MAGE Model Generates Human Antibody Chains Directly from Antigen Sequence
Start here
Researchers at Vanderbilt University, led by Ivelin Georgiev, published a model called MAGE (Monoclonal Antibody GEnerator) in the journal Cell in December of last year. Given only the amino acid sequence of an antigen protein, MAGE outputs complete human antibody heavy and light chains, without needing a template or structural information. The model also works on antigens that were not part of its training data. MAGE is built on ProGen2, a 2-billion-parameter language model pretrained on 2 billion general protein sequences. The publication describes the method and its demonstrated capabilities; no further results from the group have been covered since.
Written by a model from the reports · updated 2 hours ago
Coverage timeline
Follow the coverage to see the different angles.
- mp.weixin.qq.comCell Paper Describes MAGE Model Generating Human Antibody Heavy and Light Chains from Antigen Sequence
In December last year, Ivelin Georgiev's laboratory at Vanderbilt University published the MAGE (Monoclonal Antibody GEnerator) model in Cell. Given the amino acid sequence of an antigen protein, it outputs complete human antibody heavy and light chains, with no template or structural information required, and it works on antigens absent from its training data. MAGE is built on ProGen2, a 2-billion-parameter model pretrained on 2 billion general protein sequences.
Heat trend for this story
Heat now 7·Peak in the comparable set 10 (Oct 24 18:00)·Change over 24h in the comparable set –·Participants now 1
Participants counted above, heat below. The trend compares only the participants observed in full throughout, so its set can be smaller than the heat figure itself. Move the pointer or click the chart for an hour's reading; the left and right arrow keys work too.