NVIDIA, Google DeepMind and research partners released predicted protein-complex structures for more than 2,800 viruses through the AlphaFold Database. The structures were inferred with AlphaFold2 using NVIDIA tooling to scale the work. NVIDIA also released the workflow used to produce the dataset.
What has been made available
NVIDIA describes a collaboration involving Google DeepMind and EMBL-EBI that adds predicted viral protein complexes to the AlphaFold Database. The work uses AlphaFold 2 alongside NVIDIA computing tools and an open structure-prediction pipeline.
The release includes confidence information to help researchers assess the predictions. Making the dataset accessible lets other researchers inspect the proposed structures and choose questions for further investigation.
From a prediction to an experiment
A predicted structure is a hypothesis about molecular arrangement. It does not by itself demonstrate a biological interaction, validate a drug or establish a treatment. Experimental follow-up remains necessary to test the scientific conclusions drawn from these models.
Original source
This report summarises the source below. Analysis is labelled separately; product and research claims remain attributed to their source.
Read the original at NVIDIA