transformer / foundation model

RNAtomy is a structure-aware RNA foundation model that learns structural and functional representations from geometric and conformational features, achieving strong performance on diverse tasks.

This paper designs de novo ligases that achieve site-specific ubiquitylation of a target protein with high selectivity and efficiency.

The paper introduces a protein language model pretrained on continuous biophysical fields, showing improved performance on thermostability prediction compared to sequence-only models.

This paper evaluates generative models for predicting novel enzymatic reactions, finding they can generate correct reactions for 25% of test cases and improve over existing methods.

RPFace predicts RNA-protein interfaces directly from sequence, outperforming existing methods on RNA interfaces and generalizing to in vivo data without pretraining.

This paper introduces PDLMs, protein-DNA language models that jointly represent protein and DNA sequences to predict and engineer DNA recognition specificity.

FoldAbS repurposes protein folding models as foundation encoders to improve antibody screening by evaluating specificity, affinity, and developability in a unified framework.

This study evaluates AlphaFold 3's accuracy and reliability in predicting RNA structures and protein-RNA interfaces, finding that confidence scores do not always correlate with correct predictions.

The paper introduces AtomLM-9b, a language model using an all-atom structure tokenizer to predict protein structures with performance comparable to AlphaFold 3.

LIMBER is a hybrid RNA foundation model that predicts RNA structure and function, showing strong performance in secondary-structure prediction and classification, along with transferability to related tasks.

IDPForge generates all-atom ensembles of intrinsically disordered proteins and regions using a transformer diffusion model, showing good agreement with experimental data.

This paper uses AlphaFold 3 to predict T cell receptor antigen specificity from sequence, achieving high accuracy on unseen epitopes.

ESMfluc predicts local backbone variability in proteins using attention-based evidence decomposition, showing how sequence positions contribute to predictions of single or multi-PB states.

HighFold-Lasso predicts lasso peptide structures using topology-guided methods, achieving perfect topology success and low RMSD on a test set of 52 structures.

This paper introduces NanoPCT, a framework that predicts protein corona formation by learning universal patterns across proteins and enables zero-shot prediction for unseen proteins.

The paper uses AlphaFold2 to model class B β-lactamases without experimental structures, tests the BBL numbering scheme's validity, and identifies structural exceptions.

This paper develops an unsupervised pipeline to predict viral fitness using sequence data and finds that protein alignment, not model size, is key for accurate prediction.

The paper determines the structure of norovirus NS3 and proposes it functions as an RNA translocase.

ABCP_finder predicts anti-breast cancer peptides using transformer embeddings, achieving high accuracy and supporting biological relevance through external validation.

This paper evaluates a self-architecting transformer for protein representation, showing it achieves better performance than existing models with a smaller dataset.

This paper introduces InstaNovo-FM, a self-supervised foundation model for proteomics that learns from tandem mass spectra to enable transferable applications across datasets and instruments.

This study uses ESM3 to generate knotted proteins, finding high success rates with guided decoding and robustness in real knotted proteins.

The paper introduces AtlasFold, a protein structure prediction method using metagenomic-scale language models, achieving state-of-the-art accuracy and efficient inference.

Web resource offering predicted monomeric and oligomeric structures for 1,603 M. leprae proteins, with pockets, epitope scores, essentiality and a target priority ranking, plus a nanopore resistance-calling module.

WaterFlow predicts ordered water molecule positions in protein structures with sub-angstrom accuracy, outperforming existing methods and improving drug design and crystallographic refinement.

TaHL-PTM improves PTM prediction by addressing tokenization issues in decoder-only PLMs, achieving better performance on residue-level tasks.

AntiSite predicts antibody and nanobody paratopes using shared models with or without structural data, achieving high performance across benchmarks and generalizing to nanobodies.

The paper introduces a dataset of VHH-Fc antibodies and shows that IgG-trained models outperform VHH-Fc-specific models in predicting developability properties.

A fine-tuned ProtT5 protein language model predicts per-residue energetic frustration from sequence (Pearson r 0.86 on unseen CATH topologies) and profiles the human proteome in ~17 minutes on one GPU.

ProtJEPA predicts joint embeddings across ten biological modalities from protein sequences, outperforming baselines in zero-shot GO retrieval and enzyme class prediction.