Protein design

Narrow these by company or desk

Thu, Sep 10Protein design
Deep learning maps contact points between hepatitis B capsid and envelope
Virologists used mutation-driven deep learning alongside evolutionary sequence analysis to map binding sites between the hepatitis B capsid protein and its large envelope protein LHBs.
Carolanne van Belleghem, Sara Rescalli, Mathilde Briday, and 6 othersCarolanne van Belleghem, Sara Rescalli, Mathilde Briday, Christophe Combet, Mario Enrique Cano Contreras, Lauriane Lecoq, Marie-Laure Fogeron, Alessandra Carbone, Anja Böckmann fewer authors
CNRS; Sorbonne; INSERM; Institut Universitaire de France
bioRxiv
Thu, Sep 10Protein design
Engineered natural ligand boosts cell therapy against cancer target CD70
Researchers combined deep learning and molecular dynamics simulations to re-engineer CD27, the natural binding partner of the pan-cancer target CD70, for use in chimeric antigen receptor (CAR) T-cell therapy.
Amrik S Kang, Radhika Dalal, Mingcheng Li, and 23 othersAmrik S Kang, Radhika Dalal, Mingcheng Li, Anupam Anand Ojha, Sujata Walunj, Caitlin E Cornell, Ruyan Rahnama, Abhilash Barpanda, Sanjana Prudhvi, Trishna S Patel, Mark Aspinall-O’Dea, Beth Walker, Clare Adams, Veronica Steri, Paul Phojanakong, Juan Antonio Camara Serrano, Hua Li, Yan Zeng, Jose Rivera, Daniel A Fletcher, Benjamin J Huang, Sonya M Hanson, Pilar Cossio, Tanja Kortemme, Justin Eyquem, Arun P Wiita fewer authors
University of California, San Francisco; Flatiron Institute; University of California, Berkeley; and 3 othersUniversity of California, San Francisco; Flatiron Institute; University of California, Berkeley; Charles River Laboratories England; Parker Institute for Cancer Immunotherapy; Biohub fewer institutions
bioRxiv
Wed, Sep 9Protein design
Geometric algorithm matches deep learning on collagen structures at a fraction of the cost
Researchers built the Collagen-specific Deterministic Structure Modeler (CDSM), an empirical geometric modeling pipeline that predicts all-atom collagen triple-helical structures without large neural networks.
Bruno Howard, Maxwell Bauer, Markus J. Buehler
MIT
bioRxiv
Wed, Sep 9Protein design
Screening 1,700 AI-designed protein binders pinpoints key failure modes in CAR T cells
Bioengineers tested 1,758 de novo protein binders designed with generative AI against tumor targets BCMA, CD19, and CD22 to evaluate their real-world function as chimeric antigen receptors (CARs).
Arthur Chow, Hoyin Chu, Ruofan Li, and 6 othersArthur Chow, Hoyin Chu, Ruofan Li, Benan N Nalbant, Abdul Vehab Dozic, Laura C Kida, Zeyu Tang, Joseph R Palmeri, Caleb A Lareau fewer authors
Memorial Sloan Kettering
Nature Biomedical Engineering
Mon, Sep 7Product launch Foundation model
AtlasFold predicts protein and complex structures without sequence alignments
AtlasFold is an open-source protein structure prediction model that generates three-dimensional atomic coordinates directly from single protein sequences without requiring evolutionary multiple-sequence alignments.
Seonghwan Seo, Hyeongwoo Kim, Seokhyun Moon, and 2 othersSeonghwan Seo, Hyeongwoo Kim, Seokhyun Moon, Woo Youn Kim, Team KAIST fewer authors
KAIST; Seoul
bioRxiv

Protein design

Deep learning maps contact points between hepatitis B capsid and envelope

Engineered natural ligand boosts cell therapy against cancer target CD70

Geometric algorithm matches deep learning on collagen structures at a fraction of the cost

Screening 1,700 AI-designed protein binders pinpoints key failure modes in CAR T cells

AtlasFold predicts protein and complex structures without sequence alignments

Stay ahead of the field.