NovaFold AI Powered by AlphaFold 2
Computational biology has come a long way in the past few years, and DNASTAR is leading the way in commercial protein structure prediction. Our NovaFold AI protein structure prediction software enables you to seamlessly utilize the award-winning AlphaFold 2 method to predict the protein structure of an amino acid sequence, and then visualize and analyze your results. Both the prediction and analysis take place within the existing NovaCloud Services interface of Protean 3D.
NovaFold AI uses the world’s most accurate protein structure prediction algorithm
AlphaFold 2 was the top-ranked protein structure prediction method in the CASP14 challenge in 2020, significantly outperforming the other participating teams by determining the structure of proteins with accuracy comparable to laboratory experiments.
In the most recent CASP Challenge, AlphaFold 2 achieved an average Z-score of 2.65 over 92 targets. The closest competitor achieved only 1.00 for the same number of targets. (See graph).
AlphaFold 2 is extremely fast and typically finishes structure predictions in a few hours or less.
NovaFold AI handles even the most difficult folding problems
NovaFold AI is the first commercial software that can accurately predict even the most challenging protein structures:
• Membrane-bound proteins
• Fusion proteins
• Cytosolic domains (CDs)
• Extra-cellular regions
• G-protein-couple receptors (GPCRs)
NovaFold AI can also model multiple domains and their interactions with linkers, also known as multidomain protein structure prediction.
Please see our resources below for more information on NovaFold AI powered by AlphaFold 2.
Do you offer a free trial of NovaFold AI?
One prediction for each of our Nova Applications is included in our Lasergene free trial. We also offer free, no-obligation, personalized demos of any of our Nova Applications with one of our scientists on staff, tailored to your specific research objectives.
What protein folding prediction method does NovaFold AI use?
NovaFold AI uses the AlphaFold 2 artificial intelligence algorithm developed by DeepMind. Of over 140 prediction algorithms tested in the most recent CASP challenge, AlphaFold 2 was not only the most accurate,…
NovaFold AI uses the AlphaFold 2 artificial intelligence algorithm developed by DeepMind. Of over 140 prediction algorithms tested in the most recent CASP challenge, AlphaFold 2 was not only the most accurate, but was an astonishing 2.65 times more accurate than the next-best algorithm.
AlphaFold 2 creates distance maps between residues using a graph neural network based on its training data of ~100,000 protein structures. Part of its prediction process is to compare to existing templates in the included library. Unlike NovaFold, which uses the I-TASSER algorithm, NovaFold AI does not directly use templates to place atomic coordinates. However, NovaFold AI does use templates as part of the map building process.