The impact of noise and missing fragmentation cleavages on de novo peptide identification algorithms
peer-reviewed · Computational and Structural Biotechnology Journal · 2022
| Date | 2022-03-23 |
| Type | peer-reviewed |
| Venue | Computational and Structural Biotechnology Journal |
| Publisher | American Association for the Advancement of Science (AAAS) |
| Contribution | benchmark |
| DOI | 10.1016/j.csbj.2022.03.008 |
| Citations (OpenAlex) | 19 |
| Venue 2-year citedness | 6.57 |
Abstract
Proteomics aims to characterise system-wide protein expression and typically relies on mass-spectrometry and peptide fragmentation, followed by a database search for protein identification. It has wide ranging applications from clinical to environmental settings and virtually impacts on every area of biology. In that context, de novo peptide sequencing is becoming increasingly popular. Historically its performance lagged behind database search methods but with the integration of machine learning, this field of research is gaining momentum. To enable de novo peptide sequencing to realise its full potential, it is critical to explore the mass spectrometry data underpinning peptide identification. In this research we investigate the characteristics of tandem mass spectra using 8 published datasets. We then evaluate two state of the art de novo peptide sequencing algorithms, Novor and DeepNovo, with a particular focus on their performance with regard to missing fragmentation cleavage sites and noise. DeepNovo was found to perform better than Novor overall. However, Novor recalled more correct amino acids when 6 or more cleavage sites were missing. Furthermore, less than 11% of each algorithms’ correct peptide predictions emanate from data with more than one missing cleavage site, highlighting the issues missing cleavages pose. We further investigate how the algorithms manage to correctly identify peptides with many of these missing fragmentation cleavages. We show how noise negatively impacts the performance of both algorithms, when high intensity peaks are considered. Finally, we provide recommendations regarding further algorithms’ improvements and offer potential avenues to overcome current inherent data limitations.
Methods and tools
- De novo robustness to noise and missing cleavages: Evaluation of how de novo sequencing algorithms degrade as spectra lose fragmentation cleavages and gain noise, run over eight published datasets with Novor and DeepNovo.
Data used
- Aggregating sequences that occur in many proteins constitute weak spots of bacterial proteostasis (as deposited) · PXD008685
- Identification and quantification of small proteins and peptides in Staphylococcus aureus (as deposited) · PXD017932
- NlpI scaffolds hydrolases within peptidoglycan multi-enzyme complexes in Escherichia coli (as deposited) · PXD016825
- Optimized data analysis avoiding trypsin artefacts (as deposited) · PXD002726
- PIA - Mouse Benchmark Dataset - PIA - An intuitive protein inference engine with a web-based user interface (as deposited) · PXD000790
- PeptideShaker - A complex standard for protein identification, designed by evolution (as deposited) · PXD001077
- PeptideShaker - PeptideShaker is a user friendly tool for shotgun proteomic data interpretation and reprocessing. (as deposited) · PXD000674
- Staphylococcus aureus responds to physiologically relevant temperature changes by altering its global transcript and pro (as deposited) · PXD023039
Cites (18)
- Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning (2019) both
- Deep learning enables de novo peptide sequencing from data-independent-acquisition mass spectrometry (2018) both
- A potential golden age to come—current tools, recent use cases, and future avenues for de novo sequencing in proteomics (2018) both
- SearchGUI: A Highly Adaptable Common Interface for Proteomics Search and de Novo Engines (2018) both
- De novo peptide sequencing by deep learning (2017) both
- Evaluating de novo sequencing in proteomics: already an accurate alternative to database-driven peptide identification? (2017) both
- Novor: Real-Time Peptide de Novo Sequencing Software (2015) both
- Lessons in de novo peptide sequencing by tandem mass spectrometry (2015) both
- DeNovoGUI: An Open Source Graphical User Interface for de Novo Sequencing of Tandem Mass Spectra (2014) both
- PEAKS DB: De Novo Sequencing Assisted Database Search for Sensitive and Accurate Peptide Identification (2012) both
- Spectral Dictionaries: Integrating de novo Peptide Sequencing with Database Search of Tandem Mass Spectra (2009) both
- A comparative study of the accuracy of several de novo sequencing software packages for datasets derived by matrix‐assisted laser desorption/ionisation and electrospray (2008) both
- MSNovo: A Dynamic Programming Algorithm for de Novo Peptide Sequencing via Tandem Mass Spectrometry (2007) both
- De Novo Peptide Sequencing and Identification with Precision Mass Spectrometry (2007) both
- Peptide Sequence Tags for Fast Database Search in Mass-Spectrometry (2005) both
- Algorithms for de novo peptide sequencing using tandem mass spectrometry (2004) both
- A Dynamic Programming Approach to De Novo Peptide Sequencing via Tandem Mass Spectrometry (2001) both
- De novo peptide sequencing via tandem mass spectrometry (1999) both
Cited by (18)
- π-MNovo improves de novo peptide sequencing through microbial-domain adaptation and evidence-guided candidate selection (2026) crossref
- Protein Language Model-Aligned Spectra Embeddings for De Novo Peptide Sequencing (2025) both
- Framework for de novo sequencing of peptide mixtures via network analysis and two-dimensional tandem mass spectrometry (2025) crossref
- De novo peptide databases enable protein-based stable isotope probing of microbial communities with up to species-level resolution (2025) both
- Limitations of de novo sequencing in resolving sequence ambiguity (2025) both
- Latent Imputation before Prediction: A New Computational Paradigm for De Novo Peptide Sequencing (2025) semanticscholar
- NovoRank: Refinement for De Novo Peptide Sequencing Based on Spectral Clustering and Deep Learning (2025) crossref
- De novo peptide databases enable protein-based stable isotope probing of microbial communities with up to species-level resolution (2024) crossref
- BERT model for de novo protein sequencing using tandem mass spectra (2024) crossref
- PowerNovo: de novo peptide sequencing via tandem mass spectrometry using an ensemble of transformer and BERT models (2024) both
- NovoBench: Benchmarking Deep Learning-based De Novo Peptide Sequencing Methods in Proteomics (2024) semanticscholar
- pXg: Comprehensive Identification of Noncanonical MHC-I-Associated Peptides From De Novo Peptide Sequencing Using RNA-Seq Reads (2024) both
- Mitigating the missing-fragmentation problem in de novo peptide sequencing with a two-stage graph-based deep learning model (2023) both
- De Novo Sequencing of Synthetic Bis -cysteine Peptide Macrocycles Enabled by “Chemical Linearization” of Compound Mixtures (2023) both
- Application of a Novel Hybrid CNN-GNN for Peptide Ion Encoding (2023) both
- Comprehensive evaluation of peptide de novo sequencing tools for monoclonal antibody assembly (2023) both
- Critical evaluation of the use of artificial data for machine learning based de novo peptide identification (2023) both
- Current state, existing challenges, and promising progress for de novo sequencing and assembly of monoclonal antibodies (2022) both