InstaNovo+

algorithm · Diffusion

InstaNovo+: algorithm · Diffusion. Multinomial diffusion

Multinomial diffusion

Kind algorithm
Deep learning yes
Acquisition DDA
Family Diffusion

Code

Live stars, open issues and last-push figures are on the Code activity chart.

Data

What the papers describing this method trained it on and reported its results on, as each paper states it. A version is named only where the paper names one.

Trained on

Evaluated on

Deposited

Own results (2)

Tables from this method’s own papers that report it alone, on each of its evaluation sets, with no other method in the table. Standardised like the comparisons: every value on a 0-1 scale, the measure and then the evaluation set across.

Supplementary Table 3

De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experiments, page 59: InstaNovo+ evaluation results on all datasets. Confidence intervals are calculated as ±1.96 × se_B where se_B is a bootstrap standard error estimated from 10,000 replicates.*We do not calculate bootstrap standard errors for the ProteomeTools datasets because their size makes it prohibitively costly but also implies the standard errors would be very small.

a dataset per evaluation set Amino acid precision
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae HC-PT AC-PT Mean
InstaNovo+ 0.617 0.839 0.821 0.393 0.608 0.412 0.811 0.710 0.732 0.696 0.809 0.677
a dataset per evaluation set Amino acid accuracy
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae HC-PT AC-PT Mean
InstaNovo+ 0.616 0.839 0.820 0.395 0.606 0.406 0.810 0.709 0.699 0.694 0.809 0.673
a dataset per evaluation set Peptide accuracy
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae HC-PT AC-PT Mean
InstaNovo+ 0.517 0.697 0.736 0.198 0.464 0.229 0.719 0.562 0.579 0.589 0.710 0.545
a dataset per evaluation set Peptide AUC
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae HC-PT AC-PT Mean
InstaNovo+ 0.477 0.644 0.697 0.137 0.417 0.166 0.689 0.526 0.556 0.542 0.680 0.503

A table of the paper's OWN results: it names one method and compares nothing, so it is recorded as own results rather than as a comparison. 'Exc. Yeast' is measured ON yeast: it names the leave-one-out setup in which the model is trained on the other eight species. The paper: 'the models were trained on nine-species excluding yeast, and then evaluated on yeast' (p. 8), and 'For Yeast, we use the splits as defined in DeepNovo and PointNovo'. Fig. 2d's caption reads 'accuracy … on the high-resolution nine-species dataset excluding yeast' before saying 'evaluated on yeast', which invites the opposite reading. THE TEXT AND THIS TABLE DISAGREE: p. 10 says that after fine-tuning on nine-species exc. yeast 'IN+ reaching a peptide-level accuracy of 54.6%', while this table's Exc. Yeast row prints 0.579 ± 0.006 peptide accuracy (and 0.556 AUC). The two do not match, and the paper does not say which run each comes from. The table's value is recorded, as printed.

Extended Data Table 3

InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale proteomics experiments, page 25: InstaNovo+ evaluation results on all datasets

a dataset per evaluation set Amino acid precision
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae Bacillus subtilis Mus musculus HC-PT
InstaNovo+ 0.617 0.839 0.821 0.393 0.608 0.412 0.811 0.710 0.755 0.796 0.726 0.696
a dataset per evaluation set Amino acid precision
AC-PT Mean
InstaNovo+ 0.809 0.692
a dataset per evaluation set Amino acid accuracy
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae Bacillus subtilis Mus musculus HC-PT
InstaNovo+ 0.616 0.839 0.820 0.395 0.606 0.406 0.810 0.709 0.667 0.753 0.722 0.694
a dataset per evaluation set Amino acid accuracy
AC-PT Mean
InstaNovo+ 0.809 0.680
a dataset per evaluation set Peptide accuracy
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae Bacillus subtilis Mus musculus HC-PT
InstaNovo+ 0.517 0.697 0.736 0.198 0.464 0.229 0.719 0.562 0.624 0.674 0.490 0.589
a dataset per evaluation set Peptide accuracy
AC-PT Mean
InstaNovo+ 0.710 0.555
a dataset per evaluation set Peptide AUC
HeLa single-shot Immunopeptidomics Candidatus Scalindua brodae Snake venoms Nanobodies Wound exudates HeLa degradome Herceptin Saccharomyces cerevisiae Bacillus subtilis Mus musculus HC-PT
InstaNovo+ 0.477 0.644 0.697 0.137 0.417 0.166 0.689 0.526 0.598 0.650 0.431 0.542
a dataset per evaluation set Peptide AUC
AC-PT Mean
InstaNovo+ 0.680 0.512

Read from an IMAGE: the page has no text layer for this table, and GLM-OCR and PaddleOCR-VL, read independently, agree on every cell.

Compared against in other papers (2)

6 tables from 2 papers that report this method as a baseline, newest first. The caveats of the comparisons above hold twice over: each paper ran, retrained or quoted this method its own way, so its number here is that paper’s claim, not a measurement of the method. Where a paper says which it did, the word follows the method’s name.

MassNet: billion-scale AI-ready mass spectral corpus empowering scalable deep learning in proteomics (2026) · 4 tables

Fig. 4b

MassNet: billion-scale AI-ready mass spectral corpus empowering scalable deep learning in proteomics, page 7: Comparison of XuanjiNovo and existing de novo peptide sequencing models on the 15-species benchmark dataset. b, Cross-species peptide recall aggregated by species, illustrating overall performance differences across taxa.

15-species benchmark
Nature Methods (307 files) 		Peptide recall
Acinetobacter calcoaceticus Aedes aegypti Arabidopsis thaliana Aspergillus fumigatus Bacteroides thetaiotaomicron Drosophila melanogaster Escherichia coli Homo sapiens Mus musculus Pyrococcus furiosus Saccharomyces cerevisiae Toxoplasma gondii
π-HelixNovo · released 0.41 0.403 0.183 0.286 0.413 0.188 0.114 0.184 0.307 0.398 0.386 0.186
Casanovo V2 · released 0.6 0.57 0.266 0.395 0.652 0.319 0.329 0.224 0.458 0.516 0.598 0.262
InstaNovo+ · released 0.632 0.606 0.262 0.412 0.661 0.346 0.345 0.259 0.524 0.545 0.6 0.241
ContraNovo 0.654 0.615 0.283 0.436 0.678 0.373 0.368 0.299 0.54 0.544 0.626 0.277
XuanjiNovo 0.779 0.757 0.333 0.582 0.793 0.444 0.476 0.449 0.676 0.668 0.691 0.355
π-PrimeNovo 0.733 0.686 0.323 0.51 0.78 0.429 0.466 0.324 0.604 0.616 0.692 0.308
15-species benchmark
Nature Methods (307 files) 		Peptide recall
Trypanosoma brucei Zea mays Solanum lycopersicum
π-HelixNovo · released 0.348 0.443 0.001
Casanovo V2 · released 0.588 0.557 0.011
InstaNovo+ · released 0.623 0.649 0.004
ContraNovo 0.641 0.653 0.062
XuanjiNovo 0.762 0.799 0.252
π-PrimeNovo 0.738 0.731 0.06

Read from the paper's source data: every value is a cell of 'Source Data.xlsx', sheet Figure4B, behind a figure that prints no values. Nothing was estimated from bar heights. Values from the paper's Source Data file (sheet Figure4B); the figure prints none. XuanjiNovo is XuanjiNovo-100M, per the caption. THE TEXT AND THE SHEET DISAGREE once: p. 5 says the other models on S. lycopersicum ranged 'from as low as 0.01 (π-HelixNovo)', the sheet gives 0.001.

Supplementary Fig. 3c

MassNet: billion-scale AI-ready mass spectral corpus empowering scalable deep learning in proteomics, page 5: Evaluation of de novo peptide sequencing performance on other benchmark datasets. (A) Peptide recall across nine species using seven models: InstaNovo+, π-HelixNovo, ContraNovo, CasaNovo V2, π-PrimeNovo, XuanjiNovo-100M, and XuanjiNovo-130M. (B) Peptide recall on the HCC dataset. (C) Peptide recall on the PT dataset. The π-HelixNovo model is publicly available and was trained on datasets derived from eight species, excluding human. XuanjiNovo-100M: base model; XuanjiNovo-130M: fine-tuned on 30M PSMs from MassIVE.

ProteomeTools Peptide recall
PT
Casanovo V2 0.459
InstaNovo+ 0.479
ContraNovo 0.574
π-PrimeNovo 0.586
XuanjiNovo 0.605

Read from the paper's source data: every value is a cell of 'Supplementary Data 1.xlsx', sheet SFigure3C, behind a figure that prints no values. Nothing was estimated from bar heights. Values from Supplementary Data 1 (sheet SFigure3C); the figure prints none. The figure draws six bars and the sheet has five: π-HelixNovo (about 0.26 by its bar) has no row, so it is not recorded. THE TEXT AND THE SHEET DISAGREE: p. 8 says XuanjiNovo-100M 'achieved a recall of 0.62' on PT, the sheet gives 0.605.

Supplementary Fig. 3b

MassNet: billion-scale AI-ready mass spectral corpus empowering scalable deep learning in proteomics, page 5: Evaluation of de novo peptide sequencing performance on other benchmark datasets. (A) Peptide recall across nine species using seven models: InstaNovo+, π-HelixNovo, ContraNovo, CasaNovo V2, π-PrimeNovo, XuanjiNovo-100M, and XuanjiNovo-130M. (B) Peptide recall on the HCC dataset. (C) Peptide recall on the PT dataset. The π-HelixNovo model is publicly available and was trained on datasets derived from eight species, excluding human. XuanjiNovo-100M: base model; XuanjiNovo-130M: fine-tuned on 30M PSMs from MassIVE.

Proteomics identifies new therapeutic targets of early-stage hepatocellular carcinoma
as deposited 		Peptide recall
HCC
Casanovo V2 0.168
InstaNovo+ 0.087
ContraNovo 0.373
π-PrimeNovo 0.382
XuanjiNovo 0.676

Read from the paper's source data: every value is a cell of 'Supplementary Data 1.xlsx', sheet SFigure3B, behind a figure that prints no values. Nothing was estimated from bar heights. Values from Supplementary Data 1 (sheet SFigure3B); the figure prints none. The figure draws six bars and the sheet has five: π-HelixNovo, whose bar is barely above zero, has no row, so it is not recorded rather than read off the bar. XuanjiNovo is XuanjiNovo-100M, per the caption of Fig. 4.

Supplementary Fig. 3a

MassNet: billion-scale AI-ready mass spectral corpus empowering scalable deep learning in proteomics, page 5: Evaluation of de novo peptide sequencing performance on other benchmark datasets. (A) Peptide recall across nine species using seven models: InstaNovo+, π-HelixNovo, ContraNovo, CasaNovo V2, π-PrimeNovo, XuanjiNovo-100M, and XuanjiNovo-130M. (B) Peptide recall on the HCC dataset. (C) Peptide recall on the PT dataset. The π-HelixNovo model is publicly available and was trained on datasets derived from eight species, excluding human. XuanjiNovo-100M: base model; XuanjiNovo-130M: fine-tuned on 30M PSMs from MassIVE.

Nine-species benchmark
original (DeepNovo, 2017) 		Peptide recall
Bacillus subtilis Candidatus Thiodiazotropha endoloripes Apis mellifera Homo sapiens Methanosarcina mazei Mus musculus Vigna mungo Solanum lycopersicum Saccharomyces cerevisiae
π-HelixNovo · released 0.596 0.388 0.473 0.392 0.56 0.483 0.623 0.56 0.568
InstaNovo+ 0.576 0.417 0.467 0.456 0.528 0.471 0.565 0.585 0.56
Casanovo V2 0.622 0.446 0.499 0.455 0.565 0.486 0.594 0.624 0.598
ContraNovo 0.688 0.487 0.576 0.622 0.63 0.566 0.677 0.672 0.674
π-PrimeNovo 0.721 0.531 0.603 0.574 0.65 0.567 0.702 0.697 0.697
XuanjiNovo 100M 0.43 0.393 0.435 0.589 0.468 0.502 0.499 0.536 0.407
XuanjiNovo 130M 0.689 0.503 0.574 0.6 0.628 0.682 0.595 0.681 0.67

Read from the paper's source data: every value is a cell of 'Supplementary Data 1.xlsx', sheet SFigure3A, behind a figure that prints no values. Nothing was estimated from bar heights. Values from the paper's Supplementary Data 1 (sheet SFigure3A); the figure prints none. The sheet names InstaNovo where the legend draws InstaNovo+. THE TEXT AND THE SHEET DISAGREE once: p. 8 puts XuanjiNovo-130M on M. musculus at 'approximately 0.67', the sheet at 0.682.

Prime-DiffNovo: Accurate Peptide De Novo Sequencing via Non-autoregressive Generation and Diffusion Refinement (2026) · 2 tables

Table 1 (Nine-species · revised (build not stated))

Prime-DiffNovo: Accurate Peptide De Novo Sequencing via Non-autoregressive Generation and Diffusion Refinement, page 8: Peptide accuracy comparison of different de novo peptide sequencing methods across nine species on two benchmark datasets.

Nine-species benchmark
revised (build not stated) 		Peptide accuracy
Apis mellifera Bacillus subtilis Candidatus Thiodiazotropha endoloripes Homo sapiens Methanosarcina mazei Mus musculus Saccharomyces cerevisiae Solanum lycopersicum Vigna mungo Weighted average
InstaNovo v1.1 0.625 0.704 0.512 0.670 0.666 0.504 0.758 0.721 0.679 0.698
InstaNovo+ v1.1 0.646 0.731 0.545 0.687 0.691 0.523 0.779 0.743 0.704 0.723
π-PrimeNovo 0.602 0.702 0.549 0.704 0.681 0.537 0.671 0.699 0.731 0.682
Prime-DiffNovo 0.658 0.751 0.581 0.733 0.717 0.566 0.817 0.771 0.748 0.749

Table 1 (Nine-species · original (DeepNovo, 2017))

Prime-DiffNovo: Accurate Peptide De Novo Sequencing via Non-autoregressive Generation and Diffusion Refinement, page 8: Peptide accuracy comparison of different de novo peptide sequencing methods across nine species on two benchmark datasets.

Nine-species benchmark
original (DeepNovo, 2017) 		Peptide accuracy
Apis mellifera Bacillus subtilis Candidatus Thiodiazotropha endoloripes Homo sapiens Methanosarcina mazei Mus musculus Saccharomyces cerevisiae Solanum lycopersicum Vigna mungo Weighted average
InstaNovo v1.1 0.526 0.629 0.444 0.506 0.571 0.509 0.602 0.615 0.613 0.565
InstaNovo+ v1.1 0.555 0.662 0.475 0.516 0.596 0.525 0.634 0.642 0.662 0.593
π-PrimeNovo 0.592 0.697 0.511 0.566 0.627 0.558 0.688 0.679 0.678 0.630
Prime-DiffNovo 0.607 0.716 0.526 0.572 0.644 0.566 0.698 0.690 0.709 0.645

Papers describing it (2)

Papers using it (2)

Applications and evaluations that ran this method. They are not counted among its authors below.

Authors (19)

Kevin Eloff, Konstantinos Kalogeropoulos, Oliver Morell, Amandla Mabona, Jakob Berg Jespersen, Wesley Williams, Sam P. B. van Beljouw, Marcin J. Skwark, Andreas Hougaard Laustsen, Stan J. J. Brouns, Anne Ljungars, Erwin M. Schoof, Jeroen Van Goey, Ulrich auf dem Keller, Karim Beguir, Nicolas Lopez Carranza, Timothy P. Jenkins, Rachel Catzel, Esperanza Rivera-de-Torre

Seen in the charts

Back to the full map

Back to top