4.8 Article

AlphaPeptDeep: a modular deep learning framework to predict peptide properties for proteomics

相关参考文献

注意:仅列出部分参考文献,下载原文获取全部文献信息。
Review Biotechnology & Applied Microbiology

Identification of tumor antigens with immunopeptidomics

Chloe Chong et al.

Summary: Identification of actionable tumor antigens is crucial for various cancer immunotherapies, including T cell receptor-transduced T cells and patient-specific mRNA or peptide vaccines. By eluting human leukocyte antigen-bound peptides from tumors and subjecting these to mass spectrometry analysis, noncanonical antigens can be discovered, which may lead to more effective treatment options.

NATURE BIOTECHNOLOGY (2022)

Article Biochemical Research Methods

Prosit Transformer: A transformer for Prediction of MS2 Spectrum Intensities

Markus Ekvall et al.

Summary: Machine learning has long been essential in interpreting proteomics data from mass spectrometry. Recently, the Transformer model, successful in other fields of bioinformatics, has been applied with the convenience of transfer learning. In this study, a Transformer based on the TAPE model was implemented to predict MS2 intensities, outperforming the traditional recurrent neural network-based predictor Prosit.

JOURNAL OF PROTEOME RESEARCH (2022)

Review Biotechnology & Applied Microbiology

The emerging role of mass spectrometry-based proteomics in drug discovery

Felix Meissner et al.

Summary: This Review discusses the potential of proteomics and chemoproteomics approaches in drug discovery, specifically in target identification, validation, and identification of safety hazards. The advancements in biochemical methods and mass spectrometry-based proteomics have allowed for a better understanding of disease mechanisms and the development of targeted drugs.

NATURE REVIEWS DRUG DISCOVERY (2022)

Article Multidisciplinary Sciences

A streamlined platform for analyzing tera-scale DDA and DIA mass spectrometry data enables highly sensitive immunopeptidomics

Lei Xin et al.

Summary: In this study, the authors present a computational platform that integrates data-dependent and data-independent acquisition mass spectrometry approaches for deeper immunopeptidome analysis. The platform utilizes deep learning-based solutions to enhance sensitivity and accuracy of peptide identification, leading to the discovery of more peptide precursors compared to other systems. Additionally, the platform identifies potential T-cell epitopes for COVID-19 vaccine development in the SARS-CoV-2 immunopeptidome.

NATURE COMMUNICATIONS (2022)

Article Biochemical Research Methods

Effective gene expression prediction from sequence by integrating long-range interactions

Ziga Avsec et al.

Summary: Enformer leverages a new deep learning architecture to improve gene expression prediction accuracy based on DNA sequences, integrating information from long-range interactions in the genome and accurately predicting the impact of genetic variants on gene expression. Additionally, Enformer has learned to predict enhancer-promoter interactions directly from DNA sequences.

NATURE METHODS (2021)

Article Biochemical Research Methods

DeepLC can predict retention times for peptides that carry as-yet unseen modifications

Robbin Bouwmeester et al.

Summary: DeepLC is a deep learning-based peptide retention time predictor that can accurately predict retention times for unmodified peptides and peptides with previously unseen modifications, addressing peptide identification ambiguity in complex mass spectrometry workflows. By using atomic composition-based peptide encoding, DeepLC is able to accurately predict retention times for a wide range of modifications, including those not seen during training, potentially enabling incorrect identifications to be flagged in proteome data analysis.

NATURE METHODS (2021)

Article Multidisciplinary Sciences

Deep learning the collisional cross sections of the peptide universe from a million experimental values

Florian Meier et al.

Summary: Researchers have developed a deep learning model that can accurately predict peptide collisional cross section (CCS) values, based on a large dataset of over one million data points. In addition to peptide sequences, hydrophobicity, proportion of prolines, and position of histidines also contribute to the prediction of CCS values. This study lays the foundation for predicting CCS values and advancing proteomics workflows.

NATURE COMMUNICATIONS (2021)

Article Multidisciplinary Sciences

DeepPhospho accelerates DIA phosphoproteome profiling through in silico library generation

Ronghui Lou et al.

Summary: The authors developed a DIA workflow based on in silico spectral libraries generated by a deep neural network to expand phosphoproteome coverage. By leveraging the predicted libraries from DeepPhospho, they established a DIA workflow for phosphoproteome profiling, circumventing the need of DDA library construction and significantly expanding the coverage of phosphoproteome while maintaining high quantification performance. DeepPhospho is provided as a web server and offline app for model training, predictions, and library generation.

NATURE COMMUNICATIONS (2021)

Article Multidisciplinary Sciences

Highly accurate protein structure prediction for the human proteome

Kathryn Tunyasuvunakool et al.

Summary: Using the AlphaFold method, the structural coverage of the proteome has been significantly expanded, covering 98.5% of human proteins with 58% of residues having confident predictions and 36% having very high confidence. Introducing new metrics to interpret the dataset and identify disordered regions, this study aims to provide high-quality predictions for generating biological hypotheses.

NATURE (2021)

Article Multidisciplinary Sciences

Deep learning boosts sensitivity of mass spectrometry-based immunopeptidomics

Mathias Wilhelm et al.

Summary: The authors improved the prediction of non-tryptic peptides using deep learning and data from the ProteomeTools project, enabling various immunological discoveries. Their study demonstrated that accurate identification of HLA peptides can be enhanced, leading to the discovery of additional immunogenic neo-epitopes from patient tumors in published data.

NATURE COMMUNICATIONS (2021)

Article Biochemistry & Molecular Biology

Artificial intelligence for proteomics and biomarker discovery

Matthias Mann et al.

Summary: The rapid growth of biomedical data generation and computational capabilities has led to advancements in utilizing machine learning and deep learning in proteomics for predictive modeling and biomarker discovery. These technologies are essential for improving analytical workflows and integrating multi-omics data, while also raising concerns about model transparency, explainability, and data privacy when deploying MS-based biomarkers in clinical settings.

CELL SYSTEMS (2021)

Article Biochemical Research Methods

Sensitive Immunopeptidomics by Leveraging Available Large-Scale Multi-HLA Spectral Libraries, Data-Independent Acquisition, and MS/MS Prediction

HuiSong Pak et al.

Summary: Mass spectrometry is a cutting-edge methodology for studying the immunopeptidome, with data-dependent tandem MS (DDA) and data-independent acquisition (DIA) being two commonly used approaches. The study found that DIA method is highly sensitive for personalized neoantigen discovery, especially when combined with a comprehensive multi-HLA library and MS/MS prediction, which can significantly improve peptide identification efficiency.

MOLECULAR & CELLULAR PROTEOMICS (2021)

Article Biochemical Research Methods

Trapped Ion Mobility Spectrometry and Parallel Accumulation-Serial Fragmentation in Proteomics

Florian Meier et al.

Summary: The recent development of TIMS coupled to TOF mass analysis, particularly the PASEF operation mode, offers advantages in sequencing speed and sensitivity. By synchronizing ion release with downstream precursor selection, PASEF can increase the number of peptide fragment ion spectra obtained while maintaining sensitivity. The correlation between ion mobility and mass, as well as the ability to predict collisional cross section values based on peptide sequences, make TIMS quadrupole TOF PASEF a powerful platform for proteomics and beyond.

MOLECULAR & CELLULAR PROTEOMICS (2021)

Article Biochemical Research Methods

Mild Acid Elution and MHC Immunoaffinity Chromatography Reveal Similar Albeit Not Identical Profiles of the HLA Class I Immunopeptidome

Theo Sturm et al.

Summary: This study shows that an optimized mild acid elution (MAE) method can effectively isolate MHC peptide ligands with high purity, providing advantages such as the discovery of unmodified cysteine residues in MHC ligands and enabling high-confidence analysis of post-translational modifications. The results suggest that MAE could be a valuable alternative to MHC immunoaffinity chromatography (MHC-IAC) for suspension cells, offering a cost-effective solution for comprehensive and unbiased characterization of MHC peptide ligands.

JOURNAL OF PROTEOME RESEARCH (2021)

Article Biochemical Research Methods

DIA-NN: neural networks and interference correction enable deep proteome coverage in high throughput

Vadim Demichev et al.

NATURE METHODS (2020)

Article Biochemical Research Methods

Logomaker: beautiful sequence logos in Python

Ammar Tareen et al.

BIOINFORMATICS (2020)

Article Biotechnology & Applied Microbiology

A large peptidome dataset improves HLA class I epitope prediction across most of the human population

Siranush Sarkizova et al.

NATURE BIOTECHNOLOGY (2020)

Article Multidisciplinary Sciences

Cancer neoantigen prioritization through sensitive and reliable proteogenomics analysis

Bo Wen et al.

NATURE COMMUNICATIONS (2020)

Article Multidisciplinary Sciences

The proteome landscape of the kingdoms of life

Johannes B. Mueller et al.

NATURE (2020)

Article Biochemistry & Molecular Biology

NAguideR: performing and prioritizing missing value imputations for consistent bottom-up proteomic analyses

Shisheng Wang et al.

NUCLEIC ACIDS RESEARCH (2020)

Review Biochemical Research Methods

Deep Learning in Proteomics

Bo Wen et al.

PROTEOMICS (2020)

Article Biochemical Research Methods

Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning

Siegfried Gessulat et al.

NATURE METHODS (2019)

Article Chemistry, Analytical

MS/MS Spectrum Prediction for Modified Peptides Using pDeep2 Trained by Transfer Learning

Wen-Feng Zeng et al.

ANALYTICAL CHEMISTRY (2019)

Article Chemistry, Analytical

Improved Peptide Retention Time Prediction in Liquid Chromatography through Deep Learning

Chunwei Ma et al.

ANALYTICAL CHEMISTRY (2018)

Review Biochemistry & Molecular Biology

Data-independent acquisition-based SWATH-MS for quantitative proteomics: a tutorial

Christina Ludwig et al.

MOLECULAR SYSTEMS BIOLOGY (2018)

Article Immunology

The Length Distribution and Multiple Specificity of Naturally Presented HLA-I Ligands

David Gfeller et al.

JOURNAL OF IMMUNOLOGY (2018)

Article Biotechnology & Applied Microbiology

Comprehensive identification of peptides in tandem mass spectra using an efficient open search engine

Hao Chi et al.

NATURE BIOTECHNOLOGY (2018)

Article Biochemical Research Methods

A Molecular Basis for the Presentation of Phosphorylated Peptides by HLA-B Antigens

Adab Alpizar et al.

MOLECULAR & CELLULAR PROTEOMICS (2017)

Article Biochemical Research Methods

Building ProteomeTools based on a complete synthetic human proteome

Daniel P. Zolg et al.

NATURE METHODS (2017)

Article Biochemical Research Methods

MSFragger: ultrafast and comprehensive peptide identification in mass spectrometry-based proteomics

Andy T. Kong et al.

NATURE METHODS (2017)

Article Chemistry, Analytical

pDeep: Predicting MS/MS Spectra of Peptides with Deep Learning

Xie-Xuan Zhou et al.

ANALYTICAL CHEMISTRY (2017)

Review Multidisciplinary Sciences

Mass-spectrometric exploration of proteome structure and function

Ruedi Aebersold et al.

NATURE (2016)

Article Multidisciplinary Sciences

Direct identification of clinically relevant neoepitopes presented on native human melanoma tissue by mass spectrometry

Michal Bassani-Sternberg et al.

NATURE COMMUNICATIONS (2016)

Article Biochemical Research Methods

DIA-Umpire: comprehensive computational framework for data-independent acquisition proteomics

Chih-Chiang Tsou et al.

NATURE METHODS (2015)

Article Multidisciplinary Sciences

A repository of assays to quantify 10,000 human proteins by SWATH-MS

George Rosenberger et al.

SCIENTIFIC DATA (2014)

Article Biochemical Research Methods

A cross-validation scheme for machine learning algorithms in shotgun proteomics

Viktor Granholm et al.

BMC BIOINFORMATICS (2012)

Article Biochemical Research Methods

Using iRT, a normalized retention time for more targeted measurement of peptides

Claudia Escher et al.

PROTEOMICS (2012)

Article Biochemical Research Methods

Speeding up tandem mass spectrometry-based database searching by longest common prefix

Chen Zhou et al.

BMC BIOINFORMATICS (2010)

Article Biochemical Research Methods

Training, Selection, and Robust Calibration of Retention Time Models for Targeted Proteomics

Luminita Moruz et al.

JOURNAL OF PROTEOME RESEARCH (2010)

Article Biochemical Research Methods

Semi-supervised learning for peptide identification from shotgun proteomics datasets

Lukas Kall et al.

NATURE METHODS (2007)