Somatic hypermutation (SHM) involves activation-induced cytidine deaminase (AID)-mediated DNA targeting, followed by error-prone repair which introduces mutations. However, existing computational models do not reflect this, limiting their ability to model the mechanism of affinity maturation in silico. Here, we develop a biologically grounded, transformer-based framework that mirrors SHM. Our framework employs two independent antibody language models (AbLMs): an AID-like targeting model to select mutation sites and a DNA repair-like substitution model to predict resulting amino acids.
We sequenced memory B cell repertoires from eight healthy donors using high-accuracy bulk NGS of unpaired VH/VL chains. These data were used for pretraining the two AbLMs to perform in silico SHM. Using AntiBERTa2-derived metrics as a correlate for native-like antibodies, we compared model-generated sequences to the natural human immune repertoire.
Bulk NGS produced ∼84 million high-quality, productive memory B cell receptor sequences for AbLM training. Model predicted mutations matched the spatial distribution and overall load of true SHM. AntiBERTa2 embedding projections and log likelihood scoring revealed that model-generated sequences closely resemble native antibodies and strongly mimic SHM mutational patterns. Model-generated sequences also exhibited significantly increased levels of expression in HEK293F cells, indicating learning of beneficial mutations that enhance protein fitness.
Our framework models SHM to generate native-like human antibodies, providing a biologically grounded tool for guiding design within the natural sequence space. Unlike motif-based or single-stage SHM simulators, our framework explicitly decouples targeting and substitution to enable faithful reproduction of both hotspot localization and amino acid substitution. This approach advances in silico modeling of humoral immunity and may reveal new mechanistic insights into antibody clonal dynamics.
Endowed Fellowship in the Skaggs Graduate School of Chemical and Biological Sciences, Achievement Rewards for College Scientists (ARCS) Foundation San Diego Chapter, National Institute of Allergy and Infectious Diseases (NIAID)
Computational and Systems Immunology (COMP)
Karenna Ng, B. Némoz, Bryan S. Briney· Journal of Immunology· 0 citations
Epitope:paratope interactions define the breadth and affinity an antibody has towards an antigen. Computational structure prediction can help to model immune complexes as a method of in silico screening for broadly neutralizing antibodies (bnAbs). However, we do not currently understand how structural predictions and confidence metrics correlate with experimental data. Here, we evaluate the accuracy of AlphaFold 3 (AF3) in predicting immune complexes of a known class of bnAbs towards a conserved epitope on influenza hemagglutinin (HA).
We used single-cell RNA sequencing to generate a library of HA reactive, paired antibody sequences. AF3 was used to computationally model antibody-HA interactions. Antibodies were expressed and characterized with BLI, ELISA, and microneutralization assays. Negative stain electron microscopy (nsEM) and single particle cryo-electron microscopy (SPA cryo-EM) were used to structurally characterize immune complexes.
B cell receptor sequencing of PBMCs from 23 healthy, human donors generated 1161 H5 HA-specific, paired antibody sequences. These sequences comprised new examples within a known class of antibodies (VH1-69, of which previous examples are in the AF3 training set) likely to target the conserved central stem. We modeled these antibody sequences in complex with H1N1 and H5N1 HAs using AF3. Highly confident predictions correlated with antibodies possessing high affinity and breadth. Moreover, in silico epitope footprints strongly matched experimental nsEM data. Finally, we resolved atomistic details with high-resolution SPA cryo-EM that explain challenges with AF3 immune complex prediction.
AF3 can be used to screen for VH1-69 bnAbs targeting a conserved epitope on influenza HA with high confidence. This method of down selection may be applied to other classes of antibodies and to other epitopes. However, models still need to be tuned to accurately predict immune complexes of novel or less characterized antibody classes.
Endowed Fellowship in the Skaggs Graduate School of Chemical and Biological Sciences, National Institute of Allergy and Infectious Diseases (NIAID).
Computational and Systems Immunology (COMP)
Morgan Gee, Pragati Sharma, Alesandra J. Rodriguez et al.· Journal of Immunology· 0 citations
The AIntibody challenge shows that AI can optimize antibodies in defined, biologically grounded regimes, in addition to highlighting critical gaps including affinity prediction and library-inspired antibody design and cross-task generalization.
M. Erasmus, Daniel Bedinger, Elizabeth Hopkins et al.· Nature Biotechnology· 0 citations