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

Relevance of mutation-derived neoantigens and non-classical antigens for anticancer therapies

Efficacy of cancer immunotherapies relies on correct recognition of tumor antigens by lymphocytes, eliciting thus functional responses capable of eliminating tumor cells. Therefore, important efforts have been carried out in antigen identification, with the aim of understanding mechanisms of response to immunotherapy and to design safer and more efficient strategies. In addition to classical tumor-associated antigens identified during the last decades, implementation of next-generation sequencing methodologies is enabling the identification of neoantigens (neoAgs) arising from mutations, leading to the development of new neoAg-directed therapies. Moreover, there are numerous non-classical tumor antigens originated from other sources and identified by new methodologies. Here, we review the relevance of neoAgs in different immunotherapies and the results obtained by applying neoAg-based strategies. In addition, the different types of non-classical tumor antigens and the best approaches for their identification are described. This will help to increase the spectrum of targetable molecules useful in cancer immunotherapies.

Boosting the Sensitivity of Quantitative Single-Cell Proteomics with Infrared-Tandem Mass Tags

Single-cell proteomics is a powerful approach to precisely profile protein landscapes within individual cells toward a comprehensive understanding of proteomic functions and tissue and cellular states. The inherent challenges associated with limited starting material demand heightened analytical sensitivity. Just as advances in sample preparation maximize the amount of material that makes it from the cell to the mass spectrometer, we strive to maximize the number of ions that make it from ion source to the detector. In isobaric tagging experiments, limited reporter ion generation limits quantitative accuracy and precision. The combination of infrared photoactivation and ion parking circumvents the m/z dependence inherent in HCD, maximizing reporter generation and avoiding unintended degradation of TMT reporter molecules in infrared-tandem mass tags (IR-TMT). The method was applied to single-cell human proteomes using 18-plex TMTpro, resulting in 4-5-fold increases in reporter signal compared to conventional SPS-MS3 approaches. IR-TMT enables faster duty cycles, higher throughput, and increased peptide identification and quantification. Comparative experiments showcase 4-5-fold lower injection times for IR-TMT, providing superior sensitivity without compromising accuracy. In all, IR-TMT enhances the dynamic range of proteomic experiments and is compatible with gas-phase fractionation and real-time searching, promising increased gains in the study of cellular heterogeneity.

Fit for Purpose Approach To Evaluate Detection of Amino Acid Substitutions in Shotgun Proteomics

Amino acid substitutions (AASs) alter proteins from their genome-expected sequences. Accumulation of substitutions in proteins underlies numerous diseases and antibiotic mechanisms. Accurate global detection of AASs and their frequencies is crucial for understanding these mechanisms. Shotgun proteomics provides an untargeted method for measuring AASs but introduces biases when extrapolating from the genome to identify AASs. To characterize these biases, we created a "ground-truth" approach using the similarities betweenEscherichia coli and Salmonella typhimurium to model the complexity of AAS detection. Shotgun proteomics on mixed lysates generated libraries representing ∼100,000 peptide-spectra and 4161 peptide sequences with a single AAS and defined stoichiometry. Identifying S. typhimurium peptide-spectra with only the E. coli genome resulted in 64.1% correctly identified library peptides. Specific AASs exhibit variable identification efficiencies. There was no inherent bias from the stoichiometry of the substitutions. Short peptides and AASs localized near peptide termini had poor identification efficiency. We identify a new class of "scissor substitutions" that gain or lose protease cleavage sites. Scissor substitutions also had poor identification efficiency. This ground-truth AAS library reveals various sources of bias, which will guide the application of shotgun proteomics to validate AAS hypotheses.

Toward an Ultimate Solution for Peptide Retention Time Prediction: The Effect of Column Temperature on Separation Selectivity

We studied the effect of the column temperature on the selectivity of reversed-phase peptide separation in bottom-up proteomics. The number of peptide identifications from 2 h liquid chromatography with tandem mass spectrometry (LC-MS/MS) acquisitions reaches a plateau at 45-55 °C, driven simultaneously by improved separation efficiency, a gradual decrease in peptide retention, and possible on-column degradation of peptides at elevated temperatures. Performing 2D LC-MS/MS acquisitions at 25, 35, 45, and 55 °C resulted in the identification of ∼100,000 and ∼120,000 unique peptides for nonmodified and tandem mass tags (TMT)-labeled samples, respectively. These peptide collections were used to investigate the temperature-driven retention features. The latter is governed by the specific temperature response of individual residues, peptide hydrophobicity and length, and amphipathic helicity. On average, peptide retention decreased by 0.56 and 0.5% acetonitrile for each 10 °C increase for label-free and TMT-labeled peptides, respectively. This generally linear response of retention shifts allowed the extrapolation of predictive models beyond the studied temperature range. Thus, (trap) column cooling from room temperature to 0 °C will allow the retention of an additional 3% of detectable tryptic peptides. Meanwhile, the application of 90 °C would result in the loss of ∼20% of tryptic peptides that were amenable to MS/MS-based identification.

pXg: Comprehensive Identification of Noncanonical MHC-I-Associated Peptides From De Novo Peptide Sequencing Using RNA-Seq Reads

Discovering noncanonical peptides has been a common application of proteogenomics. Recent studies suggest that certain noncanonical peptides, known as noncanonical major histocompatibility complex-I (MHC-I)-associated peptides (ncMAPs), that bind to MHC-I may make good immunotherapeutic targets. De novo peptide sequencing is a great way to find ncMAPs since it can detect peptide sequences from their tandem mass spectra without using any sequence databases. However, this strategy has not been widely applied for ncMAP identification because there is not a good way to estimate its false-positive rates. In order to completely and accurately identify immunopeptides using de novo peptide sequencing, we describe a unique pipeline called proteomics X genomics. In contrast to current pipelines, it makes use of genomic data, RNA-Seq abundance and sequencing quality, in addition to proteomic features to increase the sensitivity and specificity of peptide identification. We show that the peptide-spectrum match quality and genetic traits have a clear relationship, showing that they can be utilized to evaluate peptide-spectrum matches. From 10 samples, we found 24,449 canonical MHC-I-associated peptides and 956 ncMAPs by using a target-decoy competition. Three hundred eighty-seven ncMAPs and 1611 canonical MHC-I-associated peptides were new identifications that had not yet been published. We discovered 11 ncMAPs produced from a squirrel monkey retrovirus in human cell lines in addition to the two ncMAPs originating from a complementarity determining region 3 in an antibody thanks to the unrestricted search space assumed by de novo sequencing. These entirely new identifications show that proteomics X genomics can make the most of de novo peptide sequencing's advantages and its potential use in the search for new immunotherapeutic targets.

Intensity and retention time prediction improves the rescoring of protein-nucleic acid cross-links

In protein-RNA cross-linking mass spectrometry, UV or chemical cross-linking introduces stable bonds between amino acids and nucleic acids in protein-RNA complexes that are then analyzed and detected in mass spectra. This analytical tool delivers valuable information about RNA-protein interactions and RNA docking sites in proteins, both in vitro and in vivo. The identification of cross-linked peptides with oligonucleotides of different length leads to a combinatorial increase in search space. We demonstrate that the peptide retention time prediction tasks can be transferred to the task of cross-linked peptide retention time prediction using a simple amino acid composition encoding, yielding improved identification rates when the prediction error is included in rescoring. For the more challenging task of including fragment intensity prediction of cross-linked peptides in the rescoring, we obtain, on average, a similar improvement. Further improvement in the encoding and fine-tuning of retention time and intensity prediction models might lead to further gains, and merit further research.

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

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.

Prediction of glycopeptide fragment mass spectra by deep learning

Deep learning has achieved a notable success in mass spectrometry-based proteomics and is now emerging in glycoproteomics. While various deep learning models can predict fragment mass spectra of peptides with good accuracy, they cannot cope with the non-linear glycan structure in an intact glycopeptide. Herein, we present DeepGlyco, a deep learning-based approach for the prediction of fragment spectra of intact glycopeptides. Our model adopts tree-structured long-short term memory networks to process the glycan moiety and a graph neural network architecture to incorporate potential fragmentation pathways of a specific glycan structure. This feature is beneficial to model explainability and differentiation ability of glycan structural isomers. We further demonstrate that predicted spectral libraries can be used for data-independent acquisition glycoproteomics as a supplement for library completeness. We expect that this work will provide a valuable deep learning resource for glycoproteomics.

MS2Rescore 3.0 Is a Modular, Flexible, and User-Friendly Platform to Boost Peptide Identifications, as Showcased with MS Amanda 3.0

Rescoring of peptide-spectrum matches (PSMs) has emerged as a standard procedure for the analysis of tandem mass spectrometry data. This emphasizes the need for software maintenance and continuous improvement for such algorithms. We introduce MS2Rescore 3.0, a versatile, modular, and user-friendly platform designed to increase peptide identifications. Researchers can install MS2Rescore across various platforms with minimal effort and benefit from a graphical user interface, a modular Python API, and extensive documentation. To showcase this new version, we connected MS2Rescore 3.0 with MS Amanda 3.0, a new release of the well-established search engine, addressing previous limitations on automatic rescoring. Among new features, MS Amanda now contains additional output columns that can be used for rescoring. The full potential of rescoring is best revealed when applied on challenging data sets. We therefore evaluated the performance of these two tools on publicly available single-cell data sets, where the number of PSMs was substantially increased, thereby demonstrating that MS2Rescore offers a powerful solution to boost peptide identifications. MS2Rescore's modular design and user-friendly interface make data-driven rescoring easily accessible, even for inexperienced users. We therefore expect the MS2Rescore to be a valuable tool for the wider proteomics community. MS2Rescore is available at https://github.com/compomics/ms2rescore.

Retention time prediction for post-translationally modified peptides: Ser, Thr, Tyr-phosphorylation

The development of a peptide retention prediction model for reversed-phase chromatography applications in proteomics is reported for peptides carrying phosphorylated Ser, Thr and Tyr-residues. The major retention features have been assessed using a collection of over 10,000 phosphorylated/non-phosphorylated peptide pairs identified in a series 1D and 2D LC-MS/MS acquisitions using formic acid as ion pairing modifier. Single modification event on average results in increased peptide retention for phosphorylation of Ser (+ 1.46), Thr (+1.33), Tyr (+0.93% acetonitrile, ACN) on gradient elution scale for Luna C18(2) stationary phase. We established several composition and sequence specific features, which drive deviations from these average values. Thus, single phosphorylation of serine results in retention shifts ranging from -2.4 to 5.5% ACN depending on position of the residue, nature of nearest neighbour residues, peptide length, hydrophobicity and pI value, and its propensity to form amphipathic helical structures. We established that the altered ion-pairing environment upon phosphorylation is detrimental for this variability. Hydrophobicity of ion-pairing modifier directly informs the magnitude of expected shifts: (most hydrophilic) 0.5 % acetic acid (larger positive shift upon phosphorylation) > 0.1 % formic acid (positive) > 0.1 % trifluoroacetic (negative) > 0.1 % heptafluorobutyric acid (larger negative shift). The effect of phosphorylation has been also evaluated for several separation conditions used in the first dimension of 2D LC applications: high pH reversed-phase (RP), hydrophilic interaction liquid chromatography (HILIC), strong cation- and strong anion exchange separations.

Thunder-DDA-PASEF enables high-coverage immunopeptidomics and is boosted by MS2Rescore with MS2PIP timsTOF fragmentation prediction model

Human leukocyte antigen (HLA) class I peptide ligands (HLAIps) are key targets for developing vaccines and immunotherapies against infectious pathogens or cancer cells. Identifying HLAIps is challenging due to their high diversity, low abundance, and patient individuality. Here, we develop a highly sensitive method for identifying HLAIps using liquid chromatography-ion mobility-tandem mass spectrometry (LC-IMS-MS/MS). In addition, we train a timsTOF-specific peak intensity MS2PIP model for tryptic and non-tryptic peptides and implement it in MS2Rescore (v3) together with the CCS predictor from ionmob. The optimized method, Thunder-DDA-PASEF, semi-selectively fragments singly and multiply charged HLAIps based on their IMS and m/z. Moreover, the method employs the high sensitivity mode and extended IMS resolution with fewer MS/MS frames (300 ms TIMS ramp, 3 MS/MS frames), doubling the coverage of immunopeptidomics analyses, compared to the proteomics-tailored DDA-PASEF (100 ms TIMS ramp, 10 MS/MS frames). Additionally, rescoring boosts the HLAIps identification by 41.7% to 33%, resulting in 5738 HLAIps from as little as one million JY cell equivalents, and 14,516 HLAIps from 20 million. This enables in-depth profiling of HLAIps from diverse human cell lines and human plasma. Finally, profiling JY and Raji cells transfected to express the SARS-CoV-2 spike protein results in 16 spike HLAIps, thirteen of which have been reported to elicit immune responses in human patients.

Post-translational modifications of proteins in cardiovascular diseases examined by proteomic approaches

Over 400 different types of post-translational modifications (PTMs) have been reported and over 200 various types of PTMs have been discovered using mass spectrometry (MS)-based proteomics. MS-based proteomics has proven to be a powerful method capable of global PTM mapping with the identification of modified proteins/peptides, the localization of PTM sites and PTM quantitation. PTMs play regulatory roles in protein functions, activities and interactions in various heart related diseases, such as ischemia/reperfusion injury, cardiomyopathy and heart failure. The recognition of PTMs that are specific to cardiovascular pathology and the clarification of the mechanisms underlying these PTMs at molecular levels are crucial for discovery of novel biomarkers and application in a clinical setting. With sensitive MS instrumentation and novel biostatistical methods for precise processing of the data, low-abundance PTMs can be successfully detected and the beneficial or unfavorable effects of specific PTMs on cardiac function can be determined. Moreover, computational proteomic strategies that can predict PTM sites based on MS data have gained an increasing interest and can contribute to characterization of PTM profiles in cardiovascular disorders. More recently, machine learning- and deep learning-based methods have been employed to predict the locations of PTMs and explore PTM crosstalk. In this review article, the types of PTMs are briefly overviewed, approaches for PTM identification/quantitation in MS-based proteomics are discussed and recently published proteomic studies on PTMs associated with cardiovascular diseases are included.

COSMIC-based mutation database enhances identification efficiency of HLA-I immunopeptidome

BACKGROUND

Neoantigens have emerged as a promising area of focus in tumor immunotherapy, with several established strategies aiming to enhance their identification. Human leukocyte antigen class I molecules (HLA-I), which present intracellular immunopeptides to T cells, provide an ideal source for identifying neoantigens. However, solely relying on a mutation database generated through commonly used whole exome sequencing (WES) for the identification of HLA-I immunopeptides, may result in potential neoantigens being missed due to limitations in sequencing depth and sample quality.

METHOD

In this study, we constructed and evaluated an extended database for neoantigen identification, based on COSMIC mutation database. This study utilized mass spectrometry-based proteogenomic profiling to identify the HLA-I immunopeptidome enriched from HepG2 cell. HepG2 WES-based and the COSMIC-based mutation database were generated and utilized to identify HepG2-specific mutant immunopeptides.

RESULT

The results demonstrated that COSMIC-based database identified 5 immunopeptides compared to only 1 mutant peptide identified by HepG2 WES-based database, indicating its effectiveness in identifying mutant immunopeptides. Furthermore, HLA-I affinity of the mutant immunopeptides was evaluated through NetMHCpan and peptide-docking modeling to validate their binding to HLA-I molecules, demonstrating the potential of mutant peptides identified by the COSMIC-based database as neoantigens.

CONCLUSION

Utilizing the COSMIC-based mutation database is a more efficient strategy for identifying mutant peptides from HLA-I immunopeptidome without significantly increasing the false positive rate. HepG2 specific WES-based database may exclude certain mutant peptides due to WES sequencing depth or sample heterogeneity. The COSMIC-based database can effectively uncover potential neoantigens within the HLA-I immunopeptidomes.

Test-Time Training for Deep MS/MS Spectrum Prediction Improves Peptide Identification

In bottom-up proteomics, peptide-spectrum matching is critical for peptide and protein identification. Recently, deep learning models have been used to predict tandem mass spectra of peptides, enabling the calculation of similarity scores between the predicted and experimental spectra for peptide-spectrum matching. These models follow the supervised learning paradigm, which trains a general model using paired peptides and spectra from standard data sets and directly employs the model on experimental data. However, this approach can lead to inaccurate predictions due to differences between the training data and the experimental data, such as sample types, enzyme specificity, and instrument calibration. To tackle this problem, we developed a test-time training paradigm that adapts the pretrained model to generate experimental data-specific models, namely, PepT3. PepT3 yields a 10-40% increase in peptide identification depending on the variability in training and experimental data. Intriguingly, when applied to a patient-derived immunopeptidomic sample, PepT3 increases the identification of tumor-specific immunopeptide candidates by 60%. Two-thirds of the newly identified candidates are predicted to bind to the patient's human leukocyte antigen isoforms. To facilitate access of the model and all the results, we have archived all the intermediate files in Zenodo.org with identifier 8231084.

Acquisition and Analysis of DIA-Based Proteomic Data: A Comprehensive Survey in 2023

Data-independent acquisition (DIA) mass spectrometry (MS) has emerged as a powerful technology for high-throughput, accurate, and reproducible quantitative proteomics. This review provides a comprehensive overview of recent advances in both the experimental and computational methods for DIA proteomics, from data acquisition schemes to analysis strategies and software tools. DIA acquisition schemes are categorized based on the design of precursor isolation windows, highlighting wide-window, overlapping-window, narrow-window, scanning quadrupole-based, and parallel accumulation-serial fragmentation-enhanced DIA methods. For DIA data analysis, major strategies are classified into spectrum reconstruction, sequence-based search, library-based search, de novo sequencing, and sequencing-independent approaches. A wide array of software tools implementing these strategies are reviewed, with details on their overall workflows and scoring approaches at different steps. The generation and optimization of spectral libraries, which are critical resources for DIA analysis, are also discussed. Publicly available benchmark datasets covering global proteomics and phosphoproteomics are summarized to facilitate performance evaluation of various software tools and analysis workflows. Continued advances and synergistic developments of versatile components in DIA workflows are expected to further enhance the power of DIA-based proteomics.

OpenProt 2.0 builds a path to the functional characterization of alternative proteins

The OpenProt proteogenomic resource (https://www.openprot.org/) provides users with a complete and freely accessible set of non-canonical or alternative open reading frames (AltORFs) within the transcriptome of various species, as well as functional annotations of the corresponding protein sequences not found in standard databases. Enhancements in this update are largely the result of user feedback and include the prediction of structure, subcellular localization, and intrinsic disorder, using cutting-edge algorithms based on machine learning techniques. The mass spectrometry pipeline now integrates a machine learning-based peptide rescoring method to improve peptide identification. We continue to help users explore this cryptic proteome by providing OpenCustomDB, a tool that enables users to build their own customized protein databases, and OpenVar, a genomic annotator including genetic variants within AltORFs and protein sequences. A new interface improves the visualization of all functional annotations, including a spectral viewer and the prediction of multicoding genes. All data on OpenProt are freely available and downloadable. Overall, OpenProt continues to establish itself as an important resource for the exploration and study of new proteins.

DeepRTAlign: toward accurate retention time alignment for large cohort mass spectrometry data analysis

Retention time (RT) alignment is a crucial step in liquid chromatography-mass spectrometry (LC-MS)-based proteomic and metabolomic experiments, especially for large cohort studies. The most popular alignment tools are based on warping function method and direct matching method. However, existing tools can hardly handle monotonic and non-monotonic RT shifts simultaneously. Here, we develop a deep learning-based RT alignment tool, DeepRTAlign, for large cohort LC-MS data analysis. DeepRTAlign has been demonstrated to have improved performances by benchmarking it against current state-of-the-art approaches on multiple real-world and simulated proteomic and metabolomic datasets. The results also show that DeepRTAlign can improve identification sensitivity without compromising quantitative accuracy. Furthermore, using the MS features aligned by DeepRTAlign, we trained and validated a robust classifier to predict the early recurrence of hepatocellular carcinoma. DeepRTAlign provides an advanced solution to RT alignment in large cohort LC-MS studies, which is currently a major bottleneck in proteomics and metabolomics research.

Using long-read CAGE sequencing to profile cryptic-promoter-derived transcripts and their contribution to the immunopeptidome

Recent studies have shown that the noncoding genome can produce unannotated proteins as antigens that induce immune response. One major source of this activity is the aberrant epigenetic reactivation of transposable elements (TEs). In tumors, TEs often provide cryptic or alternate promoters, which can generate transcripts that encode tumor-specific unannotated proteins. Thus, TE-derived transcripts (TE transcripts) have the potential to produce tumor-specific, but recurrent, antigens shared among many tumors. Identification of TE-derived tumor antigens holds the promise to improve cancer immunotherapy approaches; however, current genomics and computational tools are not optimized for their detection. Here we combined CAGE technology with full-length long-read transcriptome sequencing (long-read CAGE, or LRCAGE) and developed a suite of computational tools to significantly improve immunopeptidome detection by incorporating TE and other tumor transcripts into the proteome database. By applying our methods to human lung cancer cell line H1299 data, we show that long-read technology significantly improves mapping of promoters with low mappability scores and that LRCAGE guarantees accurate construction of uncharacterized 5' transcript structure. Augmenting a reference proteome database with newly characterized transcripts enabled us to detect noncanonical antigens from HLA-pulldown LC-MS/MS data. Lastly, we show that epigenetic treatment increased the number of noncanonical antigens, particularly those encoded by TE transcripts, which might expand the pool of targetable antigens for cancers with low mutational burden.

Accurate de novo peptide sequencing using fully convolutional neural networks

De novo peptide sequencing, which does not rely on a comprehensive target sequence database, provides us with a way to identify novel peptides from tandem mass spectra. However, current de novo sequencing algorithms suffer from low accuracy and coverage, which hinders their application in proteomics. In this paper, we present PepNet, a fully convolutional neural network for high accuracy de novo peptide sequencing. PepNet takes an MS/MS spectrum (represented as a high-dimensional vector) as input, and outputs the optimal peptide sequence along with its confidence score. The PepNet model is trained using a total of 3 million high-energy collisional dissociation MS/MS spectra from multiple human peptide spectral libraries. Evaluation results show that PepNet significantly outperforms current best-performing de novo sequencing algorithms (e.g. PointNovo and DeepNovo) in both peptide-level accuracy and positional-level accuracy. PepNet can sequence a large fraction of spectra that were not identified by database search engines, and thus could be used as a complementary tool to database search engines for peptide identification in proteomics. In addition, PepNet runs around 3x and 7x faster than PointNovo and DeepNovo on GPUs, respectively, thus being more suitable for the analysis of large-scale proteomics data.

Merging Full-Spectrum and Fragment Ion Intensity Predictions from Deep Learning for High-Quality Spectral Libraries

Spectral libraries are useful resources in proteomic data analysis. Recent advances in deep learning allow tandem mass spectra of peptides to be predicted from their amino acid sequences. This enables predicted spectral libraries to be compiled, and searching against such libraries has been shown to improve the sensitivity in peptide identification over conventional sequence database searching. However, current prediction models lack support for longer peptides, and thus far, predicted library searching has only been demonstrated for backbone ion-only spectrum prediction methods. Here, we propose a deep learning-based full-spectrum prediction method to generate predicted spectral libraries for peptide identification. We demonstrated the superiority of using full-spectrum libraries over backbone ion-only prediction approaches in spectral library searching. Furthermore, merging spectra from different prediction models, as a form of ensemble learning, can produce improved spectral libraries, in terms of identification sensitivity. We also show that a hybrid library combining predicted and experimental spectra can lead to 20% more confident identifications over experimental library searching or sequence database searching.

Deep graph convolutional network for small-molecule retention time prediction

The retention time (RT) is a crucial source of data for liquid chromatography-mass spectrometry (LCMS). A model that can accurately predict the RT for each molecule would empower filtering candidates with similar spectra but differing RT in LCMS-based molecule identification. Recent research shows that graph neural networks (GNNs) outperform traditional machine learning algorithms in RT prediction. However, all of these models use relatively shallow GNNs. This study for the first time investigates how depth affects GNNs' performance on RT prediction. The results demonstrate that a notable improvement can be achieved by pushing the depth of GNNs to 16 layers by the adoption of residual connection. Additionally, we also find that graph convolutional network (GCN) model benefits from the edge information. The developed deep graph convolutional network, DeepGCN-RT, significantly outperforms the previous state-of-the-art method and achieves the lowest mean absolute percentage error (MAPE) of 3.3% and the lowest mean absolute error (MAE) of 26.55 s on the SMRT test set. We also finetune DeepGCN-RT on seven datasets with various chromatographic conditions. The mean MAE of the seven datasets largely decreases 30% compared to previous state-of-the-art method. On the RIKEN-PlaSMA dataset, we also test the effectiveness of DeepGCN-RT in assisting molecular structure identification. By 30% lessening the number of potential structures, DeepGCN-RT is able to improve top-1 accuracy by about 11%.

Sage: An Open-Source Tool for Fast Proteomics Searching and Quantification at Scale

The growing complexity and volume of proteomics data necessitate the development of efficient software tools for peptide identification and quantification from mass spectra. Given their central role in proteomics, it is imperative that these tools are auditable and extensible─requirements that are best fulfilled by open-source and permissively licensed software. This work presents Sage, a high-performance, open-source, and freely available proteomics pipeline. Scalable and cloud-ready, Sage matches the performance of state-of-the-art software tools while running an order of magnitude faster.

lesSDRF is more: maximizing the value of proteomics data through streamlined metadata annotation

Public proteomics data often lack essential metadata, limiting its potential. To address this, we present lesSDRF, a tool to simplify the process of metadata annotation, thereby ensuring that data leave a lasting, impactful legacy well beyond its initial publication.

Retention Time and Fragmentation Predictors Increase Confidence in Identification of Common Variant Peptides

Precision medicine focuses on adapting care to the individual profile of patients, for example, accounting for their unique genetic makeup. Being able to account for the effect of genetic variation on the proteome holds great promise toward this goal. However, identifying the protein products of genetic variation using mass spectrometry has proven very challenging. Here we show that the identification of variant peptides can be improved by the integration of retention time and fragmentation predictors into a unified proteogenomic pipeline. By combining these intrinsic peptide characteristics using the search-engine post-processor Percolator, we demonstrate improved discrimination power between correct and incorrect peptide-spectrum matches. Our results demonstrate that the drop in performance that is induced when expanding a protein sequence database can be compensated, hence enabling efficient identification of genetic variation products in proteomics data. We anticipate that this enhancement of proteogenomic pipelines can provide a more refined picture of the unique proteome of patients and thereby contribute to improving patient care.

Compendium of Chromatographic Behavior of Post-translationally and Chemically Modified Peptides in Bottom-Up Proteomic Experiments

We have systematically evaluated the chromatographic behavior of post-translationally/chemically modified peptides using data spanning over 70 of the most relevant modifications. These retention properties were measured for standard bottom-up proteomic settings (fully porous C18 separation media, 0.1% formic acid as ion-pairing modifier) using collections of modified/nonmodified peptide pairs. These pairs were generated by spontaneous degradation, chemical or enzymatic treatment, analysis of synthetic peptides, or the cotranslational incorporation of noncanonical proline analogues. In addition, these measurements were validated using external data acquired for synthetic peptides and enzymatically induced citrullination. Working in units of hydrophobicity index (HI, % ACN) and evaluating the average retention shifts (ΔHI) represent the simplest approach to describe the effect of modifications from a didactic point of view. Plotting HI values for modified (y-axis) vs nonmodified (x-axis) counterparts generates unique slope and intercept values for each modification defined by the chemistry of the modifying moiety: its hydrophobicity, size, pKa of ionizable groups, and position of the altered residue. These composition-dependent correlations can be used for coarse incorporation of PTMs into models for prediction of peptide retention. More accurate predictions would require the development of specific sequence-dependent algorithms to predict ΔHI values.

What Can Ribo-Seq, Immunopeptidomics, and Proteomics Tell Us About the Noncanonical Proteome?

Ribosome profiling (Ribo-Seq) has proven transformative for our understanding of the human genome and proteome by illuminating thousands of noncanonical sites of ribosome translation outside the currently annotated coding sequences (CDSs). A conservative estimate suggests that at least 7000 noncanonical ORFs are translated, which, at first glance, has the potential to expand the number of human protein CDSs by 30%, from ∼19,500 annotated CDSs to over 26,000 annotated CDSs. Yet, additional scrutiny of these ORFs has raised numerous questions about what fraction of them truly produce a protein product and what fraction of those can be understood as proteins according to conventional understanding of the term. Adding further complication is the fact that published estimates of noncanonical ORFs vary widely by around 30-fold, from several thousand to several hundred thousand. The summation of this research has left the genomics and proteomics communities both excited by the prospect of new coding regions in the human genome but searching for guidance on how to proceed. Here, we discuss the current state of noncanonical ORF research, databases, and interpretation, focusing on how to assess whether a given ORF can be said to be "protein coding."

Potential of Negative-Ion-Mode Proteomics: An MS1-Only Approach

Current proteomics approaches rely almost exclusively on using the positive ionization mode, resulting in inefficient ionization of many acidic peptides. This study investigates protein identification efficiency in the negative ionization mode using the DirectMS1 method. DirectMS1 is an ultrafast data acquisition method based on accurate peptide mass measurements and predicted retention times. Our method achieves the highest rate of protein identification in the negative ion mode to date, identifying over 1000 proteins in a human cell line at a 1% false discovery rate. This is accomplished using a single-shot 10 min separation gradient, comparable to lengthy MS/MS-based analyses. Optimizing separation and experimental conditions was achieved by utilizing mobile buffers containing 2.5 mM imidazole and 3% isopropanol. The study emphasized the complementary nature of data obtained in positive and negative ion modes. Combining the results from all replicates in both polarities increased the number of identified proteins to 1774. Additionally, we analyzed the method's efficiency using different proteases for protein digestion. Among the four studied proteases (LysC, GluC, AspN, and trypsin), trypsin and LysC demonstrated the highest protein identification yield. This suggests that digestion procedures utilized in positive-mode proteomics can be effectively applied in the negative ion mode. Data are deposited to ProteomeXchange: PXD040583.

N-terminal proteoforms may engage in different protein complexes

Alternative translation initiation and alternative splicing may give rise to N-terminal proteoforms, proteins that differ at their N-terminus compared with their canonical counterparts. Such proteoforms can have altered localizations, stabilities, and functions. Although proteoforms generated from splice variants can be engaged in different protein complexes, it remained to be studied to what extent this applies to N-terminal proteoforms. To address this, we mapped the interactomes of several pairs of N-terminal proteoforms and their canonical counterparts. First, we generated a catalogue of N-terminal proteoforms found in the HEK293T cellular cytosol from which 22 pairs were selected for interactome profiling. In addition, we provide evidence for the expression of several N-terminal proteoforms, identified in our catalogue, across different human tissues, as well as tissue-specific expression, highlighting their biological relevance. Protein-protein interaction profiling revealed that the overlap of the interactomes for both proteoforms is generally high, showing their functional relation. We also showed that N-terminal proteoforms can be engaged in new interactions and/or lose several interactions compared with their canonical counterparts, thus further expanding the functional diversity of proteomes.

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

Peptide identification in liquid chromatography-tandem mass spectrometry (LC-MS/MS) experiments relies on computational algorithms for matching acquired MS/MS spectra against sequences of candidate peptides using database search tools, such as MSFragger. Here, we present a new tool, MSBooster, for rescoring peptide-to-spectrum matches using additional features incorporating deep learning-based predictions of peptide properties, such as LC retention time, ion mobility, and MS/MS spectra. We demonstrate the utility of MSBooster, in tandem with MSFragger and Percolator, in several different workflows, including nonspecific searches (immunopeptidomics), direct identification of peptides from data independent acquisition data, single-cell proteomics, and data generated on an ion mobility separation-enabled timsTOF MS platform. MSBooster is fast, robust, and fully integrated into the widely used FragPipe computational platform.

HLA-II immunopeptidome profiling and deep learning reveal features of antigenicity to inform antigen discovery

CD4+ T cell responses are exquisitely antigen specific and directed toward peptide epitopes displayed by human leukocyte antigen class II (HLA-II) on antigen-presenting cells. Underrepresentation of diverse alleles in ligand databases and an incomplete understanding of factors affecting antigen presentation in vivo have limited progress in defining principles of peptide immunogenicity. Here, we employed monoallelic immunopeptidomics to identify 358,024 HLA-II binders, with a particular focus on HLA-DQ and HLA-DP. We uncovered peptide-binding patterns across a spectrum of binding affinities and enrichment of structural antigen features. These aspects underpinned the development of context-aware predictor of T cell antigens (CAPTAn), a deep learning model that predicts peptide antigens based on their affinity to HLA-II and full sequence of their source proteins. CAPTAn was instrumental in discovering prevalent T cell epitopes from bacteria in the human microbiome and a pan-variant epitope from SARS-CoV-2. Together CAPTAn and associated datasets present a resource for antigen discovery and the unraveling genetic associations of HLA alleles with immunopathologies.

Deep learning from harmonized peptide libraries enables retention time prediction of diverse post translational modifications

In proteomics experiments, peptide retention time (RT) is an orthogonal property to fragmentation when assessing detection confidence. Advances in deep learning enable accurate RT prediction for any peptide from sequence alone, including those yet to be experimentally observed. Here we present Chronologer, an open-source software tool for rapid and accurate peptide RT prediction. Using new approaches to harmonize and false-discovery correct across independently collected datasets, Chronologer is built on a massive database with >2.2 million peptides including 10 common post-translational modification (PTM) types. By linking knowledge learned across diverse peptide chemistries, Chronologer predicts RTs with less than two-thirds the error of other deep learning tools. We show how RT for rare PTMs, such as OGlcNAc, can be learned with high accuracy using as few as 10-100 example peptides in newly harmonized datasets. This iteratively updatable workflow enables Chronologer to comprehensively predict RTs for PTM-marked peptides across entire proteomes.

Mistle: bringing spectral library predictions to metaproteomics with an efficient search index

MOTIVATION

Deep learning has moved to the forefront of tandem mass spectrometry-driven proteomics and authentic prediction for peptide fragmentation is more feasible than ever. Still, at this point spectral prediction is mainly used to validate database search results or for confined search spaces. Fully predicted spectral libraries have not yet been efficiently adapted to large search space problems that often occur in metaproteomics or proteogenomics.

RESULTS

In this study, we showcase a workflow that uses Prosit for spectral library predictions on two common metaproteomes and implement an indexing and search algorithm, Mistle, to efficiently identify experimental mass spectra within the library. Hence, the workflow emulates a classic protein sequence database search with protein digestion but builds a searchable index from spectral predictions as an in-between step. We compare Mistle to popular search engines, both on a spectral and database search level, and provide evidence that this approach is more accurate than a database search using MSFragger. Mistle outperforms other spectral library search engines in terms of run time and proves to be extremely memory efficient with a 4- to 22-fold decrease in RAM usage. This makes Mistle universally applicable to large search spaces, e.g. covering comprehensive sequence databases of diverse microbiomes.

AVAILABILITY AND IMPLEMENTATION

Mistle is freely available on GitHub at https://github.com/BAMeScience/Mistle.

Improved LC-MS identification of short homologous peptides using sequence-specific retention time predictors

Peptides are an important group of compounds contributing to the desired, as well as the undesired taste of a food product. Their taste impressions can include aspects of sweetness, bitterness, savoury, umami and many other impressions depending on the amino acids present as well as their sequence. Identification of short peptides in foods is challenging. We developed a method to assign identities to short peptides including homologous structures, i.e. peptides containing the same amino acids with a different sequence order, by accurate prediction of the retention times during reversed phase separation. To train the method, a large set of well-defined short peptides with systematic variations in the amino acid sequence was prepared by a novel synthesis strategy called 'swapped-sequence synthesis'. Additionally, several proteins were enzymatically digested to yield short peptides. Experimental retention times were determined after reversed phase separation and peptide MS2 data was acquired using a high-resolution mass spectrometer operated in data-dependent acquisition mode (DDA). A support vector regression model was trained using a combination of existing sequence-independent peptide descriptors and a newly derived set of selected amino acid index derived sequence-specific peptide (ASP) descriptors. The model was trained and validated using the experimental retention times of the 713 small food-relevant peptides prepared. Whilst selecting the most useful ASP descriptors for our model, special attention was given to predict the retention time differences between homologous peptide structures. Inclusion of ASP descriptors greatly improved the ability to accurately predict retention times, including retention time differences between 157 homologous peptide pairs. The final prediction model had a goodness-of-fit (Q2) of 0.94; moreover for 93% of the short peptides, the elution order was correctly predicted.

What can Ribo-seq and proteomics tell us about the non-canonical proteome?

Ribosome profiling (Ribo-seq) has proven transformative for our understanding of the human genome and proteome by illuminating thousands of non-canonical sites of ribosome translation outside of the currently annotated coding sequences (CDSs). A conservative estimate suggests that at least 7,000 non-canonical open reading frames (ORFs) are translated, which, at first glance, has the potential to expand the number of human protein-coding sequences by 30%, from ∼19,500 annotated CDSs to over 26,000. Yet, additional scrutiny of these ORFs has raised numerous questions about what fraction of them truly produce a protein product and what fraction of those can be understood as proteins according to conventional understanding of the term. Adding further complication is the fact that published estimates of non-canonical ORFs vary widely by around 30-fold, from several thousand to several hundred thousand. The summation of this research has left the genomics and proteomics communities both excited by the prospect of new coding regions in the human genome, but searching for guidance on how to proceed. Here, we discuss the current state of non-canonical ORF research, databases, and interpretation, focusing on how to assess whether a given ORF can be said to be "protein-coding".

In brief

The human genome encodes thousands of non-canonical open reading frames (ORFs) in addition to protein-coding genes. As a nascent field, many questions remain regarding non-canonical ORFs. How many exist? Do they encode proteins? What level of evidence is needed for their verification? Central to these debates has been the advent of ribosome profiling (Ribo-seq) as a method to discern genome-wide ribosome occupancy, and immunopeptidomics as a method to detect peptides that are processed and presented by MHC molecules and not observed in traditional proteomics experiments. This article provides a synthesis of the current state of non-canonical ORF research and proposes standards for their future investigation and reporting.

Highlights

Combined use of Ribo-seq and proteomics-based methods enables optimal confidence in detecting non-canonical ORFs and their protein products.Ribo-seq can provide more sensitive detection of non-canonical ORFs, but data quality and analytical pipelines will impact results.Non-canonical ORF catalogs are diverse and span both high-stringency and low-stringency ORF nominations.A framework for standardized non-canonical ORF evidence will advance the research field.

Updated MS²PIP web server supports cutting-edge proteomics applications

Interest in the use of machine learning for peptide fragmentation spectrum prediction has been strongly on the rise over the past years, especially for applications in challenging proteomics identification workflows such as immunopeptidomics and the full-proteome identification of data independent acquisition spectra. Since its inception, the MS²PIP peptide spectrum predictor has been widely used for various downstream applications, mostly thanks to its accuracy, ease-of-use, and broad applicability. We here present a thoroughly updated version of the MS²PIP web server, which includes new and more performant prediction models for both tryptic- and non-tryptic peptides, for immunopeptides, and for CID-fragmented TMT-labeled peptides. Additionally, we have also added new functionality to greatly facilitate the generation of proteome-wide predicted spectral libraries, requiring only a FASTA protein file as input. These libraries also include retention time predictions from DeepLC. Moreover, we now provide pre-built and ready-to-download spectral libraries for various model organisms in multiple DIA-compatible spectral library formats. Besides upgrading the back-end models, the user experience on the MS²PIP web server is thus also greatly enhanced, extending its applicability to new domains, including immunopeptidomics and MS3-based TMT quantification experiments. MS²PIP is freely available at https://iomics.ugent.be/ms2pip/.

The peptide woods are lovely, dark and deep: Hunting for novel cancer antigens

Harnessing the patient's immune system to control a tumor is a proven avenue for cancer therapy. T cell therapies as well as therapeutic vaccines, which target specific antigens of interest, are being explored as treatments in conjunction with immune checkpoint blockade. For these therapies, selecting the best suited antigens is crucial. Most of the focus has thus far been on neoantigens that arise from tumor-specific somatic mutations. Although there is clear evidence that T-cell responses against mutated neoantigens are protective, the large majority of these mutations are not immunogenic. In addition, most somatic mutations are unique to each individual patient and their targeting requires the development of individualized approaches. Therefore, novel antigen types are needed to broaden the scope of such treatments. We review high throughput approaches for discovering novel tumor antigens and some of the key challenges associated with their detection, and discuss considerations when selecting tumor antigens to target in the clinic.

A high-throughput approach reveals distinct peptide charging behaviors in electrospray ionization mass spectrometry

Electrospray ionization is a powerful and prevalent technique used to ionize analytes in mass spectrometry. The distribution of charges that an analyte receives (charge state distribution, CSD) is an important consideration for interpreting mass spectra. However, due to an incomplete understanding of the ionization mechanism, the analyte properties that influence CSDs are not fully understood. Here, we employ a machine learning-based high-throughput approach and analyze CSDs of hundreds of thousands of peptides. Interestingly, half of the peptides exhibit charges that differ from what one would naively expect (number of basic sites). We find that these peptides can be classified into two regimes-undercharging and overcharging-and that these two regimes display markedly different charging characteristics. Strikingly, peptides in the overcharging regime show minimal dependence on basic site count, and more generally, the two regimes exhibit distinct sequence determinants. These findings highlight the rich ionization behavior of peptides and the potential of CSDs for enhancing peptide identification.

Workflow enabling deepscale immunopeptidome, proteome, ubiquitylome, phosphoproteome, and acetylome analyses of sample-limited tissues

Serial multi-omic analysis of proteome, phosphoproteome, and acetylome provides insights into changes in protein expression, cell signaling, cross-talk and epigenetic pathways involved in disease pathology and treatment. However, ubiquitylome and HLA peptidome data collection used to understand protein degradation and antigen presentation have not together been serialized, and instead require separate samples for parallel processing using distinct protocols. Here we present MONTE, a highly sensitive multi-omic native tissue enrichment workflow, that enables serial, deep-scale analysis of HLA-I and HLA-II immunopeptidome, ubiquitylome, proteome, phosphoproteome, and acetylome from the same tissue sample. We demonstrate that the depth of coverage and quantitative precision of each 'ome is not compromised by serialization, and the addition of HLA immunopeptidomics enables the identification of peptides derived from cancer/testis antigens and patient specific neoantigens. We evaluate the technical feasibility of the MONTE workflow using a small cohort of patient lung adenocarcinoma tumors.

Recent advances in kinase signaling network profiling by mass spectrometry

Mass spectrometry-based phosphoproteomics is currently the leading methodology for the study of global kinase signaling. The scientific community is continuously releasing technological improvements for sensitive and fast identification of phosphopeptides, and their accurate quantification. To interpret large-scale phosphoproteomics data, numerous bioinformatic resources are available that help understanding kinase network functional role in biological systems upon perturbation. Some of these resources are databases of phosphorylation sites, protein kinases and phosphatases; others are bioinformatic algorithms to infer kinase activity, predict phosphosite functional relevance and visualize kinase signaling networks. In this review, we present the latest experimental and bioinformatic tools to profile protein kinase signaling networks and provide examples of their application in biomedicine.

Comparison of Database Searching Programs for the Analysis of Single-Cell Proteomics Data

Single-cell proteomics is emerging as an important subfield in the proteomics and mass spectrometry communities, with potential to reshape our understanding of cell development, cell differentiation, disease diagnosis, and the development of new therapies. Compared with significant advancements in the "hardware" that is used in single-cell proteomics, there has been little work comparing the effects of using different "software" packages to analyze single-cell proteomics datasets. To this end, seven popular proteomics programs were compared here, applying them to search three single-cell proteomics datasets generated by three different platforms. The results suggest that MSGF+, MSFragger, and Proteome Discoverer are generally more efficient in maximizing protein identifications, that MaxQuant is better suited for the identification of low-abundance proteins, that MSFragger is superior in elucidating peptide modifications, and that Mascot and X!Tandem are better for analyzing long peptides. Furthermore, an experiment with different loading amounts was carried out to investigate changes in identification results and to explore areas in which single-cell proteomics data analysis may be improved in the future. We propose that this comparative study may provide insight for experts and beginners alike operating in the emerging subfield of single-cell proteomics.

Toward an Integrated Machine Learning Model of a Proteomics Experiment

In recent years machine learning has made extensive progress in modeling many aspects of mass spectrometry data. We brought together proteomics data generators, repository managers, and machine learning experts in a workshop with the goals to evaluate and explore machine learning applications for realistic modeling of data from multidimensional mass spectrometry-based proteomics analysis of any sample or organism. Following this sample-to-data roadmap helped identify knowledge gaps and define needs. Being able to generate bespoke and realistic synthetic data has legitimate and important uses in system suitability, method development, and algorithm benchmarking, while also posing critical ethical questions. The interdisciplinary nature of the workshop informed discussions of what is currently possible and future opportunities and challenges. In the following perspective we summarize these discussions in the hope of conveying our excitement about the potential of machine learning in proteomics and to inspire future research.

Antigen discovery for the development of cancer immunotherapy

Central to successful cancer immunotherapy is effective T cell antitumor immunity. Multiple targeted immunotherapies engineered to invigorate T cell-driven antitumor immunity rely on identifying the repertoire of T cell antigens expressed on the tumor cell surface. Mass spectrometry-based survey of such antigens ("immunopeptidomics") combined with other omics platforms and computational algorithms has been instrumental in identifying and quantifying tumor-derived T cell antigens. In this review, we discuss the types of tumor antigens that have emerged for targeted cancer immunotherapy and the immunopeptidomics methods that are central in MHC peptide identification and quantification. We provide an overview of the strength and limitations of mass spectrometry-driven approaches and how they have been integrated with other technologies to discover targetable T cell antigens for cancer immunotherapy. We highlight some of the emerging cancer immunotherapies that successfully capitalized on immunopeptidomics, their challenges, and mass spectrometry-based strategies that can support their development.

Noncanonical splicing junctions between exons and transposable elements represent a source of immunogenic recurrent neo-antigens in patients with lung cancer

Although most characterized tumor antigens are encoded by canonical transcripts (such as differentiation or tumor-testis antigens) or mutations (both driver and passenger mutations), recent results have shown that noncanonical transcripts including long noncoding RNAs and transposable elements (TEs) can also encode tumor-specific neo-antigens. Here, we investigate the presentation and immunogenicity of tumor antigens derived from noncanonical mRNA splicing events between coding exons and TEs. Comparing human non-small cell lung cancer (NSCLC) and diverse healthy tissues, we identified a subset of splicing junctions that is both tumor specific and shared across patients. We used HLA-I peptidomics to identify peptides encoded by tumor-specific junctions in primary NSCLC samples and lung tumor cell lines. Recurrent junction-encoded peptides were immunogenic in vitro, and CD8+ T cells specific for junction-encoded epitopes were present in tumors and tumor-draining lymph nodes from patients with NSCLC. We conclude that noncanonical splicing junctions between exons and TEs represent a source of recurrent, immunogenic tumor-specific antigens in patients with NSCLC.

Exploring the variable space of shallow machine learning models for reversed-phase retention time prediction

Peptide retention time (RT) prediction algorithms are tools to study and identify the physicochemical properties that drive the peptide-sorbent interaction. Traditional RT algorithms use multiple linear regression with manually curated parameters to determine the degree of direct contribution for each parameter and improvements to RT prediction accuracies relied on superior feature engineering. Deep learning led to a significant increase in RT prediction accuracy and automated feature engineering via chaining multiple learning modules. However, the significance and the identity of these extracted variables are not well understood due to the inherent complexity when interpreting "relationships-of-relationships" found in deep learning variables. To achieve both accuracy and interpretability simultaneously, we isolated individual modules used in deep learning and the isolated modules are the shallow learners employed for RT prediction in this work. Using a shallow convolutional neural network (CNN) and gated recurrent unit (GRU), we find that the spatial features obtained via the CNN correlate with real-world physicochemical properties namely cross-collisional sections (CCS) and variations of assessable surface area (ASA). Furthermore, we determined that the discovered parameters are "micro-coefficients" that contribute to the "macro-coefficient" - hydrophobicity. Manually embedding CCS and the variations of ASA to the GRU model yielded an R2 = 0.981 using only 525 variables and can represent 88% of the ∼110,000 tryptic peptides used in our dataset. This work highlights the feature discovery process of our shallow learners can achieve beyond traditional RT models in performance and have better interpretability when compared with the deep learning RT algorithms found in the literature.

Identification of Alternative Splicing in Proteomes of Human Melanoma Cell Lines without RNA Sequencing Data

Alternative splicing is one of the main regulation pathways in living cells beyond simple changes in the level of protein expression. Most of the approaches proposed in proteomics for the identification of specific splicing isoforms require a preliminary deep transcriptomic analysis of the sample under study, which is not always available, especially in the case of the re-analysis of previously acquired data. Herein, we developed new algorithms for the identification and validation of protein splice isoforms in proteomic data in the absence of RNA sequencing of the samples under study. The bioinformatic approaches were tested on the results of proteome analysis of human melanoma cell lines, obtained earlier by high-resolution liquid chromatography and mass spectrometry (LC-MS). A search for alternative splicing events for each of the cell lines studied was performed against the database generated from all known transcripts (RefSeq) and the one composed of peptide sequences, which included all biologically possible combinations of exons. The identifications were filtered using the prediction of both retention times and relative intensities of fragment ions in the corresponding mass spectra. The fragmentation mass spectra corresponding to the discovered alternative splicing events were additionally examined for artifacts. Selected splicing events were further validated at the mRNA level by quantitative PCR.

ProteomicsML: An Online Platform for Community-Curated Data sets and Tutorials for Machine Learning in Proteomics

Data set acquisition and curation are often the most difficult and time-consuming parts of a machine learning endeavor. This is especially true for proteomics-based liquid chromatography (LC) coupled to mass spectrometry (MS) data sets, due to the high levels of data reduction that occur between raw data and machine learning-ready data. Since predictive proteomics is an emerging field, when predicting peptide behavior in LC-MS setups, each lab often uses unique and complex data processing pipelines in order to maximize performance, at the cost of accessibility and reproducibility. For this reason we introduce ProteomicsML, an online resource for proteomics-based data sets and tutorials across most of the currently explored physicochemical peptide properties. This community-driven resource makes it simple to access data in easy-to-process formats, and contains easy-to-follow tutorials that allow new users to interact with even the most advanced algorithms in the field. ProteomicsML provides data sets that are useful for comparing state-of-the-art machine learning algorithms, as well as providing introductory material for teachers and newcomers to the field alike. The platform is freely available at https://www.proteomicsml.org/, and we welcome the entire proteomics community to contribute to the project at https://github.com/ProteomicsML/ProteomicsML.

Prediction of peptide mass spectral libraries with machine learning

The recent development of machine learning methods to identify peptides in complex mass spectrometric data constitutes a major breakthrough in proteomics. Longstanding methods for peptide identification, such as search engines and experimental spectral libraries, are being superseded by deep learning models that allow the fragmentation spectra of peptides to be predicted from their amino acid sequence. These new approaches, including recurrent neural networks and convolutional neural networks, use predicted in silico spectral libraries rather than experimental libraries to achieve higher sensitivity and/or specificity in the analysis of proteomics data. Machine learning is galvanizing applications that involve large search spaces, such as immunopeptidomics and proteogenomics. Current challenges in the field include the prediction of spectra for peptides with post-translational modifications and for cross-linked pairs of peptides. Permeation of machine-learning-based spectral prediction into search engines and spectrum-centric data-independent acquisition workflows for diverse peptide classes and measurement conditions will continue to push sensitivity and dynamic range in proteomics applications in the coming years.