InstaDeep’s Nicolás Lopez Carranza and Marcin J. Skwark recently presented the DeepChain protein design platform at the highly prestigious ISMB/ECCP 2021 conference. Speaking to a large virtual audience of expert bioinformaticians and computational biologists, Nicolás (DeepChain Lead) and Marcin (Bioinformatics Lead) explained how DeepChain helps life science researchers design and optimize protein-protein interactions, and how that is directly applicable to drug discovery.
The pair demonstrated the novel, accessible and user-friendly platform during the presentation, from initial PDB file input, right through to applying AI-directed mutagenesis, where DeepChain finds a biologically plausible sequence of mutations that improve binding energy.
They went on to outline the platform’s simulation-based scoring metric, based on a realistic, fully-flexible macromolecular simulation; as well as the open-source DeepChain Apps framework that enables users to build custom metrics based on neural networks trained on amino acid sequence data, thus leveraging the predictive power of pre-trained embeddings. The presentation was followed by a Q&A session for interested viewers.
The annual international conference on Intelligent Systems for Molecular Biology (ISMB) is the flagship meeting of the International Society for Computational Biology (ISCB); and is the world’s largest bioinformatics and computational biology conference. In 2021, ISMB joined forces with the European Conference on Computational Biology (ECCB) to provide an intense multidisciplinary forum for disseminating the latest developments in bioinformatics and computational biology.
Nicolás and Marcin’s presentation is available to view at the top of this post; and more information and video recaps from ISMB/ECCB 2021 are available at www.iscb.org/ismbeccb2021
