// StartMathJax Script window.MathJax = {loader: {load: [ 'input/asciimath', 'ui/lazy', 'output/chtml', 'ui/menu']} }; (function() { var script = document.createElement('script'); script.src = "https://cdn.jsdelivr.net/npm/mathjax@3/es5/startup.js"; script.async = true; document.head.appendChild(script); })(); // UpdateTypeset Script config = { attributes: true, childList: true, subtree: true }; // Callback function to execute when mutations are observed callback = (mutationList, observer) => { for (mutation of mutationList) { if (mutation.type === 'childList') { console.log('A child node has been added or removed.'); MathJax.typeset(); } else if (mutation.type === 'attributes') { console.log(`The ${mutation.attributeName} attribute was modified.`); } } }; // Create an observer instance linked to the callback function observer = new MutationObserver(callback); document.onreadystatechange = () => { if (document.readyState === 'complete') { console.log("Loaded fully according to readyState") targetNode = document.getElementById('content-wrapper') console.log(targetNode) // Start observing the target node for configured mutations observer.observe(targetNode, config); } }
top of page

A summary on research paper of AlphaFold protein model structure search

The original paper can be found here. All figures and tables are directly from the paper.


AlphaFold is a structure prediction model in structural biology. It consistently produce reliable computational structure models with atomic accuracy. The latest version, AlphaFold2 generate structure models that are so accurate which would be something expected from X-ray crystallography. Soon after the release of the AlphaFold2 code, predicted structure models by AlphaFold2 for proteins from 21 major model species have been released at the AlphaFold Protein Structure Database.


In this paper, The authors offered a real-time search algorithm, 3D-AF-Surfer, which finds proteins with similar structures once an protein structure is offered. In 3D-AF-Surfer, structures are represented with 3D Zernike descriptors (3DZD), which is a rotationally invariant, mathematical representation of 3D shapes.


Both AlphaFold and 3D-AF-Surfer utilised accumulated protein sequence andstructure information in the Protein Data Bank (PDB).


This paper measured the performance of the method using Accuracy, Precision, Recall and F-measure.


The result showed that the search algorithm provided in this paper outperforms other protein search method, achieving a real-time search result.

For secondary structure classes classification, a machine learning method is used: a bagged ensemble of support vector machine classifiers (SVMs) using the secondary structure content of SCOPe domains. The bagged ensemble had an accuracy of 91.5%.

2 views0 comments


bottom of page