MSBooster: improving peptide identification rates using deep learning-based features

Yang, Kevin; Yu, Fengchao; Teo, Guo Ci; Li, Kai; Demichev, Vadim; Ralser, Markus; Nesvizhskii, Alexey I.

doi:10.1038/s41467-023-40129-9

Cited by 74 publications

(54 citation statements)

References 91 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All other settings were set to default. Validation: the following validation tools were used with the default settings: MSBooster, 22 Predict RT, Predict spectra, Percolator, 23 ProteinProphet, and reports were filtered at a 1% false discovery rate (FDR). Quantitation: IonQuant 24 was used for MS1 quantitation, with matching-between-runs at 1% FDR, intensities were normalized across runs, and unique peptides were considered.…”

Section: Database Search Analysis and Data Analysismentioning

confidence: 99%

Accelerating Proteomics Using Broad Specificity Proteases

Jiang,

Yeung,

Liu

et al. 2024

J. Proteome Res.

View full text Add to dashboard Cite

Trypsin is the gold-standard protease in bottom-up proteomics, but many sequence stretches of the proteome are inaccessible to trypsin and standard LC-MS approaches. Thus, multienzyme strategies are used to maximize sequence coverage in posttranslational modification profiling. We present fast and robust SP3-and STRAP-based protocols for the broad-specificity proteases subtilisin, proteinase K, and thermolysin. All three enzymes are remarkably fast, producing near-complete digests in 1−5 min, and cost 200− 1000× less than proteomics-grade trypsin. Using FragPipe resolved a major challenge by drastically reducing the duration of the required "unspecific" searches. In-depth analyses of proteinase K, subtilisin, and thermolysin Jurkat digests identified 7374, 8178, and 8753 unique proteins with average sequence coverages of 21, 29, and 37%, including 10,000s of amino acids not reported in PeptideAtlas' >2400 experiments. While we could not identify distinct cleavage patterns, machine learning could distinguish true protease products from random cleavages, potentially enabling the prediction of cleavage products. Finally, proteinase K, subtilisin, and thermolysin enabled label-free quantitation of 3111, 3659, and 4196 unique Jurkat proteins, which in our hands is comparable to trypsin. Our data demonstrate that broad-specificity proteases enable quantitative proteomics of uncharted areas of the proteome. Their fast kinetics may allow "on-the-fly" digestion of samples in the future.

show abstract

Section: Database Search Analysis and Data Analysismentioning

confidence: 99%

Accelerating Proteomics Using Broad Specificity Proteases

Jiang,

Yeung,

Liu

et al. 2024

J. Proteome Res.

View full text Add to dashboard Cite

show abstract

“…In addition to the existing commercially based software which identify peptides based on a PSM, development of script‐based high‐performance peptide search algorithms have also been reported. The search scripts of MSFragger and FragPipe, together with machine learning nodes of MSBooster and MSFragger‐Glyco, have been developed recently to enhance the identification of immunopeptides and glycosylated immunopeptides, respectively 60‐64 . For more variations of immunopeptides with post‐translational modifications (PTMs), PROMISE has been developed, which delineated the preference of the PTMs in a certain HLA allotype 65 …”

Section: Transformative Technologies That Shed Light Upon the Dark Ma...mentioning

confidence: 99%

Emerging potential of immunopeptidomics by mass spectrometry in cancer immunotherapy

Minegishi,

Haga,

Ueda

2024

Cancer Science

View full text Add to dashboard Cite

With significant advances in analytical technologies, research in the field of cancer immunotherapy, such as adoptive T cell therapy, cancer vaccine, and immune checkpoint blockade (ICB), is currently gaining tremendous momentum. Since the efficacy of cancer immunotherapy is recognized only by a minority of patients, more potent tumor‐specific antigens (TSAs, also known as neoantigens) and predictive markers for treatment response are of great interest. In cancer immunity, immunopeptides, presented by human leukocyte antigen (HLA) class I, play a role as initiating mediators of immunogenicity. The latest advancement in the interdisciplinary multiomics approach has rapidly enlightened us about the identity of the “dark matter” of cancer and the associated immunopeptides. In this field, mass spectrometry (MS) is a viable option to select because of the naturally processed and actually presented TSA candidates in order to grasp the whole picture of the immunopeptidome. In the past few years the search space has been enlarged by the multiomics approach, the sensitivity of mass spectrometers has been improved, and deep/machine‐learning‐supported peptide search algorithms have taken immunopeptidomics to the next level. In this review, along with the introduction of key technical advancements in immunopeptidomics, the potential and further directions of immunopeptidomics will be reviewed from the perspective of cancer immunotherapy.

show abstract

“…Existing rescoring tools mainly differ from each other by their use of distinct feature sets and prediction models. Some PSM rescoring tools use only a few features by default [30–32], while others use dozens [33, 34] to 100 features [35, 36]. Additionally, some tools allow adjusting the number of features used.…”

Section: Common Feature Types Used During Immunopeptide Psm Rescoringmentioning

confidence: 99%

Machine learning‐based peptide‐spectrum match rescoring opens up the immunopeptidome

Adams,

Laukens,

Bittremieux

et al. 2023

Proteomics

View full text Add to dashboard Cite

Immunopeptidomics is a key technology in the discovery of targets for immunotherapy and vaccine development. However, identifying immunopeptides remains challenging due to their non‐tryptic nature, which results in distinct spectral characteristics. Moreover, the absence of strict digestion rules leads to extensive search spaces, further amplified by the incorporation of somatic mutations, pathogen genomes, unannotated open reading frames, and post‐translational modifications. This inflation in search space leads to an increase in random high‐scoring matches, resulting in fewer identifications at a given false discovery rate. Peptide‐spectrum match rescoring has emerged as a machine learning‐based solution to address challenges in mass spectrometry‐based immunopeptidomics data analysis. It involves post‐processing unfiltered spectrum annotations to better distinguish between correct and incorrect peptide‐spectrum matches. Recently, features based on predicted peptidoform properties, including fragment ion intensities, retention time, and collisional cross section, have been used to improve the accuracy and sensitivity of immunopeptide identification. In this review, we describe the diverse bioinformatics pipelines that are currently available for peptide‐spectrum match rescoring and discuss how they can be used for the analysis of immunopeptidomics data. Finally, we provide insights into current and future machine learning solutions to boost immunopeptide identification.

show abstract

MSBooster: improving peptide identification rates using deep learning-based features

Cited by 74 publications

References 91 publications

Accelerating Proteomics Using Broad Specificity Proteases

Accelerating Proteomics Using Broad Specificity Proteases

Emerging potential of immunopeptidomics by mass spectrometry in cancer immunotherapy

Machine learning‐based peptide‐spectrum match rescoring opens up the immunopeptidome

Contact Info

Product

Resources

About