurate protein structure prediction with AlphaFold. Question: Develop a computational method that predicts protein structures with atomic accuracy, including when no similar structure is known.
Accurate structure prediction of biomolecular interactions with AlphaFold 3 Question: Predict the joint structures of complexes containing diverse biomolecular components within a single model.
De novo design of protein structure and function with RFdiffusion Question: Develop a general protein-design framework supporting tasks such as binder design, symmetric assemblies and functional motif scaffolding.
Illuminating protein space with a programmable generative model Question: Generate novel protein structures and sequences while allowing users to steer generation toward desired properties and functions.
ColabFold: making protein folding accessible to all Question: Accelerate protein structure and complex prediction and make it freely accessible through Google Colaboratory.

See a complete example, explained step by step →

Create a free account, then Premium runs it on your own work. Continue with Google Create a free account Log in

The whole result, as its owner would share it →

Other examples · Biology and medicine

Every example →