To Fly, or Not to Fly, That Is the Question: A Deep Learning Model for Peptide Detectability Prediction in Mass Spectrometry
Uložené v:
| Názov: | To Fly, or Not to Fly, That Is the Question: A Deep Learning Model for Peptide Detectability Prediction in Mass Spectrometry |
|---|---|
| Autori: | Naim Abdul-Khalek, Mario Picciani, Omar Shouman, Reinhard Wimmer, Michael Toft Overgaard, Mathias Wilhelm, Simon Gregersen Echers |
| Zdroj: | J Proteome Res Abdul-Khalek, N, Picciani, M, Shouman, O, Wimmer, R, Overgaard, M T, Wilhelm, M & Gregersen Echers, S 2025, 'To Fly, or Not to Fly, That Is the Question : A Deep Learning Model for Peptide Detectability Prediction in Mass Spectrometry', Journal of Proteome Research, vol. 24, no. 6, pp. 2709-2726. https://doi.org/10.1021/acs.jproteome.4c00973 |
| Informácie o vydavateľovi: | American Chemical Society (ACS), 2025. |
| Rok vydania: | 2025 |
| Predmety: | Attention mechanism, Computational proteomics, Encoder-decoder, Peptides/analysis, Classification, Article, ddc, Flyability, Rescoring, Deep Learning, Bottom-up proteomics, Peptide Library, ddc:630, Animals, Humans, Amino Acid Sequence, Proteomics/methods, Peptide detectability, Mass Spectrometry/methods, Software |
| Popis: | Identifying detectable peptides, known as flyers, is key in mass spectrometry-based proteomics. Peptide detectability is strongly related to peptide sequences and their resulting physicochemical properties. Moreover, the high variability in MS data challenges the development of a generic model for detectability prediction, underlining the need for customizable tools. We present Pfly, a deep learning model developed to predict peptide detectability based solely on peptide sequence. Pfly is a versatile and reliable state-of-the-art tool, offering high performance, accessibility, and easy customizability for end-users. This adaptability allows researchers to tailor Pfly to specific experimental conditions, improving accuracy and expanding applicability across various research fields. Pfly is an encoder-decoder with an attention mechanism, classifying peptides as flyers or non-flyers, and providing both binary and categorical probabilities for four distinct classes defined in this study. The model was initially trained on a synthetic peptide library and subsequently fine-tuned with a biological dataset to mitigate bias toward synthesizability, improving predictive capacity and outperforming state-of-the-art predictors in benchmark comparisons across different human and cross-species datasets. The study further investigates the influence of protein abundance and rescoring, illustrating the negative impact on peptide identification due to misclassification. Pfly has been integrated into the DLOmix framework and is accessible on GitHub at https://github.com/wilhelm-lab/dlomix. |
| Druh dokumentu: | Article Other literature type |
| Popis súboru: | application/pdf |
| Jazyk: | English |
| ISSN: | 1535-3907 1535-3893 |
| DOI: | 10.1021/acs.jproteome.4c00973 |
| Prístupová URL adresa: | https://pubmed.ncbi.nlm.nih.gov/40344201 https://mediatum.ub.tum.de/doc/1784920/document.pdf |
| Rights: | CC BY NC ND URL: http://creativecommons.org/licenses/by-nc-nd/4.0/This article is licensed under CC-BY-NC-ND 4.0 |
| Prístupové číslo: | edsair.doi.dedup.....121b5ccdab3110ca1e2f993f57cdd94d |
| Databáza: | OpenAIRE |
Nájsť tento článok vo Web of Science | Abstrakt: | Identifying detectable peptides, known as flyers, is key in mass spectrometry-based proteomics. Peptide detectability is strongly related to peptide sequences and their resulting physicochemical properties. Moreover, the high variability in MS data challenges the development of a generic model for detectability prediction, underlining the need for customizable tools. We present Pfly, a deep learning model developed to predict peptide detectability based solely on peptide sequence. Pfly is a versatile and reliable state-of-the-art tool, offering high performance, accessibility, and easy customizability for end-users. This adaptability allows researchers to tailor Pfly to specific experimental conditions, improving accuracy and expanding applicability across various research fields. Pfly is an encoder-decoder with an attention mechanism, classifying peptides as flyers or non-flyers, and providing both binary and categorical probabilities for four distinct classes defined in this study. The model was initially trained on a synthetic peptide library and subsequently fine-tuned with a biological dataset to mitigate bias toward synthesizability, improving predictive capacity and outperforming state-of-the-art predictors in benchmark comparisons across different human and cross-species datasets. The study further investigates the influence of protein abundance and rescoring, illustrating the negative impact on peptide identification due to misclassification. Pfly has been integrated into the DLOmix framework and is accessible on GitHub at https://github.com/wilhelm-lab/dlomix. |
|---|---|
| ISSN: | 15353907 15353893 |
| DOI: | 10.1021/acs.jproteome.4c00973 |