Transformer (AR)
40 methods · 2022–2026
Transformer (AR): Autoregressive transformer decoders, which emit the peptide one residue at a time conditioned on the spectrum and on the residues already chosen. The dominant architecture since Casanovo, and the one most benchmarked tools share.
Autoregressive transformer decoders, which emit the peptide one residue at a time conditioned on the spectrum and on the residues already chosen. The dominant architecture since Casanovo, and the one most benchmarked tools share.
The earliest of its 40 methods is Casanovo (2022); 39 more have followed.
| Methods | 40 |
| Papers describing them | 64 |
| Authors | 245 |
| Active | 2022-02-07 to 2026-10-05 |
| Deep learning | 40 of 40 |
| Kinds | algorithm (36), adjacent (2), downstream-application, post-processor |
| Acquisition | DDA (35), DIA (3), both (2) |
Methods (40)
Oldest first, by the paper that describes each one.
- Casanovo (2022): First Transformer
- DPST (2022): Amino-acid-aware transformer
- PaSER Novor (2023): Real-time 4D-proteomics
- BiATNovo (2023): Bidirectional self-attention
- DpNovo (2023): Transformer + dynamic programming
- π-HelixNovo (2023): Complementary spectra
- InstaNovo (2023): Knapsack beam search
- SeqNovo (2023): Seq2Seq for IoMT
- DiaTrans (2023): Transformer for DIA
- Casanovo-DB (2024): Database search scoring
- NovoB (2024): Bidirectional decoding
- AdaNovo (2024): Mutual info for PTMs
- ContraNovo (2024): Contrastive learning
- Cascadia (2024): Transformer for DIA
- PowerNovo (2024): Transformer + BERT ensemble
- TransNovo (2024): Transformer-based sequencing
- π-xNovo (2024): Explainable AI sequencing
- SearchNovo (2024): DB-search + de novo fusion
- ByNovo (2024): RankNovo’s in-house base model with an auxiliary b/y ion head
- RankNovo (2024): Universal reranking
- DIANovo (2024): Transformer-based de novo sequencer for DIA: disentangles coeluted/multiplexed precursor spectra. Presented in Zheng Ma’s arXiv preprint (2024) and his PhD thesis (2025).
- ReNovo (2024): Retrieval-based sequencing
- XA-Novo (2025): NAR knowledge distillation
- TSARseqNovo (2025): Semi-autoregressive
- PepGo (2025): Tree search-based decoding
- DiNovo (2025): Mirror proteases + DL
- Pairwise (2025): Mass difference attention
- InstaNovo-P (2025): Phosphoproteomics
- Casanovo Foundation (2025): Foundation model for tandem-MS proteomics: pre-trained the Casanovo spectrum encoder on 30M labelled spectra from MassIVE-KB, then reused the encoder off-the-shelf for downstream tasks (de novo sequencing, spectrum quality, chimericity, phosphorylation, glycosylation prediction). Successor in spirit to Casanovo v1/v2/v5. The peer-reviewed version (MCP 2026) names the framework CasanovoTL.
- InstaNovo glycopeptide fine-tuning (2025): Adaptation of InstaNovo, a transformer de novo peptide sequencer, to glycoproteomics by fine-tuning on glycopeptide spectra. Learning from glyco spectra is measurable but limited, and every fine-tuning setting suffers catastrophic forgetting, losing accuracy on peptides the base model handled well; fine-tuning on unfiltered spectra that partly overlap the original training set gives the strongest learning signal but does not remove the effect. PCA of spectrum embeddings shows a large domain shift that explains the difficulty, and the study argues for treating overlapping glyco spectra as informative examples rather than outliers to discard, alongside multi-task and contrastive objectives and embedding transformations that pull glyco spectra toward the original distribution.
- RNovA (2025): Zero-shot open PTM discovery
- pUniFind (2025): Multimodal pre-trained transformer for mass spectra that unifies peptide-spectrum scoring and zero-shot de novo sequencing in a single model. Trained on >100M open-search-derived spectra; reports +60% PSMs over prior de novo methods with 1,300+ modifications supported, and a DL-based QC step that recovers 38.5% additional peptides.
- Modanovo (2025): Unified PTM-aware model
- PLMNovo (2025): Casts peptide-spectrum matching as a constrained optimisation that aligns the spectrum encoder’s embeddings with those of a pre-trained protein language model, and trains encoder and decoder with a Lagrangian primal-dual algorithm, optionally fine-tuning the language model.
- DyCoNovo (2025): A deep learning de novo sequencing model that adds lightweight dynamic convolution for local peak clusters and phased contrastive learning to a Transformer.
- CausalNovo (2026): Causality-informed framework
- MemNovo (2026): Autoregressive transformer de novo sequencer with a memory mechanism that lets the decoder revisit the input spectrum at each step to balance error accumulation.
- π-HelixNovo2 (2026): Successor to π-HelixNovo with an emphasis on availability: an online inference service alongside the model architecture refinement. Same Tsinghua / Pengcheng Lab / NCPSB collaboration as the original.
- GyroNovo (2026): Attacks missing b- and y-ion fragments on two fronts. Rather than treating imputation as a fixed reconstruction task, it uses the decoder errors seen during training to steer the imputation objective toward the fragments that actually cause mistakes, and to build easy and hard augmented views of each spectrum so the decoder learns under varying spectral corruption. It also gives self-attention a mass-aware inductive bias, using rotary embeddings to encode pairwise mass differences between peaks. Inference needs no extra inputs or search. Reports about 9 points of peptide-level and 7 points of amino-acid-level precision over the previous best on NovoBench.
- dIon (2026): Self-supervised DINO-style pretraining of a spectrum encoder, from a precursor-conditioned spectrum mixture and a partial spectrum with the precursor withheld; initialising a Casanovo-style de novo sequencer from it raises peptide precision over training from scratch.
How they score
7 of the 40 have been run on denovo_benchmarks, which ranks 14 tools over 86 datasets. The family’s best median rank is 1.
- InstaNovo: median peptide-level average precision 0.853, median rank 1 of 14
- π-HelixNovo: median peptide-level average precision 0.825, median rank 4 of 14
- ContraNovo: median peptide-level average precision 0.808, median rank 4 of 14
- Pairwise: median peptide-level average precision 0.808, median rank 4 of 14
- Casanovo: median peptide-level average precision 0.777, median rank 7 of 14
- AdaNovo: median peptide-level average precision 0.535, median rank 10 of 14
- BiATNovo: median peptide-level average precision 0.469, median rank 11 of 14
Read these next to the rest of the field, not on their own: what the numbers mean.
Applied in
Papers describing them (64)
- De novo mass spectrometry peptide sequencing with a transformer model (2022, bioRxiv, preprint)
- DPST: De Novo Peptide Sequencing with Amino-Acid-Aware Transformers (2022, arXiv, preprint)
- De novo mass spectrometry peptide sequencing with a transformer model (2022, ICML 2022, ML conference)
- Sequence-to-sequence translation from mass spectra to peptides with a transformer model (2023, bioRxiv, preprint)
- PaSER Novor: Real-time de novo sequencing for 4D-Proteomics applications (2023, preprint)
- BiATNovo: A Self-Attention based Bidirectional Peptide Sequencing Method (2023, bioRxiv, preprint)
- DpNovo: A DEEP LEARNING MODEL COMBINED WITH DYNAMIC PROGRAMMING FOR DE NOVO PEPTIDE SEQUENCING (2023, thesis)
- Introducing PandaNovo for practical large-scale de novo peptide sequencing (2023, bioRxiv, preprint)
- De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experiments (2023, bioRxiv, preprint)
- SeqNovo: De Novo Peptide Sequencing Prediction in IoMT via Seq2Seq (2023, IEEE Journal of Biomedical and Health Informatics, peer-reviewed)
- Transformer-Based De Novo Peptide Sequencing for Data-Independent Acquisition Mass Spectrometry (2023, 2023 IEEE 23rd International Conference on Bioinformatics and Bioengineering (BIBE), peer-reviewed)
- A learned score function improves the power of mass spectrometry database search (2024, bioRxiv, preprint)
- Introducing π-HelixNovo for practical large-scale de novo peptide sequencing (2024, Briefings in Bioinformatics, peer-reviewed)
- Transformer-Based De Novo Peptide Sequencing for Data-Independent Acquisition Mass Spectrometry (2024, arXiv, postprint)
- Bidirectional de novo peptide sequencing using a transformer model (2024, PLOS Computational Biology, peer-reviewed)
- AdaNovo: Adaptive De Novo Peptide Sequencing with Conditional Mutual Information (2024, arXiv, preprint)
- ContraNovo: A Contrastive Learning Approach to Enhance De Novo Peptide Sequencing (2024, AAAI 2024, peer-reviewed)
- A transformer model for de novo sequencing of data independent acquisition mass spectrometry data (2024, bioRxiv, preprint)
- PowerNovo: de novo peptide sequencing via tandem mass spectrometry using an ensemble of transformer and BERT models (2024, Scientific Reports, peer-reviewed)
- TransNovo (2024, preprint)
- Sequence-to-sequence translation from mass spectra to peptides with a transformer model (2024, Nature Communications, peer-reviewed)
- Transforming de novo peptide sequencing by explainable AI (2024, Research Square, preprint)
- Accounting for Digestion Enzyme Bias in Casanovo (2024, Journal of Proteome Research, peer-reviewed)
- Bridging the Gap between Database Search and De Novo Peptide Sequencing with SearchNovo (2024, bioRxiv, preprint)
- BERT model for de novo protein sequencing using tandem mass spectra (2024, Proceedings of the International Conference “Mathematical Biology and Bioinformatics”, abstract)
- BiATNovo: An Attention-based Bidirectional De Novo Sequencing Framework for Data-Independent-Acquisition Mass Spectrometry (2024, bioRxiv, preprint)
- RankNovo: A Universal Reranking Approach for Robust De Novo Peptide Sequencing (2024, OpenReview, preprint)
- Disentangling the Complex Multiplexed DIA Spectra in De Novo Peptide Sequencing (2024, arXiv, preprint)
- ReNovo: Retrieval-Based De Novo Mass Spectrometry Peptide Sequencing (2024, ICLR 2025, ML conference)
- AdaNovo: Towards Robust De Novo Peptide Sequencing in Proteomics against Data Biases (2024, NeurIPS 2024, ML conference)
- Bridging the Gap between Database Search and De Novo Peptide Sequencing with SearchNovo (2025, ICLR 2025, ML conference)
- XA-Novo: an accurate and high-throughput mass spectrometry-based de novo sequencing technology for monoclonal antibodies and antibody mixtures (2025, Research Square, preprint)
- A transformer-based semi-autoregressive framework for high-speed and accurate de novo peptide sequencing (2025, Communications Biology, peer-reviewed)
- PepGo: a deep learning and tree search-based model for de novo peptide sequencing (2025, bioRxiv, preprint)
- DiNovo: high-coverage, high-confidence de novo peptide sequencing using mirror proteases and deep learning (2025, bioRxiv, preprint)
- Pairwise Attention: Leveraging Mass Differences to Enhance De Novo Sequencing of Mass Spectra (2025, bioRxiv, preprint)
- InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale proteomics experiments (2025, Nature Machine Intelligence, peer-reviewed)
- InstaNovo-P: A de novo peptide sequencing model for phosphoproteomics (2025, bioRxiv, preprint)
- Foundation model for mass spectrometry proteomics (2025, arXiv, preprint)
- Universal Biological Sequence Reranking for Improved De Novo Peptide Sequencing (2025, arXiv, preprint)
- Pairwise Attention: Leveraging Mass Differences to Enhance De Novo Sequencing of Mass Spectra (2025, Journal of Proteome Research, peer-reviewed)
- Advancing De Novo Glycopeptide Sequencing with InstaNovo in Glycoproteomics (2025, thesis)
- Zero-Shot De Novo Peptide Sequencing with Open Post-Translational Modification Discovery (2025, Research Square, preprint)
- pUniFind: a unified large pre-trained deep learning model pushing the limit of mass spectra interpretation (2025, arXiv, preprint)
- A transformer model for de novo sequencing of data independent acquisition mass spectrometry data (2025, Nature Methods, peer-reviewed)
- Universal Biological Sequence Reranking for Improved De Novo Peptide Sequencing (2025, ICML 2025, ML conference)
- Improvements to CasaNovo, a deep learning de novo peptide sequencer (2025, bioRxiv, preprint)
- Advancing Proteomic Analyses with Graph-Based Deep Learning: Protein Inference and DIA De Novo Peptide Sequencing (2025, thesis)
- Modanovo: A Unified Model for Post-Translational Modification-Aware de Novo Sequencing Using Experimental Spectra from In Vivo and Synthetic Peptides (2025, bioRxiv, preprint)
- Protein Language Model-Aligned Spectra Embeddings for De Novo Peptide Sequencing (2025, bioRxiv, preprint)
- DyCoNovo: a De Novo Peptide Prediction Model Based on Dynamic Convolution and Phased Contrastive Learning (2025, 2025 18th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), peer-reviewed)
- Modanovo: A Unified Model for Post-translational Modification-Aware De Novo Sequencing Using Experimental Spectra From In Vivo and Synthetic Peptides (2025, Molecular & Cellular Proteomics, peer-reviewed)
- Improvements to Casanovo, a Deep Learning De Novo Peptide Sequencer (2025, Journal of Proteome Research, peer-reviewed)
- CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework (2026, OpenReview, preprint)
- XA-Novo: an accurate and high-throughput mass spectrometry-based de novo sequencing technology for monoclonal antibodies and antibody mixtures (2026, Nature Communications, peer-reviewed)
- DiNovo enables high-coverage and high-confidence de novo peptide sequencing via mirror proteases and deep learning (2026, Nature Communications, peer-reviewed)
- Zero-shot de novo peptide sequencing with open posttranslational modification discovery (2026, Nature Biotechnology, peer-reviewed)
- A large-scale unified deep learning model for peptide mass spectrum interpretation trained on multimodal data (2026, Nature Machine Intelligence, peer-reviewed)
- MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry (2026, arXiv, preprint)
- π-HelixNovo2: Making Accurate Online De Novo Peptide Sequencing Available to All (2026, Genomics, Proteomics & Bioinformatics, peer-reviewed)
- InstaNovo-P: a de novo peptide sequencing model for phosphoproteomics (2026, Nature Communications, peer-reviewed)
- GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing (2026, arXiv, preprint)
- Pretrained spectrum representations for diverse tandem mass spectrometry proteomics tasks (2026, Molecular & Cellular Proteomics, peer-reviewed)
- dIon: Fragmentation-Based Invariance for Self-Supervised Learning of Tandem Mass Spectra (2026, arXiv, preprint)