Computational Antibody Papers
Filter by tags
All
Filter by published year
All
TitleKey points
- Novel LLM (MINT) that natively encapsulates protein protein interactions.
- MINT (Multimeric INteraction Transformer) extends the ESM-2 protein language model by incorporating a cross-chain attention mechanism. This allows it to process multiple protein sequences simultaneously while preserving inter-sequence relationships and contextual information critical for modeling protein-protein interactions.
- MINT was trained on a large, curated subset of the STRING database, consisting of 96 million high-quality physical protein-protein interactions and 16.4 million unique protein sequences. The training employed a masked language modeling objective adapted for multimeric inputs.
- MINT was benchmarked on several general protein interaction tasks including binary interaction classification, binding affinity prediction (PDB-Bind), and mutational impact prediction (e.g., SKEMPI and MutationalPPI). It consistently outperformed existing PLMs, achieving state-of-the-art performance on multiple datasets such as a 29% improvement over baselines in SKEMPI.
- MINT outperformed antibody-specific models (e.g., IgBert, IgT5, and AbMap) on the FLAB benchmark and SARS-CoV-2 antibody mutant binding prediction tasks. It showed >10% performance improvement on three FLAB datasets and a 14% gain in low-data settings (0.5% training data) for SARS-CoV-2 binding predictions.
2025-06-05
AbBFN2: A flexible antibody foundation model based on Bayesian Flow Networks
- developability
- generative methods
- protein design
- Novel generative modeling framework (AbBFN2) using Bayesian Flow Networks (BFNs) for antibody sequence optimization.
- Trains on sequences from Observed Antibody Space (OAS) combined with genetic and biophysical annotations, leveraging a denoising approach for both conditional and unconditional sequence generation. Targets include optimizing Therapeutic Antibody Profiler (TAP) annotations.
- Computationally validated for germline assignment accuracy, species prediction (humanness), and TAP parameter optimization.
- Combines multiple antibody design objectives into a unified, single-step optimization process, unlike existing software methods which are typically specialized for individual tasks.
2025-06-05
Adapting ProteinMPNN for antibody design without retraining
- protein design
- generative methods
- Novel method to bias ProteinMPNN for antibody design, without modifying model weights.
- Logits from protein-general ProteinMPNN and antibody-specific AbLANG are added and softmaxed. Addition of AbLANG is supposed to push the model into the antibody-acceptable space.
- On in-silico experiments ProteinMPNN+AbLang outperformed ProteinMPNN alone and rivalled antibody-specific AbMPNN.
- Authors designed 96 variants of Trastuzumab CDR-H3 using ProteinMPNN, AbLang and ProteinMPNN+AbLang each. AbLANG and ProteinMPNN produced 1 and 3 successful variants respecitively (both out of 96) whereas their combination produced 36 successful variants.
- None of the variants were better variants than WT Trastuzumab.
2025-06-05
PSBench: a large-scale benchmark for estimating the accuracy of protein complex structural models:
- structure prediction
- PSBench is a large benchmark dataset (>1M models) for training and evaluating model accuracy estimation (EMA) methods for protein complex structures, using data from CASP15 & CASP16.
- Models were generated by AlphaFold2-Multimer and AlphaFold3 under blind prediction conditions and annotated with 10 detailed global, local, and interface quality scores.
- The dataset enables development of advanced EMA methods (e.g. GATE), which showed top performance in blind CASP16 assessments.
- Fast and reliable numbering tool with an inbuilt free germline database, unifying functionalities of tools such as IgBlast, ANARCI etc.
- It can number both amino acid and nucleotide sequences.
- Rather than using statistical methods such as HMMs, MMSeqs-like methodology was used for rapid alignment.
- Alignments are more accurate than existing methods, with speed improvement, running on a CPU.
- Fast, alignment-based antibody numbering tool, significantly outperforming existing software in processing speed.
- Uses a simplified global alignment with a custom scoring matrix, facilitating rapid numbering of millions of sequences efficiently.
- Ensures accuracy comparable to established methods (ANARCI, AbNum) while numbering large-scale antibody datasets.
- Emphasizes interpretability and robustness, providing transparent sequence scoring useful for humanization tasks.
- New version of ANARCI - using language models.
- Employs a Seq2Seq language model eliminating the need for alignment-based numbering, thus generalizing well to novel sequences.
- Provides numbering that matches existing methods for >99.99% conserved residues and >99.94% CDR regions.
- Improved speed of the original HMM-based ANARCI when GPU is available.
- Can be fine-tuned for rare immunoglobulin domains (e.g., shark VNAR sequences, T-cell receptors), offering customizable antibody numbering workflows.
- Offline toolkit for antibody numbering and CDR delineation (ABRs).
- Provides an offline toolkit integrating five established antibody numbering schemes (Kabat, Chothia, IMGT, Aho, Martin).
- Uses IMGT as the source of Germlines.
- Allows prediction of Complementarity-Determining Regions (CDRs) and Antigen-Binding Regions (ABRs) through Hidden Markov Models (HMMs).
- Addresses data security concerns by enabling offline usage, beneficial for therapeutic antibody development.
- Achieves high recall (0.92) in identifying ABRs, making it superior to existing tools which rely heavily on online services.
2025-04-28
Scaling unlocks broader generation and deeper functional understanding of proteins
- protein design
- non-antibody stuff
- Novel protein generative language model — ProGen3
- The model can do autoregressive generation N-to-C, C-to-N, and also supports span infilling.
- The architecture is a Transformer with a Sparse Mixture of Experts (MoE), activating about 27% of parameters per forward pass to improve computational efficiency.
- They studied how sampling affects training by trying different family-level weighting schemes. Uniform sampling across families (where small and large families have equal chance) gave better diversity and generalization, while unmodified sampling (letting big families dominate) performed worst.
- They validated the models by showing that generated proteins express well in wet lab experiments (split-GFP assays, spanning both highly novel and moderately novel sequence spaces).
- They used a large thermostability dataset to align model predictions to stability. This alignment is not standard fine-tuning — instead, preference optimization was applied, teaching the model to prefer sequences predicted to have higher stability. Upon experimental validation, aligned models indeed produced proteins with higher expression and stability.
2025-04-28
Atom level enzyme active site scaffolding using RFdiffusion2
- protein design
- non-antibody stuff
- Improvement upon earlier RFDiffusion, enhancing stability and accuracy in designing enzyme active sites.
- Catalytic sites can now be specified at the atomic level instead of the residue backbone level used previously. This eliminates the need to explicitly enumerate side-chain rotamers.
- Training uses flow matching, a technique that simplifies and stabilizes the diffusion training process.
- Benchmarked on a set of 41 diverse enzyme active sites; RFdiffusion2 succeeded in all 41 cases, significantly outperforming the earlier RFDiffusion, which succeeded in only 16.