Computational Antibody Papers
Filter by tags
All
Filter by published year
All
TitleKey points
2026-06-24
BoltzProt-1: Towards Efficient De Novo Binder Design with Good Developability
- protein design
- binding prediction
- Novel de novo protein binder design pipeline from Boltz that pairs an optimized candidate generation engine with BoltzPPI, a novel, interaction-aware scoring model built to rank designs based on binding confidence rather than just geometric plausibility.
- To train BoltzPPI, the authors created a set of positive labels (PDB and patent complexes) and negative labels (synthetic non-interacting protein pairs co-localized via Boltz-2). This input pool (token, pair, distance, and mask features) is processed through a Pairformer stack using specific optimization tricks (co-trained focal loss, multi-view representation dropping, and Gaussian noise injection) to predict a binary binding-confidence score.
- Significantly improves experimental VHH design performance, boosting the confirmed-binder hit rate from 3.3% to 8.0% on novel targets and successfully discovering screening hits for 7 out of 10 benchmark targets from the Chai-2 dataset.
- Yields highly manufacturable binders, with 58% of its confirmed candidates passing a comprehensive panel of strict biophysical filters (such as thermal stability, purity, and low aggregation Propensity), outperforming both BoltzGen (40%) and clinical-stage VHH controls (21%).
- Generates structurally distinct binders whose CDR loops remain highly distinct from any known entries in the SAbDab database, with the entire pipeline made available to researchers via the Boltz API and Lab platforms.
- Case study of JAM-2, a generative biomolecular design model that engineered drug-like VHH antibodies against five distinct peptide-MHC class I (pMHC-I) targets across two HLA alleles using only target amino acid sequences as input.
- When reformatted into bispecific T-cell engagers (TCEs) without experimental optimization, the designs mediated potent, sub-nanomolar T-cell activation and successfully directed primary human T cells to kill target-presenting cells.
- Binders achieved stringent selectivity, showing a minimum 216-fold preference for NY-ESO-1 over highly similar human self-peptides and safely discriminating mutant KRAS variants (G12V and G12C) from wild-type sequences. A cryo-EM structure verified this atomic accuracy with a whole-complex Ca RMSD of 0.93 Å.
- Bypassing the need for downstream engineering, 78% of the designed antibodies passed five core industry-standard biophysical developability criteria, exhibiting strong expression titers, favorable monomericity, thermal stability, and low polyreactivity
- Theoretical framework for predicting molecular recognition from sequence alone structures are not required at inference time.
- The selection probability governing which molecules bind which is provably unique, it is the Boltzmann distribution, derived from first principles rather than assumed.
- To use it, we need sequence representations whose dot product approximates relative binding energies, not just higher for binders vs non-binders, but correctly ranking the full competitor pool.
- This separates two problems: learning the embedding geometry (cheap, binary binding pairs) from recovering absolute binding energies (cheap, just two calibration parameters fitted against a small number of experimental KdK_d Kd measurements).
2026-06-23
GermRL: Alleviating The Germline Bias In Autoregressive Antibody Language Models Through Reinforcement Learning
- language models
- New language model addressing the germline bias in NGS training data.
- GermRL fine-tunes an autoregressive ProGen2-OAS base model using a modified, outcome-supervised Group Relative Policy Optimization (GRPO) framework that features epoch-end weight syncing and automated "prefix grafting." So instead of using an absolute baseline, GRPO generates a local batch of sequences simultaneously; since the reward function favors mutations, any safely mutated sequence scores higher than its germline-heavy peers, resulting in a positive relative z-score that "up-votes" the mutated path while "down-voting" the underperforming germline ones.
- This is the first framework to fix germline bias in generative, autoregressive models (unlike previous efforts that targeted masked models), and it acts as a lightweight, modular plugin requiring no scratch training or data pre-processing.
- It was evaluated on one-shot generation success (pass@1) across specific mutation bounds (LD5 to LD35), measuring structural plausibility via ESMFold, sequence/V-gene diversity, developability metrics (GRAVY/instability), and semantic overlap with natural human antibodies using AntiBERTy UMAP embeddings.
2026-06-23
AbCVista: a deep learning framework for predicting antibody conformational ensembles
- structure prediction
- Deep learning framework designed to predict both highly accurate single antibody structures and biologically realistic 3D conformational ensembles.
- It combines a fast, antibody-specific sequence alignment tool (Abalign) with an optimized OpenFold engine, utilizing a novel "split-and-intersect" cross-chain clustering strategy to properly model heavy-light chain coordination without chain-segregation bias.
- In static benchmarks, it outperforms top-tier models like AlphaFold3 and IgFold, especially on the notoriously difficult hypervariable CDR-H3 loop, while cutting down structure inference time to roughly 2 minutes per antibody.
- Validated against extensive Molecular Dynamics (MD) simulations, its generated ensembles accurately capture true biological flexibility rather than random sampling noise, which significantly improves downstream therapeutic evaluations like antibody aggregation and developability mapping.
2026-05-29
ABRefine: An Accurate Antibody Structure Refinement Method by Equivariant Graph Transformer with Rigid Body Constraint
- structure prediction
- An end-to-end structural refinement method specifically developed for antibodies and nanobodies.
- Curates structures from SAbDab to train a pure equivariant graph transformer (not a standard EGNN) that directly predicts 3D coordinate shift vectors for all backbone atoms (N, Ca, C, O).
- Benchmarked against five baseline methods, achieving an average Ca RMSD improvement typically on the order of 0.01 Å to 0.05 Å - which essentially is raising questions of statistical significance (as the error bars are order of 1.0A at least).
2026-05-29
Protein language models trained on biophysical dynamics inform mutation effects
- language models
- Protein Language model that understands protein dynamics.
- The authors leveraged existing structural data and datasets like mdCATH to gather equilibrium fluctuations for 64,403 proteins. Instead of raw time-series trajectories, they extracted calculated biophysical properties like root-mean-square fluctuations (RMSF) and Normal Mode Analysis (NMA) to serve as training labels.
- They trained two models, SeqDance 'from scratch' and ESMDance as an extension of ESM2. ESMDance: Built by fine-tuning the pre-trained ESM-2 transformer, teaching it to map its existing evolutionary knowledge to these new physical flexibility profiles. SeqDance: Trained completely from scratch using only raw sequences and the target dynamics data, forcing it to learn pure, unbiased residue co-movement and physics.
- To test zero-shot mutation prediction, the models compare the wild-type flexibility against the mutated sequence's flexibility. A large mathematical discrepancy flags a highly disruptive, damaging mutation. These predicted shifts were correlated against deep mutational scanning (DMS) lab data measuring actual cellular fitness and stability changes. ESMDance is the go-to for mutation prediction (especially on viral and de novo designed proteins with no evolutionary history), while SeqDance wins at modeling highly flexible Intrinsically Disordered Regions (IDRs).
2026-05-29
Language Modeling Materializes a World Model of Protein Biology
- language models
- protein design
- structure prediction
- New versions of ESM and ESMFold, ESMC (ESM Cambrian) and ESMFold2 to model protein sequence, structure, and function.
- Like previous versions, ESMC is trained entirely on sequences using a masked language modeling (MLM) objective. However, it scales up to 2.8 billion metagenomic sequences, nearly a 100-fold increase over the 50 million sequences used for ESM2.
- ESMFold2 achieves state-of-the-art atomic resolution and directly outperforms AlphaFold3 on complex antibody-antigen predictions (even when AlphaFold3 is given multiple sequence alignments and ESMFold2 operates from sequence alone).
- To design therapeutic antibody fragments (scFvs), they input the target sequence and lock in a stable, known antibody framework template, leaving only the target-recognizing loops (CDRs) to be filled in.
- AI-guided optimization: The system uses mathematical backpropagation through both ESMC and ESMFold2 to iteratively optimize the CDR sequences. It automatically mutates the loops to maximize structural interface confidence scores.
- They synthesized and tested these computational designs in the wet lab, achieving high experimental hit rates and discovering entirely novel binders with therapeutically relevant nanomolar affinities.
2026-05-22
GYDE: A collaborative drug discovery platform for AI-powered protein design and engineering
- server/webapp
- No-code environment from Genentech/Roche, giving access to latest tools in protein/antibody design.
- Integrates models (BindCraft, OpenFold etc.) and datasets such as PDB, Uniprot.
- A Unified Framework for Unsupervised AIRR Analytics
- immuneML introduces the first standardized environment to discover patterns, cluster sequences, and run robust stability validations on partially or imperfectly labeled adaptive immune receptor data.
- The platform systematically evaluates and compares generative machine learning models (such as LSTM and VAE) to determine how effectively they can engineer novel, antigen-specific immune sequences versus simply memorizing training data.
- It rigorously assesses how well different data representations, including advanced protein language models, capture true biological properties like epitope specificity and MHC restrictions, a utility proven on 48,000 experimental TCRβ sequences.
- It provides vital exploratory and dimensionality reduction tools to identify sequencing batch effects and data biases before running supervised diagnostics, demonstrated using a real-world single-cell dataset from 143 inflammatory bowel disease patients.