Researchers have developed CoSiNE, a novel deep learning model for antibody engineering that moves beyond traditional methods. Unlike approaches that treat antibody sequences as independent samples, CoSiNE explicitly models the evolutionary process, incorporating information from affinity maturation. This allows it to better capture complex epistatic interactions and disentangle selection from somatic hypermutation, outperforming current language models in variant effect prediction. AI
IMPACT This model could accelerate the discovery and optimization of therapeutic antibodies by providing a more accurate prediction of sequence evolution and binding affinity.
RANK_REASON The cluster contains a research paper detailing a new AI model for antibody engineering. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
