InstaNovo

algorithm · Transformer (AR)

InstaNovo: algorithm · Transformer (AR). Knapsack beam search

Knapsack beam search

Kind	algorithm
Deep learning	yes
Acquisition	DDA
Family	Transformer (AR)

Code

• https://github.com/instadeepai/instanovo

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

• Nine-species benchmark (InstaNovo split) · 2025 peer-reviewed
• Nine-species benchmark (original (DeepNovo, 2017)) · 2023 preprint, 2025 peer-reviewed
• ProteomeTools (Parts I-III) · 2023 preprint, 2025 peer-reviewed
• ProteomeTools (all-confidence (InstaNovo)) · 2023 preprint
• ProteomeTools (high-confidence (InstaNovo)) · 2023 preprint, 2025 peer-reviewed

Evaluated on

• High-throughput MS-based immunopeptidomics (as deposited) · 2023 preprint, 2025 peer-reviewed
• High-throughput proteomics of the 26 medically most important elapids and vipers from sub-Saharan Africa (as deposited) · 2023 preprint, 2025 peer-reviewed
• InstaNovo enables diffusion-powered de novo peptide identification in large scale proteomics experiments (as deposited) · 2023 preprint, 2025 peer-reviewed
• Longitudinal evaluation of biomarkers in wound fluids of venous leg ulcers treated with a protease-modulating wound dressing (as deposited) · 2023 preprint, 2025 peer-reviewed
• Nine-species benchmark (InstaNovo split) · 2023 preprint, 2025 peer-reviewed
• Nine-species benchmark (original (DeepNovo, 2017)) · 2023 preprint, 2025 peer-reviewed
• ProteomeTools (all-confidence (InstaNovo)) · 2023 preprint, 2025 peer-reviewed
• ProteomeTools (high-confidence (InstaNovo)) · 2023 preprint, 2025 peer-reviewed
• Supplementary Data for “Comprehensive evaluation of peptide de novo sequencing tools for monoclonal antibody assembly” (as deposited) · 2023 preprint, 2025 peer-reviewed

Deposited

• InstaNovo enables diffusion-powered de novo peptide identification in large scale proteomics experiments (as deposited) · 2023 preprint, 2025 peer-reviewed

Benchmarks

• denovo_benchmarks: median peptide-level average precision 0.853 over 86 datasets, median rank 2 of 15 (version 1.2.2).

Both are mass-based matches on the tool’s most recent run. What these numbers mean.

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 2

De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experiments, page 58: 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	0.609	0.778	0.815	0.396	0.597	0.411	0.798	0.659	0.695	0.687	0.794	0.658

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	0.608	0.779	0.815	0.398	0.595	0.406	0.798	0.658	0.664	0.685	0.794	0.655

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	0.503	0.581	0.724	0.196	0.447	0.225	0.695	0.494	0.533	0.573	0.685	0.514

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	0.465	0.532	0.697	0.167	0.412	0.190	0.676	0.472	0.497	0.550	0.666	0.490

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.

Extended Data Table 2

InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale proteomics experiments, page 24: 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	0.609	0.778	0.815	0.396	0.597	0.411	0.798	0.659	0.709	0.762	0.695	0.687

a dataset per evaluation set	Amino acid precision
	AC-PT	Mean
InstaNovo	0.794	0.670

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	0.608	0.779	0.815	0.398	0.595	0.406	0.798	0.658	0.626	0.721	0.692	0.685

a dataset per evaluation set	Amino acid accuracy
	AC-PT	Mean
InstaNovo	0.794	0.660

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	0.503	0.581	0.724	0.196	0.447	0.225	0.695	0.494	0.559	0.624	0.466	0.573

a dataset per evaluation set	Peptide accuracy
	AC-PT	Mean
InstaNovo	0.685	0.521

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	0.465	0.532	0.697	0.167	0.412	0.190	0.676	0.472	0.528	0.595	0.428	0.550

a dataset per evaluation set	Peptide AUC
	AC-PT	Mean
InstaNovo	0.666	0.491

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 (12)

44 tables from 12 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.

ScopeNovo: a De Novo Peptide Sequencing Method Using Spectrum-Spectrum Similarity to Refine the Prediction Accuracy (2026) · 2 tables

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

ScopeNovo: a De Novo Peptide Sequencing Method Using Spectrum-Spectrum Similarity to Refine the Prediction Accuracy, page 5: Comparison of de novo peptide sequencing methods on 9-species-V1 and 9-species-V2 benchmarks. For ScopeNovo we report two operating regimes: Setting 1 (External Library, spectrum-consistency reranking against a curated external spectral library), which characterizes the upper-bound performance attainable when high-quality external spectral resources are available; and Setting 2 (Self-Constructed Library), in which the reference library is bootstrapped entirely from high-confidence in-batch de novo predictions on the test data itself, placing ScopeNovo under the same resource assumptions as the baselines and yielding a direct comparison. In each column, the best value among the baselines and ScopeNovo (Setting 2) is in bold; ScopeNovo (Setting 1) is reported separately for reference and excluded from the bold comparison.

Nine-species benchmark revised (build not stated)	Amino acid precision
	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Apis mellifera	Homo sapiens	Methanosarcina mazei	Mus musculus	Vigna mungo	Solanum lycopersicum	Saccharomyces cerevisiae	Average
Casanovo	0.923	0.787	0.776	0.856	0.828	0.806	0.854	0.874	0.865	0.881
Casanovo V2	0.843	0.799	0.788	0.865	0.836	0.813	0.867	0.882	0.875	0.865
InstaNovo	0.852	0.751	0.799	0.847	0.848	0.634	0.894	0.864	0.877	0.850
ContraNovo	0.863	0.802	0.817	0.909	0.864	0.830	0.903	0.881	0.883	0.865
AdaNovo	0.792	0.816	0.756	0.800	0.804	0.757	0.759	0.831	0.801	0.794
RankNovo	0.836	0.836	0.891	0.836	0.879	0.920	0.896	0.893	0.879	0.880
π-PrimeNovo	0.874	0.806	0.824	0.870	0.869	0.808	0.921	0.885	0.891	0.866
ScopeNovo Setting 1	0.885	0.822	0.839	0.884	0.882	0.841	0.931	0.897	0.900	0.885
ScopeNovo Setting 2	0.877	0.814	0.829	0.879	0.874	0.821	0.926	0.889	0.892	0.868

Nine-species benchmark revised (build not stated)	Peptide accuracy
	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Apis mellifera	Homo sapiens	Methanosarcina mazei	Mus musculus	Vigna mungo	Solanum lycopersicum	Saccharomyces cerevisiae	Average
Casanovo	0.687	0.522	0.579	0.662	0.656	0.522	0.645	0.730	0.738	0.682
Casanovo V2	0.693	0.532	0.587	0.670	0.661	0.531	0.654	0.737	0.743	0.683
InstaNovo	0.704	0.512	0.625	0.670	0.666	0.504	0.679	0.721	0.758	0.698
ContraNovo	0.709	0.533	0.615	0.759	0.698	0.567	0.722	0.729	0.725	0.701
AdaNovo	0.636	0.624	0.597	0.590	0.641	0.472	0.538	0.693	0.653	0.635
RankNovo	0.713	0.603	0.617	0.717	0.712	0.559	0.744	0.724	0.680	0.682
π-PrimeNovo	0.712	0.572	0.612	0.710	0.714	0.548	0.766	0.719	0.684	0.697
ScopeNovo Setting 1	0.748	0.664	0.669	0.762	0.772	0.664	0.816	0.771	0.702	0.737
ScopeNovo Setting 2	0.728	0.619	0.641	0.750	0.743	0.611	0.788	0.745	0.691	0.716

The paper bolds the best of the baselines and ScopeNovo (Setting 2) only: 'ScopeNovo (Setting 1) is reported separately for reference and excluded from the bold comparison.' Bold and underline here rank every row, Setting 1 included; the paper's own marks were checked against its rule, and match it.

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

ScopeNovo: a De Novo Peptide Sequencing Method Using Spectrum-Spectrum Similarity to Refine the Prediction Accuracy, page 5: Comparison of de novo peptide sequencing methods on 9-species-V1 and 9-species-V2 benchmarks. For ScopeNovo we report two operating regimes: Setting 1 (External Library, spectrum-consistency reranking against a curated external spectral library), which characterizes the upper-bound performance attainable when high-quality external spectral resources are available; and Setting 2 (Self-Constructed Library), in which the reference library is bootstrapped entirely from high-confidence in-batch de novo predictions on the test data itself, placing ScopeNovo under the same resource assumptions as the baselines and yielding a direct comparison. In each column, the best value among the baselines and ScopeNovo (Setting 2) is in bold; ScopeNovo (Setting 1) is reported separately for reference and excluded from the bold comparison.

Nine-species benchmark original (DeepNovo, 2017)	Amino acid precision
	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Apis mellifera	Homo sapiens	Methanosarcina mazei	Mus musculus	Vigna mungo	Solanum lycopersicum	Saccharomyces cerevisiae	Average
Casanovo	0.788	0.676	0.708	0.676	0.754	0.765	0.761	0.792	0.756	0.743
Casanovo V2	0.801	0.687	0.721	0.687	0.767	0.774	0.768	0.802	0.768	0.754
InstaNovo	0.802	0.656	0.718	0.691	0.756	0.742	0.794	0.774	0.767	0.745
ContraNovo	0.828	0.700	0.764	0.778	0.798	0.803	0.804	0.812	0.787	0.785
AdaNovo	0.858	0.798	0.767	0.690	0.804	0.778	0.878	0.818	0.791	0.798
RankNovo	0.807	0.771	0.880	0.784	0.838	0.823	0.876	0.846	0.828	0.827
π-PrimeNovo	0.840	0.715	0.767	0.725	0.799	0.797	0.823	0.822	0.795	0.789
ScopeNovo Setting 1	0.849	0.715	0.778	0.729	0.811	0.809	0.816	0.825	0.796	0.797
ScopeNovo Setting 2	0.844	0.722	0.774	0.732	0.806	0.809	0.829	0.828	0.798	0.795

Nine-species benchmark original (DeepNovo, 2017)	Peptide accuracy
	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Apis mellifera	Homo sapiens	Methanosarcina mazei	Mus musculus	Vigna mungo	Solanum lycopersicum	Saccharomyces cerevisiae	Average
Casanovo	0.621	0.437	0.494	0.446	0.568	0.485	0.593	0.622	0.601	0.551
Casanovo V2	0.630	0.445	0.504	0.455	0.579	0.493	0.600	0.629	0.607	0.559
InstaNovo	0.652	0.459	0.535	0.515	0.595	0.525	0.635	0.635	0.618	0.582
ContraNovo	0.688	0.486	0.580	0.625	0.633	0.574	0.678	0.674	0.673	0.628
AdaNovo	0.753	0.638	0.620	0.494	0.671	0.546	0.797	0.684	0.660	0.660
RankNovo	0.765	0.592	0.642	0.621	0.688	0.595	0.765	0.720	0.726	0.687
π-PrimeNovo	0.725	0.532	0.602	0.575	0.654	0.572	0.701	0.701	0.703	0.650
ScopeNovo Setting 1	0.789	0.610	0.665	0.628	0.731	0.660	0.755	0.752	0.744	0.711
ScopeNovo Setting 2	0.749	0.572	0.643	0.624	0.690	0.637	0.733	0.732	0.721	0.684

The paper bolds the best of the baselines and ScopeNovo (Setting 2) only: 'ScopeNovo (Setting 1) is reported separately for reference and excluded from the bold comparison.' Bold and underline here rank every row, Setting 1 included; the paper's own marks were checked against its rule, and match it.

dIon: Fragmentation-Based Invariance for Self-Supervised Learning of Tandem Mass Spectra (2026) · 3 tables

Table 26

dIon: Fragmentation-Based Invariance for Self-Supervised Learning of Tandem Mass Spectra, page 33: Complete MSKB-final de novo results, including amino-acid metrics and the dIoninitialized models fine-tuned with a 1,000-peak cap. Within each training regime, scratch and dIon initialization use an identical end-to-end training protocol.

Casanovo digestion enzyme bias data as deposited	Amino acid precision	Amino acid recall	Peptide precision at coverage 1
dIon from scratch, end-to-end; trained on MSKB-final	0.895	0.895	0.747
dIon init., end-to-end; trained on MSKB-final	0.927	0.926	0.802
dIon init., end-to-end, 1,000 peaks; trained on MSKB-final	0.935	0.934	0.822
dIon from scratch, end-to-end; trained on DNL-v1	0.933	0.934	0.819
dIon init., end-to-end; trained on DNL-v1	0.945	0.945	0.842
dIon init., end-to-end, 1,000 peaks; trained on DNL-v1	0.949	0.949	0.856
Casanovo 5.2.1 · released	0.911	0.912	0.783
InstaNovo 1.2.2 · released	0.961	0.961	0.916

MSKB-final, rows grouped by training corpus as in Table 1. InstaNovo 1.2.2 is printed in grey with an asterisk because it 'is documented as training on MassIVE-KB v1, the source corpus of this test set', so its numbers are not a held-out evaluation.
In the paper: column amino acid precision: the original table bolded dIon init., end-to-end, 1,000 peaks; trained on DNL-v1 (0.949).
In the paper: column peptide precision at coverage 1: the original table bolded dIon init., end-to-end, 1,000 peaks; trained on DNL-v1 (0.856).
In the paper: column amino acid recall: the original table bolded dIon init., end-to-end, 1,000 peaks; trained on DNL-v1 (0.949).

Table25

dIon: Fragmentation-Based Invariance for Self-Supervised Learning of Tandem Mass Spectra, page 32: Per-species peptide precision at full coverage on the Kingdoms de novo test set (deterministic run-disjoint split, at most 100k spectra per species).

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Akkermansia muciniphila	Arabidopsis thaliana (callus)	Arabidopsis thaliana (root)	Arabidopsis thaliana (sprout)	Bacillus subtilis	Bacteroides fragilis	Bacteroides thetaiotaomicron	Bacteroides uniformis	Bacteroides vulgatus	Bifidobacterium adolescentis	Bifidobacterium longum subsp. longum	Blautia obeum
dIon init.; DNL-v1, 200 peaks	0.584	0.617	0.620	0.652	0.559	0.631	0.607	0.632	0.612	0.563	0.525	0.606
dIon from scratch; DNL-v1, 200 peaks	0.535	0.573	0.574	0.605	0.527	0.583	0.561	0.585	0.562	0.519	0.487	0.554
Casanovo 5.2.1 · released	0.487	0.528	0.510	0.537	0.506	0.505	0.493	0.512	0.491	0.471	0.451	0.502
InstaNovo 1.2.2 · released	0.546	0.586	0.581	0.620	0.546	0.574	0.554	0.576	0.554	0.537	0.503	0.572

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Bos taurus	Caenorhabditis elegans	Canis lupus	Chlamydomonas reinhardtii	Chromera velia	Clostridium bolteae	Clostridium perfringens	Clostridium saccharolyticum	Collinsella aerofaciens	Coprococcus comes	Cricetulus griseus	Danio rerio
dIon init.; DNL-v1, 200 peaks	0.672	0.633	0.631	0.537	0.535	0.587	0.642	0.608	0.570	0.612	0.551	0.603
dIon from scratch; DNL-v1, 200 peaks	0.617	0.584	0.576	0.504	0.492	0.534	0.594	0.556	0.525	0.560	0.500	0.554
Casanovo 5.2.1 · released	0.540	0.519	0.504	0.476	0.450	0.485	0.538	0.502	0.475	0.500	0.451	0.485
InstaNovo 1.2.2 · released	0.649	0.600	0.607	0.520	0.519	0.553	0.600	0.570	0.538	0.569	0.545	0.575

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Dictyostelium discoideum	Didelphis didelphinae	Dorea formicigenerans	Drosophila melanogaster	Eggerthella lenta	Emiliania huxleyi	Escherichia coli	Eubacterium rectale	Fusarium oxysporum	Fusobacterium nucleatum subsp. nucleatum	Gallus gallus	Glycine max
dIon init.; DNL-v1, 200 peaks	0.622	0.645	0.555	0.640	0.519	0.560	0.619	0.564	0.494	0.641	0.659	0.608
dIon from scratch; DNL-v1, 200 peaks	0.566	0.589	0.514	0.586	0.462	0.515	0.567	0.516	0.457	0.594	0.607	0.561
Casanovo 5.2.1 · released	0.476	0.508	0.485	0.501	0.408	0.449	0.495	0.467	0.438	0.538	0.528	0.494
InstaNovo 1.2.2 · released	0.569	0.611	0.530	0.596	0.482	0.530	0.573	0.521	0.476	0.599	0.622	0.573

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Gossypium hirsutum	Guillardia theta	Haloarcula marismortui	Halobacterium sp.	Homo sapiens	Lactobacillus paracasei	Mus musculus	Mycoplasma intracellular	Neurospora crassa	Odoribacter splanchnicus	Oryctolagus cuniculus	Oryzias melastigma
dIon init.; DNL-v1, 200 peaks	0.607	0.579	0.652	0.556	0.679	0.586	0.611	0.761	0.553	0.598	0.648	0.593
dIon from scratch; DNL-v1, 200 peaks	0.560	0.533	0.616	0.522	0.624	0.541	0.555	0.698	0.511	0.552	0.582	0.539
Casanovo 5.2.1 · released	0.504	0.472	0.581	0.489	0.543	0.489	0.479	0.636	0.461	0.500	0.497	0.471
InstaNovo 1.2.2 · released	0.575	0.550	0.589	0.518	0.662	0.554	0.596	0.743	0.522	0.551	0.611	0.557

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Parabacteroides distasonis	Parabacteroides merdae	Phaeodactylum tricornutum	Porphyra umbilicalis	Prevotella copri	Rattus norvegicus	Roseburia intestinalis	Ruminococcus bromii	Ruminococcus gnavus	Ruminococcus torques	Saccharomyces cerevisiae	Stentor coeruleus
dIon init.; DNL-v1, 200 peaks	0.637	0.627	0.555	0.617	0.646	0.580	0.522	0.595	0.623	0.649	0.650	0.615
dIon from scratch; DNL-v1, 200 peaks	0.585	0.577	0.511	0.574	0.598	0.531	0.476	0.549	0.573	0.600	0.602	0.562
Casanovo 5.2.1 · released	0.523	0.520	0.457	0.497	0.537	0.482	0.445	0.505	0.505	0.530	0.517	0.486
InstaNovo 1.2.2 · released	0.588	0.577	0.532	0.583	0.599	0.580	0.497	0.562	0.578	0.604	0.602	0.572

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Streptococcus parasanguinis	Sulfolobus solfataricus	Sus scrofa	Symbiodinium microadriaticum	Synechocystis nigrescens	Tardigrade	Thalassiosira pseudonana	Triticum aestivum	Veillonella parvula	Vitis vinifera
dIon init.; DNL-v1, 200 peaks	0.627	0.567	0.577	0.551	0.652	0.618	0.562	0.616	0.554	0.581
dIon from scratch; DNL-v1, 200 peaks	0.577	0.514	0.525	0.510	0.615	0.564	0.515	0.569	0.509	0.543
Casanovo 5.2.1 · released	0.518	0.473	0.462	0.477	0.577	0.493	0.454	0.488	0.454	0.510
InstaNovo 1.2.2 · released	0.583	0.543	0.572	0.546	0.633	0.578	0.531	0.577	0.511	0.559

Peptide precision at full coverage per species of dIon's Kingdoms de novo test split. The 'Spectra' column, the number of test spectra per species, is not recorded as a measurement.

Table1 (Kingdoms de novo)

dIon: Fragmentation-Based Invariance for Self-Supervised Learning of Tandem Mass Spectra, page 7: De novo sequencing on the Kingdoms de novo (all 70 species) and MSKB-final test sets. Within each training regime, scratch and dIon initialization use an identical end-to-end training protocol.

The Proteome Landscape of the Kingdoms of Life	Peptide precision at coverage 1
	Kingdoms
dIon from scratch, end-to-end; trained on MSKB-final	0.449
dIon init., end-to-end; trained on MSKB-final	0.533
dIon from scratch, end-to-end; trained on DNL-v1	0.554
dIon init., end-to-end; trained on DNL-v1	0.602
Casanovo 5.2.1 · released	0.492
InstaNovo 1.2.2 · released	0.570

Peptide precision here is at FULL coverage, unanswered spectra counted wrong, which is Casanovo's precision at coverage 1. Rows are grouped by training corpus, recorded in each variant. InstaNovo 1.2.2 has no MSKB-final value: it 'was trained on MassIVE-KB v1, the source of this test set'.

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing (2026) · 6 tables

Table2 (Seven-species)

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing, page 7: PTM-levelperformanceontheNine-Species,SevenSpecies, and HC-PT datasets. Best results are shown in bold.

Seven-species benchmark NovoBench split	Ptm precision	Ptm recall
DeepNovo · quoted	0.391	0.373
PointNovo · quoted	0.117	0.094
InstaNovo · quoted	0.126	0.115
Casanovo · quoted	0.360	0.251
AdaNovo · quoted	0.448	0.321
LIPNovo · quoted	0.604	0.498
LIPNovo+ · quoted	0.648	0.561
LIPNovo reproduced · retrained	0.602	0.500
GyroNovo	0.665	0.565

Table2 (Nine-species)

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing, page 7: PTM-levelperformanceontheNine-Species,SevenSpecies, and HC-PT datasets. Best results are shown in bold.

Nine-species benchmark	Ptm precision	Ptm recall
DeepNovo · quoted	0.576	0.529
PointNovo · quoted	0.629	0.546
InstaNovo · quoted	0.443	0.294
Casanovo · quoted	0.706	0.566
AdaNovo · quoted	0.652	0.570
LIPNovo · quoted	0.765	0.656
LIPNovo+ · quoted	0.801	0.678
LIPNovo reproduced · retrained	0.773	0.646
GyroNovo	0.815	0.731

Table2 (HC-PT)

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing, page 7: PTM-levelperformanceontheNine-Species,SevenSpecies, and HC-PT datasets. Best results are shown in bold.

ProteomeTools HC-PT (NovoBench)	Ptm precision	Ptm recall
DeepNovo · quoted	0.626	0.615
PointNovo · quoted	0.676	0.740
InstaNovo · quoted	0.350	0.261
Casanovo · quoted	0.501	0.460
AdaNovo · quoted	0.552	0.482
LIPNovo · quoted	0.732	0.745
LIPNovo+ · quoted	0.757	0.805
LIPNovo reproduced · retrained	0.690	0.762
GyroNovo	0.790	0.817

Table1 (Seven-species)

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing, page 7: Amino-acid- and peptide-level performance on the Nine-Species, Seven-Species, and HC-PT datasets. Best results are shown in bold.

Seven-species benchmark NovoBench split	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · quoted	0.492	0.433	0.204	0.136
PointNovo · quoted	0.196	0.169	0.022	0.007
Casanovo · quoted	0.322	0.327	0.119	0.084
AdaNovo · quoted	0.379	0.385	0.174	0.135
InstaNovo · quoted	0.192	0.176	0.031	0.009
LIPNovo · quoted	0.557	0.560	0.327	0.281
LIPNovo+ · quoted	0.619	0.622	0.383	0.337
LIPNovo reproduced · retrained	0.553	0.554	0.325	0.270
GyroNovo	0.632	0.635	0.421	0.378

Table1 (Nine-species)

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing, page 7: Amino-acid- and peptide-level performance on the Nine-Species, Seven-Species, and HC-PT datasets. Best results are shown in bold.

Nine-species benchmark	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
PEAKS · quoted	0.748		0.428
DeepNovo · quoted	0.696	0.638	0.428	0.376
PointNovo · quoted	0.740	0.671	0.480	0.436
Casanovo · quoted	0.697	0.696	0.481	0.439
AdaNovo · quoted	0.698	0.709	0.505	0.469
InstaNovo · quoted	0.420	0.395	0.164	0.123
LIPNovo · quoted	0.797	0.797	0.582	0.547
LIPNovo+ · quoted	0.831	0.831	0.637	0.612
LIPNovo reproduced · retrained	0.794	0.796	0.583	0.549
GyroNovo	0.848	0.848	0.674	0.646

Table1 (HC-PT)

GyroNovo: Error-Guided Fragment Imputation with Mass-Aware Attention for De Novo Peptide Sequencing, page 7: Amino-acid- and peptide-level performance on the Nine-Species, Seven-Species, and HC-PT datasets. Best results are shown in bold.

ProteomeTools HC-PT (NovoBench)	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · quoted	0.531	0.534	0.313	0.255
PointNovo · quoted	0.623	0.622	0.419	0.373
Casanovo · quoted	0.442	0.453	0.211	0.177
AdaNovo · quoted	0.442	0.451	0.212	0.178
InstaNovo · quoted	0.289	0.285	0.057	0.034
LIPNovo · quoted	0.637	0.643	0.458	0.427
LIPNovo+ · quoted	0.698	0.694	0.519	0.490
LIPNovo reproduced · retrained	0.653	0.654	0.469	0.438
GyroNovo	0.704	0.702	0.532	0.502

Learning from tandem mass spectra at scale with a self-supervised foundation model for proteomics (2026) · 2 tables

Table S13

Learning from tandem mass spectra at scale with a self-supervised foundation model for proteomics, page 84: Peptide recall across the biological validation datasets for the de novo sequencing InstaNovo-FM and InstaNovo model variant benchmarks. The best model per dataset is in bold. We abbreviate “InstaNovo (FM size matched)” to “IN (FM-SM)” to conserve page space.

a dataset per evaluation set	Peptide recall
	HeLa degradome	Candidatus Scalindua brodae	Snake venoms	HeLa single-shot	Nanobodies	Wound exudates
InstaNovo-FM fine-tuned	0.821	0.740	0.230	0.662	0.528	0.366
InstaNovo-FM from scratch	0.819	0.747	0.239	0.657	0.522	0.354
InstaNovo-FM frozen	0.775	0.726	0.156	0.577	0.474	0.266
InstaNovo	0.821	0.756	0.219	0.652	0.520	0.345
InstaNovo FM size matched	0.823	0.748	0.226	0.650	0.516	0.362

No accession is printed for the six validation sets. They are mapped from the Methods, which call them InstaNovo's application datasets minus 'the Immuno and Herceptin datasets': GluC is InstaNovo's 'HeLa GluC degradome', TPL Antibodies its nanobodies, and Hela QC, by elimination, its HeLa single-shot set.

Table S12

Learning from tandem mass spectra at scale with a self-supervised foundation model for proteomics, page 84: Peptide recall across the biological validation datasets for the de novo sequencing benchmark. The best model per dataset is in bold.

a dataset per evaluation set	Peptide recall
	HeLa degradome	Candidatus Scalindua brodae	Snake venoms	HeLa single-shot	Nanobodies	Wound exudates
InstaNovo-FM fine-tuned	0.821	0.740	0.230	0.662	0.528	0.366
InstaNovo v1.2	0.868	0.764	0.227	0.708	0.517	0.431
Casanovo	0.767	0.743	0.217	0.531	0.465	0.262
XuanjiNovo	0.808	0.741	0.233	0.673	0.529	0.466

No accession is printed for the six validation sets. They are mapped from the Methods, which call them InstaNovo's application datasets minus 'the Immuno and Herceptin datasets': GluC is InstaNovo's 'HeLa GluC degradome', TPL Antibodies its nanobodies, and Hela QC, by elimination, its HeLa single-shot set.

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

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing (2026) · 6 tables

Table2 (Seven-species)

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing, page 7: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance.

Seven-species benchmark NovoBench split	Ptm precision	Ptm recall
DeepNovo · retrained	0.391	0.373
PointNovo · retrained	0.117	0.094
InstaNovo · retrained	0.126	0.115
Casanovo · retrained	0.360	0.251
AdaNovo · retrained	0.448	0.321
π-HelixNovo · retrained	0.473	0.366
LIPNovo · retrained	0.604	0.498
PhysNovo	0.654	0.549

Table2 (Nine-species)

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing, page 7: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance.

Nine-species benchmark revised (balanced)	Ptm precision	Ptm recall
DeepNovo · retrained	0.576	0.529
PointNovo · retrained	0.629	0.546
InstaNovo · retrained	0.443	0.294
Casanovo · retrained	0.706	0.566
AdaNovo · retrained	0.652	0.570
π-HelixNovo · retrained	0.680	0.598
LIPNovo · retrained	0.765	0.656
PhysNovo	0.793	0.686

Table2 (HC-PT)

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing, page 7: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance.

ProteomeTools HC-PT (NovoBench)	Ptm precision	Ptm recall
DeepNovo · retrained	0.626	0.615
PointNovo · retrained	0.676	0.740
InstaNovo · retrained	0.350	0.261
Casanovo · retrained	0.501	0.460
AdaNovo · retrained	0.552	0.482
π-HelixNovo · retrained	0.568	0.667
LIPNovo · retrained	0.732	0.745
PhysNovo	0.763	0.770

Table1 (Seven-species)

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing, page 7: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptidelevel performance. The best results are marked in bold, and the second best are underlined.

Seven-species benchmark NovoBench split	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · retrained	0.492	0.433	0.204	0.136
PointNovo · retrained	0.196	0.169	0.022	0.007
InstaNovo · retrained	0.192	0.176	0.031	0.009
Casanovo · retrained	0.322	0.327	0.119	0.084
AdaNovo · retrained	0.379	0.385	0.174	0.135
π-HelixNovo · retrained	0.481	0.472	0.234	0.173
LIPNovo · retrained	0.557	0.560	0.327	0.281
ReNovo	0.512	0.514	0.278	0.228
PhysNovo	0.568	0.569	0.346	0.290

Table1 (Nine-species)

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing, page 7: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptidelevel performance. The best results are marked in bold, and the second best are underlined.

Nine-species benchmark revised (balanced)	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
PEAKS	0.748		0.428
DeepNovo · retrained	0.696	0.638	0.428	0.376
PointNovo · retrained	0.740	0.671	0.480	0.436
InstaNovo · retrained	0.420	0.395	0.164	0.123
Casanovo · retrained	0.697	0.696	0.481	0.439
AdaNovo · retrained	0.698	0.709	0.505	0.469
π-HelixNovo · retrained	0.765	0.758	0.517	0.453
LIPNovo · retrained	0.797	0.797	0.582	0.547
π-PrimeNovo	0.790
ReNovo	0.770	0.769	0.568	0.528
PhysNovo	0.815	0.814	0.588	0.553

Table1 (HC-PT)

Discrete Diffusion with Physical Mass Constraints for De Novo Peptide Sequencing, page 7: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptidelevel performance. The best results are marked in bold, and the second best are underlined.

ProteomeTools HC-PT (NovoBench)	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · retrained	0.531	0.534	0.313	0.255
PointNovo · retrained	0.623	0.622	0.419	0.373
InstaNovo · retrained	0.289	0.285	0.057	0.034
Casanovo · retrained	0.442	0.453	0.211	0.177
AdaNovo · retrained	0.442	0.451	0.212	0.178
π-HelixNovo · retrained	0.588	0.582	0.356	0.318
LIPNovo · retrained	0.637	0.643	0.458	0.427
ReNovo	0.651	0.648	0.467	0.436
PhysNovo	0.649	0.652	0.473	0.444

MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry (2026) · 2 tables

Table 6

MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry, page 10: Complete four-metric comparison on the Nine Species benchmark for InstaNovo and InstaNovo + MemNovo.

Nine-species benchmark revised (balanced)	Amino acid precision
	Bacillus subtilis	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Homo sapiens	Mus musculus	Candidatus Thiodiazotropha endoloripes	Average
InstaNovo	0.7307	0.7665	0.7410	0.6740	0.7862	0.7851	0.7186	0.6767	0.6242	0.7226
MemNovo on InstaNovo	0.7372	0.7798	0.7383	0.6775	0.7945	0.7912	0.7314	0.6739	0.6433	0.7308

Nine-species benchmark revised (balanced)	Amino acid recall
	Bacillus subtilis	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Homo sapiens	Mus musculus	Candidatus Thiodiazotropha endoloripes	Average
InstaNovo	0.7124	0.7482	0.7281	0.6542	0.7710	0.7623	0.7010	0.6608	0.6028	0.7045
MemNovo on InstaNovo	0.7109	0.7441	0.7334	0.6590	0.7704	0.7553	0.6990	0.6686	0.6053	0.7051

Nine-species benchmark revised (balanced)	Peptide precision
	Bacillus subtilis	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Homo sapiens	Mus musculus	Candidatus Thiodiazotropha endoloripes	Average
InstaNovo	0.6366	0.6877	0.6096	0.5526	0.6904	0.6425	0.6138	0.4922	0.4660	0.5991
MemNovo on InstaNovo	0.6357	0.6925	0.6036	0.5477	0.6904	0.6419	0.6146	0.4843	0.4698	0.5978

Nine-species benchmark revised (balanced)	Peptide recall
	Bacillus subtilis	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Homo sapiens	Mus musculus	Candidatus Thiodiazotropha endoloripes	Average
InstaNovo	0.6249	0.6772	0.5997	0.5384	0.6821	0.6293	0.6031	0.4823	0.4555	0.5881
MemNovo on InstaNovo	0.6259	0.6780	0.6015	0.5408	0.6824	0.6303	0.6026	0.4834	0.4567	0.5891

Table2

MemNovo: Look Back at the Spectrum for Balanced De Novo Peptide Sequencing from Mass Spectrometry, page 7: Performance comparison on the Nine Species benchmark. We report amino acid precision (AA) and peptide precision (Pep.) for each method. For Casanovo and InstaNovo, we additionally show results with MemNovo applied and the corresponding relative improvement (Rel. Imp.). Best results for each species are shown in bold. Results marked with † are reported from original publications, which may use different model versions and evaluation protocols; direct numerical comparison with our evaluation should be interpreted with caution.

Nine-species benchmark revised (balanced)	Amino acid precision
	Mus musculus	Homo sapiens	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Average
DeepNovo · quoted	0.6230	0.6100	0.7500	0.6940	0.6300	0.7310	0.6790	0.7420	0.6020	0.6734
PointNovo · quoted	0.6260	0.6060	0.7790	0.7120	0.6440	0.7330	0.7300	0.7680	0.5890	0.6874
Casanovo	0.6352	0.6885	0.7672	0.7253	0.6730	0.7302	0.7446	0.7701	0.6895	0.7137
AdaNovo · quoted	0.6460	0.6180	0.7930	0.7280	0.6500	0.7400	0.7190	0.7390	0.6420	0.6972
InstaNovo	0.7351	0.7934	0.8447	0.8214	0.7754	0.8181	0.8450	0.8367	0.7631	0.8037
MemNovo on Casanovo	0.6524	0.7031	0.7688	0.7287	0.7749	0.7442	0.7469	0.7941	0.7055	0.7354
MemNovo on InstaNovo	0.7614	0.8130	0.8553	0.8350	0.7934	0.8324	0.8566	0.8504	0.7912	0.8210

Nine-species benchmark revised (balanced)	Peptide precision
	Mus musculus	Homo sapiens	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Average
DeepNovo · quoted	0.2860	0.2930	0.4620	0.4220	0.3300	0.4540	0.4360	0.4490	0.2530	0.3761
PointNovo · quoted	0.3550	0.3510	0.5340	0.4780	0.3960	0.5130	0.5110	0.5180	0.2980	0.4393
Casanovo	0.2404	0.3105	0.4089	0.3254	0.2875	0.3873	0.3314	0.3585	0.2274	0.3197
AdaNovo · quoted	0.4670	0.3730	0.5930	0.4960	0.4310	0.5300	0.5460	0.5280	0.3720	0.4818
InstaNovo	0.3485	0.4562	0.5559	0.4927	0.4318	0.5116	0.5177	0.5185	0.3435	0.4640
MemNovo on Casanovo	0.3372	0.4541	0.4968	0.4295	0.5226	0.5250	0.3530	0.5309	0.3529	0.4447
MemNovo on InstaNovo	0.3522	0.4675	0.5981	0.5051	0.4419	0.5305	0.5481	0.5298	0.3668	0.4822

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing (2026) · 6 tables

Table3 (Seven-species)

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing, page 16: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance. The best and second-best results are marked in bold and underline, respectively.

Seven-species benchmark NovoBench split	Ptm precision	Ptm recall
DeepNovo	0.391	0.373
PointNovo	0.117	0.094
InstaNovo	0.126	0.115
Casanovo · retrained	0.360	0.251
AdaNovo	0.448	0.321
π-HelixNovo	0.473	0.366
SearchNovo	0.472	0.447
Casanovo retrained · retrained	0.368	0.292
LIPNovo	0.604	0.498
LIPNovo+	0.648	0.561

Table3 (Nine-species)

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing, page 16: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance. The best and second-best results are marked in bold and underline, respectively.

Nine-species benchmark	Ptm precision	Ptm recall
DeepNovo	0.576	0.529
PointNovo	0.629	0.546
InstaNovo	0.443	0.294
Casanovo · retrained	0.706	0.566
AdaNovo	0.652	0.570
π-HelixNovo	0.680	0.598
SearchNovo	0.764	0.599
Casanovo retrained · retrained	0.755	0.601
LIPNovo	0.765	0.656
LIPNovo+	0.801	0.678

Table3 (HC-PT)

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing, page 16: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance. The best and second-best results are marked in bold and underline, respectively.

ProteomeTools HC-PT (NovoBench)	Ptm precision	Ptm recall
DeepNovo	0.626	0.615
PointNovo	0.676	0.740
InstaNovo	0.350	0.261
Casanovo · retrained	0.501	0.460
AdaNovo	0.552	0.482
π-HelixNovo	0.568	0.667
SearchNovo	0.715	0.772
Casanovo retrained · retrained	0.550	0.582
LIPNovo	0.732	0.745
LIPNovo+	0.757	0.805

In the paper: column ptm recall: the original table underlined SearchNovo (0.772), LIPNovo (0.745).

Table2 (Seven-species)

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing, page 15: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptide-level performance. The best and second-best results are marked in bold and underline, respectively.

Seven-species benchmark NovoBench split	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo	0.492	0.433	0.204	0.136
PointNovo	0.196	0.169	0.022	0.007
Casanovo · retrained	0.322	0.327	0.119	0.084
AdaNovo	0.379	0.385	0.174	0.135
π-HelixNovo	0.481	0.472	0.234	0.173
InstaNovo	0.192	0.176	0.031	0.009
SearchNovo	0.489	0.488	0.259	0.174
ReNovo	0.512	0.514	0.278	0.228
DiffuNovo	0.437	0.435	0.193	0.162
Casanovo retrained · retrained	0.357	0.366	0.159	0.119
LIPNovo	0.557	0.560	0.327	0.281
LIPNovo+	0.619	0.622	0.383	0.337

Table2 (Nine-species)

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing, page 15: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptide-level performance. The best and second-best results are marked in bold and underline, respectively.

Nine-species benchmark	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
PEAKS	0.748		0.428
DeepNovo	0.696	0.638	0.428	0.376
PointNovo	0.740	0.671	0.480	0.436
Casanovo · retrained	0.697	0.696	0.481	0.439
AdaNovo	0.698	0.709	0.505	0.469
π-HelixNovo	0.765	0.758	0.517	0.453
InstaNovo	0.420	0.395	0.164	0.123
SearchNovo	0.748	0.746	0.550	0.489
ReNovo	0.770	0.769	0.568	0.528
DiffuNovo	0.791	0.789	0.565	0.536
Casanovo retrained · retrained	0.741	0.740	0.529	0.493
LIPNovo	0.797	0.797	0.582	0.547
LIPNovo+	0.831	0.831	0.637	0.612

Table2 (HC-PT)

LIPNovo+: Self-Reflective Latent Imputation for Robust De Novo Peptide Sequencing, page 15: Comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptide-level performance. The best and second-best results are marked in bold and underline, respectively.

ProteomeTools HC-PT (NovoBench)	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo	0.531	0.534	0.313	0.255
PointNovo	0.623	0.622	0.419	0.373
Casanovo · retrained	0.442	0.453	0.211	0.177
AdaNovo	0.442	0.451	0.212	0.178
π-HelixNovo	0.588	0.582	0.356	0.318
InstaNovo	0.289	0.285	0.057	0.034
SearchNovo	0.652	0.658	0.447	0.413
ReNovo	0.651	0.648	0.467	0.436
DiffuNovo	0.654	0.654	0.458	0.434
Casanovo retrained · retrained	0.525	0.530	0.324	0.290
LIPNovo	0.637	0.643	0.458	0.427
LIPNovo+	0.698	0.694	0.519	0.490

In the paper: column amino acid precision: the original table underlined LIPNovo (0.637).

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework (2026) · 6 tables

Table2 (Seven-species)

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework, page 6: Comparison with state-of-the-art models in PTM-level performance. † denotes our retrained results, and others are given by NovoBench. The best is marked in bold.

Seven-species benchmark NovoBench split	Ptm precision	Ptm recall
DeepNovo · quoted	0.391	0.373
PointNovo · quoted	0.117	0.094
InstaNovo · quoted	0.126	0.115
SearchNovo · quoted	0.472	0.447
Casanovo · quoted	0.360	0.251
Casanovo · retrained	0.368	0.292
CausalNovo on †CasaNovo	0.503	0.422
AdaNovo · quoted	0.448	0.321
AdaNovo · retrained	0.430	0.356
CausalNovo on †AdaNovo	0.469	0.398
π-HelixNovo · quoted	0.473	0.366
π-HelixNovo · retrained	0.362	0.370
CausalNovo on †π-HelixNovo	0.513	0.427

Table2 (Nine-species)

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework, page 6: Comparison with state-of-the-art models in PTM-level performance. † denotes our retrained results, and others are given by NovoBench. The best is marked in bold.

Nine-species benchmark	Ptm precision	Ptm recall
DeepNovo · quoted	0.576	0.529
PointNovo · quoted	0.629	0.546
InstaNovo · quoted	0.443	0.294
SearchNovo · quoted	0.764	0.599
Casanovo · quoted	0.706	0.566
Casanovo · retrained	0.755	0.601
CausalNovo on †CasaNovo	0.791	0.608
AdaNovo · quoted	0.652	0.570
AdaNovo · retrained	0.678	0.552
CausalNovo on †AdaNovo	0.769	0.607
π-HelixNovo · quoted	0.680	0.598
π-HelixNovo · retrained	0.723	0.593
CausalNovo on †π-HelixNovo	0.731	0.616

Table2 (HC-PT)

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework, page 6: Comparison with state-of-the-art models in PTM-level performance. † denotes our retrained results, and others are given by NovoBench. The best is marked in bold.

ProteomeTools HC-PT (NovoBench)	Ptm precision	Ptm recall
DeepNovo · quoted	0.626	0.615
PointNovo · quoted	0.676	0.740
InstaNovo · quoted	0.350	0.261
SearchNovo · quoted	0.715	0.772
Casanovo · quoted	0.501	0.460
Casanovo · retrained	0.550	0.582
CausalNovo on †CasaNovo	0.671	0.741
AdaNovo · quoted	0.552	0.482
AdaNovo · retrained	0.562	0.532
CausalNovo on †AdaNovo	0.652	0.743
π-HelixNovo · quoted	0.568	0.667
π-HelixNovo · retrained	0.632	0.566
CausalNovo on †π-HelixNovo	0.737	0.746

Table1 (Seven-species)

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework, page 6: Comparison with state-of-the-art models in amino acid-level and peptide-level performance. † denotes our retrained results, and others are provided by NovoBench. The best results are marked in bold, and the second-best results are underlined.

Seven-species benchmark NovoBench split	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · quoted	0.492	0.433	0.204	0.136
PointNovo · quoted	0.196	0.169	0.022	0.007
InstaNovo · quoted	0.192	0.176	0.031	0.009
SearchNovo · quoted	0.489	0.488	0.259	0.174
Casanovo · quoted	0.322	0.327	0.119	0.084
Casanovo · retrained	0.357	0.366	0.159	0.119
CausalNovo on †CasaNovo	0.477	0.478	0.245	0.200
AdaNovo · quoted	0.379	0.385	0.174	0.135
AdaNovo · retrained	0.403	0.405	0.189	0.149
CausalNovo on †AdaNovo	0.453	0.453	0.233	0.192
π-HelixNovo · quoted	0.481	0.472	0.234	0.173
π-HelixNovo · retrained	0.465	0.462	0.218	0.156
CausalNovo on †π-HelixNovo	0.536	0.534	0.282	0.229

Table1 (Nine-species)

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework, page 6: Comparison with state-of-the-art models in amino acid-level and peptide-level performance. † denotes our retrained results, and others are provided by NovoBench. The best results are marked in bold, and the second-best results are underlined.

Nine-species benchmark	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
PEAKS · quoted	0.748		0.428
DeepNovo · quoted	0.696	0.638	0.428	0.376
PointNovo · quoted	0.740	0.671	0.480	0.436
InstaNovo · quoted	0.420	0.395	0.164	0.123
SearchNovo · quoted	0.748	0.746	0.550	0.489
Casanovo · quoted	0.697	0.696	0.481	0.439
Casanovo · retrained	0.741	0.740	0.529	0.493
CausalNovo on †CasaNovo	0.765	0.766	0.564	0.528
AdaNovo · quoted	0.698	0.709	0.505	0.469
AdaNovo · retrained	0.681	0.681	0.473	0.439
CausalNovo on †AdaNovo	0.744	0.746	0.542	0.507
π-HelixNovo · quoted	0.765	0.758	0.517	0.453
π-HelixNovo · retrained	0.765	0.752	0.509	0.431
CausalNovo on †π-HelixNovo	0.787	0.784	0.543	0.483

Table1 (HC-PT)

CausalNovo: Advancing De Novo Peptide Sequencing via a Causality-Informed Framework, page 6: Comparison with state-of-the-art models in amino acid-level and peptide-level performance. † denotes our retrained results, and others are provided by NovoBench. The best results are marked in bold, and the second-best results are underlined.

ProteomeTools HC-PT (NovoBench)	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · quoted	0.531	0.534	0.313	0.255
PointNovo · quoted	0.623	0.622	0.419	0.373
InstaNovo · quoted	0.289	0.285	0.057	0.034
SearchNovo · quoted	0.652	0.658	0.447	0.413
Casanovo · quoted	0.442	0.453	0.211	0.177
Casanovo · retrained	0.525	0.530	0.324	0.290
CausalNovo on †CasaNovo	0.635	0.639	0.459	0.426
AdaNovo · quoted	0.442	0.451	0.212	0.178
AdaNovo · retrained	0.492	0.496	0.289	0.254
CausalNovo on †AdaNovo	0.634	0.637	0.453	0.420
π-HelixNovo · quoted	0.588	0.582	0.356	0.318
π-HelixNovo · retrained	0.532	0.537	0.301	0.261
CausalNovo on †π-HelixNovo	0.656	0.658	0.450	0.415

Bidirectional Representations Augmented Autoregressive Biological Sequence Generation (2025) · 1 table

Table1

Bidirectional Representations Augmented Autoregressive Biological Sequence Generation, page 8: Comparison of the performance of CROSSNOVO and baseline methods on the 9-species-v1 test set.

The same table is also printed in Bidirectional Representations Augmented Autoregressive Biological Sequence Generation (Table1, page 8).

Nine-species benchmark original (DeepNovo, 2017)	Amino acid precision
	Mus musculus	Homo sapiens	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Average
PEAKS	0.600	0.639	0.748	0.673	0.633	0.728	0.644	0.719	0.586	0.663
π-PrimeNovo	0.784	0.729	0.802	0.801	0.763	0.815	0.822	0.846	0.734	0.788
DeepNovo	0.623	0.610	0.750	0.694	0.630	0.731	0.679	0.742	0.602	0.673
PointNovo	0.626	0.606	0.779	0.712	0.644	0.733	0.730	0.768	0.589	0.687
Casanovo	0.689	0.586	0.684	0.679	0.629	0.721	0.668	0.749	0.603	0.667
InstaNovo	0.703	0.636	0.691	0.712	0.660	0.732	0.711	0.739	0.619	0.689
Casanovo V2	0.760	0.676	0.752	0.755	0.706	0.785	0.748	0.790	0.681	0.739
π-HelixNovo	0.765	0.665	0.768	0.784	0.757	0.721	0.793	0.816	0.681	0.750
ContraNovo	0.798	0.771	0.797	0.799	0.745	0.810	0.807	0.828	0.711	0.785
CrossNovo	0.816	0.800	0.814	0.826	0.785	0.830	0.831	0.856	0.740	0.811

Nine-species benchmark original (DeepNovo, 2017)	Peptide recall
	Mus musculus	Homo sapiens	Saccharomyces cerevisiae	Methanosarcina mazei	Apis mellifera	Solanum lycopersicum	Vigna mungo	Bacillus subtilis	Candidatus Thiodiazotropha endoloripes	Average
PEAKS	0.197	0.277	0.428	0.356	0.287	0.403	0.362	0.387	0.203	0.322
π-PrimeNovo	0.567	0.574	0.697	0.650	0.603	0.697	0.702	0.721	0.531	0.638
DeepNovo	0.286	0.293	0.462	0.422	0.330	0.454	0.436	0.449	0.253	0.376
PointNovo	0.355	0.351	0.534	0.478	0.396	0.513	0.511	0.518	0.298	0.439
Casanovo	0.426	0.341	0.490	0.478	0.406	0.521	0.506	0.537	0.330	0.448
InstaNovo	0.471	0.455	0.559	0.528	0.466	0.732	0.564	0.576	0.416	0.530
Casanovo V2	0.483	0.446	0.599	0.557	0.493	0.618	0.589	0.622	0.446	0.539
π-HelixNovo	0.483	0.392	0.568	0.560	0.473	0.560	0.623	0.596	0.388	0.517
ContraNovo	0.567	0.622	0.674	0.630	0.576	0.672	0.677	0.688	0.486	0.621
CrossNovo	0.596	0.661	0.698	0.660	0.610	0.695	0.716	0.726	0.518	0.654

In the paper: column peptide recall, Mouse: the original table only underlined ContraNovo (Contra.) (0.567); π-PrimeNovo (Prime.) (0.567) ties with it and is underlined here too.
In the paper: column peptide recall, Tomato: the original table bolded π-PrimeNovo (Prime.) (0.697), InstaNovo (Insta) (0.732) and underlined CrossNovo (Ours) (0.695).

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (2025) · 6 tables

Table2 (Seven-species)

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing, page 6: Empirical comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance. † denotes our retrained results, and other results are given by NovoBench. The best results are marked in bold.

The same table is also printed in Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (Table2, page 6).

Seven-species benchmark NovoBench split	Ptm precision	Ptm recall
DeepNovo · quoted	0.391	0.373
PointNovo · quoted	0.117	0.094
InstaNovo · quoted	0.126	0.115
Casanovo · quoted	0.360	0.251
AdaNovo · quoted	0.448	0.321
AdaNovo · retrained	0.430	0.356
π-HelixNovo · quoted	0.473	0.366
π-HelixNovo · retrained	0.362	0.370
Casanovo · retrained	0.368	0.292
LIPNovo	0.604	0.498

Table2 (Nine-species)

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing, page 6: Empirical comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance. † denotes our retrained results, and other results are given by NovoBench. The best results are marked in bold.

The same table is also printed in Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (Table2, page 6).

Nine-species benchmark	Ptm precision	Ptm recall
DeepNovo · quoted	0.576	0.529
PointNovo · quoted	0.629	0.546
InstaNovo · quoted	0.443	0.294
Casanovo · quoted	0.706	0.566
AdaNovo · quoted	0.652	0.570
AdaNovo · retrained	0.678	0.552
π-HelixNovo · quoted	0.680	0.598
π-HelixNovo · retrained	0.723	0.593
Casanovo · retrained	0.755	0.601
LIPNovo	0.765	0.656

Table2 (HC-PT)

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing, page 6: Empirical comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in PTM-level performance. † denotes our retrained results, and other results are given by NovoBench. The best results are marked in bold.

The same table is also printed in Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (Table2, page 6).

ProteomeTools HC-PT (NovoBench)	Ptm precision	Ptm recall
DeepNovo · quoted	0.626	0.615
PointNovo · quoted	0.676	0.740
InstaNovo · quoted	0.350	0.261
Casanovo · quoted	0.501	0.460
AdaNovo · quoted	0.552	0.482
AdaNovo · retrained	0.562	0.532
π-HelixNovo · quoted	0.568	0.667
π-HelixNovo · retrained	0.632	0.566
Casanovo · retrained	0.550	0.582
LIPNovo	0.732	0.745

Table1 (Seven-species)

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing, page 6: Empirical comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptide-level performance. † denotes our retrained results, and other results are provided by NovoBench. The best is marked in bold.

The same table is also printed in Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (Table1, page 6).

Seven-species benchmark NovoBench split	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · quoted	0.492	0.433	0.204	0.136
PointNovo · quoted	0.196	0.169	0.022	0.007
InstaNovo · quoted	0.192	0.176	0.031	0.009
Casanovo · quoted	0.322	0.327	0.119	0.084
AdaNovo · quoted	0.379	0.385	0.174	0.135
AdaNovo · retrained	0.403	0.405	0.189	0.149
π-HelixNovo · quoted	0.481	0.472	0.234	0.173
π-HelixNovo · retrained	0.465	0.462	0.218	0.156
Casanovo · retrained	0.357	0.366	0.159	0.119
LIPNovo	0.557	0.560	0.327	0.281

Table1 (Nine-species)

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing, page 6: Empirical comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptide-level performance. † denotes our retrained results, and other results are provided by NovoBench. The best is marked in bold.

The same table is also printed in Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (Table1, page 6).

Nine-species benchmark	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
PEAKS · quoted	0.748		0.428
DeepNovo · quoted	0.696	0.638	0.428	0.376
PointNovo · quoted	0.740	0.671	0.480	0.436
InstaNovo · quoted	0.420	0.395	0.164	0.123
Casanovo · quoted	0.697	0.696	0.481	0.439
AdaNovo · quoted	0.698	0.709	0.505	0.469
AdaNovo · retrained	0.681	0.681	0.473	0.439
π-HelixNovo · quoted	0.765	0.758	0.517	0.453
π-HelixNovo · retrained	0.765	0.752	0.509	0.431
Casanovo · retrained	0.741	0.740	0.529	0.493
LIPNovo	0.797	0.797	0.582	0.547

In the paper: column amino acid precision: the original table only underlined π-HelixNovo (π-HelixNovo† (Yang et al., 2024)) (0.765); π-HelixNovo (π-HelixNovo (Yang et al., 2024)) (0.765) ties with it and is underlined here too.

Table1 (HC-PT)

Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing, page 6: Empirical comparison with state-of-the-art methods on Nine-species, Seven-species, and HC-PT datasets in amino acid-level and peptide-level performance. † denotes our retrained results, and other results are provided by NovoBench. The best is marked in bold.

The same table is also printed in Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (Table1, page 6).

ProteomeTools HC-PT (NovoBench)	Amino acid precision	Amino acid recall	Peptide precision	Peptide AUC
DeepNovo · quoted	0.531	0.534	0.313	0.255
PointNovo · quoted	0.623	0.622	0.419	0.373
InstaNovo · quoted	0.289	0.285	0.057	0.034
Casanovo · quoted	0.442	0.453	0.211	0.177
AdaNovo · quoted	0.442	0.451	0.212	0.178
AdaNovo · retrained	0.492	0.496	0.289	0.254
π-HelixNovo · quoted	0.588	0.582	0.356	0.318
π-HelixNovo · retrained	0.532	0.537	0.301	0.261
Casanovo · retrained	0.525	0.530	0.324	0.290
LIPNovo	0.637	0.643	0.458	0.427

Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing (2025) · 2 tables

Table6 (Nine-species)

Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing, page 18: Performance comparison on the NovoBench benchmark (yeast test species). Scores for models marked with * are quoted from the NovoBench paper or original publications. CV denotes cross-validation results from the original PrimeNovo paper. “–” indicates data not available.

The same table is also printed in Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing (Table6, page 18).

Nine-species benchmark	Peptide precision
	Saccharomyces cerevisiae
Casanovo · quoted	0.48
InstaNovo · quoted	0.53
AdaNovo · quoted	0.50
π-HelixNovo · quoted	0.52
SearchNovo · quoted	0.55
π-PrimeNovo CV · quoted	0.58
Casanovo pretrained · released	0.60
π-PrimeNovo	0.70
RefineNovo	0.71

Table6 (HC-PT)

Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing, page 18: Performance comparison on the NovoBench benchmark (yeast test species). Scores for models marked with * are quoted from the NovoBench paper or original publications. CV denotes cross-validation results from the original PrimeNovo paper. “–” indicates data not available.

The same table is also printed in Curriculum Learning for Biological Sequence Prediction: The Case of De Novo Peptide Sequencing (Table6, page 18).

ProteomeTools HC-PT (NovoBench)	Peptide precision
Casanovo · quoted	0.21
InstaNovo · quoted	0.57
AdaNovo · quoted	0.21
π-HelixNovo · quoted	0.21
SearchNovo · quoted	0.45
π-PrimeNovo	0.85
RefineNovo	0.88

Papers describing it (2)

• De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experiments (2023, bioRxiv, preprint)
• InstaNovo enables diffusion-powered de novo peptide sequencing in large-scale proteomics experiments (2025, Nature Machine Intelligence, peer-reviewed)

Papers using it (6)

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

• Can we Confidently Sequence the „Peptide Dark Matter“? Reaching a Consensus for De Novo Peptide Sequencing (2026, Proceedings of the 38th European Peptide Symposium, presentation)
• INSearch: AI-native framework for large scale proteomic database search via contrastive joint embeddings and transformer-based scoring function (2026, thesis)
• Sparse Autoencoders for Interpretability of InstaNovo (2026, thesis)
• Generalizable Direct Protein Sequencing With InstaNexus (2026, Molecular & Cellular Proteomics, peer-reviewed)
• Generalizable direct protein sequencing with InstaNexus (2025, bioRxiv, preprint)
• Advancing De Novo Glycopeptide Sequencing with InstaNovo in Glycoproteomics (2025, thesis)

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

• The architectures
• Code activity
• Models and the authors behind them

Back to the full map