ScopeNovo: a De Novo Peptide Sequencing Method Using Spectrum-Spectrum Similarity to Refine the Prediction Accuracy.

ScopeNovo enhances de novo peptide sequencing accuracy by 6.1% and 4.0% on benchmark datasets using spectrum-spectrum similarity reranking.

• Why it matters: Accurate peptide sequencing is crucial for characterizing unknown proteins and identifying modifications, but current models often rely solely on sequence scores without spectrum support, limiting reliability.
• What they did: The method generates multiple candidate peptides with an existing deep learning model and refines predictions by reranking based on spectrum-spectrum similarity, tested with π-PrimeNovo on nine-species datasets.
• The result: ScopeNovo significantly improves peptide accuracy over π-PrimeNovo, achieving 0.711 and 0.737 on V1 and V2 datasets in well-annotated species, and maintains comparable performance in less-annotated species compared to RankNovo.