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

Rescoring Peptide Spectrum Matches: Boosting Proteomics Performance by Integrating Peptide Property Predictors Into Peptide Identification

Rescoring of peptide spectrum matches originating from database search engines enabled by peptide property predictors is exceeding the performance of peptide identification from traditional database search engines. In contrast to the peptide spectrum match scores calculated by traditional database search engines, rescoring peptide spectrum matches generates scores based on comparing observed and predicted peptide properties, such as fragment ion intensities and retention times. These newly generated scores enable a more efficient discrimination between correct and incorrect peptide spectrum matches. This approach was shown to lead to substantial improvements in the number of confidently identified peptides, facilitating the analysis of challenging datasets in various fields such as immunopeptidomics, metaproteomics, proteogenomics, and single-cell proteomics. In this review, we summarize the key elements leading up to the recent introduction of multiple data-driven rescoring pipelines. We provide an overview of relevant post-processing rescoring tools, introduce prominent data-driven rescoring pipelines for various applications, and highlight limitations, opportunities, and future perspectives of this approach and its impact on mass spectrometry-based proteomics.

ProPept-MT: A Multi-Task Learning Model for Peptide Feature Prediction

In the realm of quantitative proteomics, data-independent acquisition (DIA) has emerged as a promising approach, offering enhanced reproducibility and quantitative accuracy compared to traditional data-dependent acquisition (DDA) methods. However, the analysis of DIA data is currently hindered by its reliance on project-specific spectral libraries derived from DDA analyses, which not only limits proteome coverage but also proves to be a time-intensive process. To overcome these challenges, we propose ProPept-MT, a novel deep learning-based multi-task prediction model designed to accurately forecast key features such as retention time (RT), ion intensity, and ion mobility (IM). Leveraging advanced techniques such as multi-head attention and BiLSTM for feature extraction, coupled with Nash-MTL for gradient coordination, ProPept-MT demonstrates superior prediction performance. Integrating ion mobility alongside RT, mass-to-charge ratio (m/z), and ion intensity forms 4D proteomics. Then, we outline a comprehensive workflow tailored for 4D DIA proteomics research, integrating the use of 4D in silico libraries predicted by ProPept-MT. Evaluation on a benchmark dataset showcases ProPept-MT's exceptional predictive capabilities, with impressive results including a 99.9% Pearson correlation coefficient (PCC) for RT prediction, a median dot product (DP) of 96.0% for fragment ion intensity prediction, and a 99.3% PCC for IM prediction on the test set. Notably, ProPept-MT manifests efficacy in predicting both unmodified and phosphorylated peptides, underscoring its potential as a valuable tool for constructing high-quality 4D DIA in silico libraries.

Systematic Assessment of Deep Learning-Based Predictors of Fragmentation Intensity Profiles

In recent years, several deep learning-based methods have been proposed for predicting peptide fragment intensities. This study aims to provide a comprehensive assessment of six such methods, namely Prosit, DeepMass:Prism, pDeep3, AlphaPeptDeep, Prosit Transformer, and the method proposed by Guan et al. To this end, we evaluated the accuracy of the predicted intensity profiles for close to 1.7 million precursors (including both tryptic and HLA peptides) corresponding to more than 18 million experimental spectra procured from 40 independent submissions to the PRIDE repository that were acquired for different species using a variety of instruments and different dissociation types/energies. Specifically, for each method, distributions of similarity (measured by Pearson's correlation and normalized angle) between the predicted and the corresponding experimental b and y fragment intensities were generated. These distributions were used to ascertain the prediction accuracy and rank the prediction methods for particular types of experimental conditions. The effect of variables like precursor charge, length, and collision energy on the prediction accuracy was also investigated. In addition to prediction accuracy, the methods were evaluated in terms of prediction speed. The systematic assessment of these six methods may help in choosing the right method for MS/MS spectra prediction for particular needs.

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.

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.

Mapping protein states and interactions across the tree of life with co-fractionation mass spectrometry

We present CFdb, a harmonized resource of interaction proteomics data from 411 co-fractionation mass spectrometry (CF-MS) datasets spanning 21,703 fractions. Meta-analysis of this resource charts protein abundance, phosphorylation, and interactions throughout the tree of life, including a reference map of the human interactome. We show how large-scale CF-MS data can enhance analyses of individual CF-MS datasets, and exemplify this strategy by mapping the honey bee interactome.

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.

Algorithms for de-novo sequencing of peptides by tandem mass spectrometry: A review

Peptide sequencing is of great significance to fundamental and applied research in the fields such as chemical, biological, medicinal and pharmaceutical sciences. With the rapid development of mass spectrometry and sequencing algorithms, de-novo peptide sequencing using tandem mass spectrometry (MS/MS) has become the main method for determining amino acid sequences of novel and unknown peptides. Advanced algorithms allow the amino acid sequence information to be accurately obtained from MS/MS spectra in short time. In this review, algorithms from exhaustive search to the state-of-art machine learning and neural network for high-throughput and automated de-novo sequencing are introduced and compared. Impacts of datasets on algorithm performance are highlighted. The current limitations and promising direction of de-novo peptide sequencing are also discussed in this review.

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.

Analysis of DIA proteomics data using MSFragger-DIA and FragPipe computational platform

Liquid chromatography (LC) coupled with data-independent acquisition (DIA) mass spectrometry (MS) has been increasingly used in quantitative proteomics studies. Here, we present a fast and sensitive approach for direct peptide identification from DIA data, MSFragger-DIA, which leverages the unmatched speed of the fragment ion indexing-based search engine MSFragger. Different from most existing methods, MSFragger-DIA conducts a database search of the DIA tandem mass (MS/MS) spectra prior to spectral feature detection and peak tracing across the LC dimension. To streamline the analysis of DIA data and enable easy reproducibility, we integrate MSFragger-DIA into the FragPipe computational platform for seamless support of peptide identification and spectral library building from DIA, data-dependent acquisition (DDA), or both data types combined. We compare MSFragger-DIA with other DIA tools, such as DIA-Umpire based workflow in FragPipe, Spectronaut, DIA-NN library-free, and MaxDIA. We demonstrate the fast, sensitive, and accurate performance of MSFragger-DIA across a variety of sample types and data acquisition schemes, including single-cell proteomics, phosphoproteomics, and large-scale tumor proteome profiling studies.

AIomics: Exploring More of the Proteome Using Mass Spectral Libraries Extended by Artificial Intelligence

The unbounded permutations of biological molecules, including proteins and their constituent peptides, present a dilemma in identifying the components of complex biosamples. Sequence search algorithms used to identify peptide spectra can be expanded to cover larger classes of molecules, including more modifications, isoforms, and atypical cleavage, but at the cost of false positives or false negatives due to the simplified spectra they compute from sequence records. Spectral library searching can help solve this issue by precisely matching experimental spectra to library spectra with excellent sensitivity and specificity. However, compiling spectral libraries that span entire proteomes is pragmatically difficult. Neural networks that predict complete spectra containing a full range of annotated and unannotated ions can be used to replace these simplified spectra with libraries of fully predicted spectra, including modified peptides. Using such a network, we created predicted spectral libraries that were used to rescore matches from a sequence search done over a large search space, including a large number of modifications. Rescoring improved the separation of true and false hits by 82%, yielding an 8% increase in peptide identifications, including a 21% increase in nonspecifically cleaved peptides and a 17% increase in phosphopeptides.

Artificial intelligence reinforced upconversion nanoparticle-based lateral flow assay via transfer learning

The combination of upconverting nanoparticles (UCNPs) and immunochromatography has become a widely used and promising new detection technique for point-of-care testing (POCT). However, their low luminescence efficiency, non-specific adsorption, and image noise have always limited their progress toward practical applications. Recently, artificial intelligence (AI) has demonstrated powerful representational learning and generalization capabilities in computer vision. We report for the first time a combination of AI and upconversion nanoparticle-based lateral flow assays (UCNP-LFAs) for the quantitative detection of commercial internet of things (IoT) devices. This universal UCNPs quantitative detection strategy combines high accuracy, sensitivity, and applicability in the field detection environment. By using transfer learning to train AI models in a small self-built database, we not only significantly improved the accuracy and robustness of quantitative detection, but also efficiently solved the actual problems of data scarcity and low computing power of POCT equipment. Then, the trained AI model was deployed in IoT devices, whereby the detection process does not require detailed data preprocessing to achieve real-time inference of quantitative results. We validated the quantitative detection of two detectors using eight transfer learning models on a small dataset. The AI quickly provided ultra-high accuracy prediction results (some models could reach 100% accuracy) even when strong noise was added. Simultaneously, the high flexibility of this strategy promises to be a general quantitative detection method for optical biosensors. We believe that this strategy and device have a scientific significance in revolutionizing the existing POCT technology landscape and providing excellent commercial value in the in vitro diagnostics (IVD) industry.

DeepFLR facilitates false localization rate control in phosphoproteomics

Protein phosphorylation is a post-translational modification crucial for many cellular processes and protein functions. Accurate identification and quantification of protein phosphosites at the proteome-wide level are challenging, not least because efficient tools for protein phosphosite false localization rate (FLR) control are lacking. Here, we propose DeepFLR, a deep learning-based framework for controlling the FLR in phosphoproteomics. DeepFLR includes a phosphopeptide tandem mass spectrum (MS/MS) prediction module based on deep learning and an FLR assessment module based on a target-decoy approach. DeepFLR improves the accuracy of phosphopeptide MS/MS prediction compared to existing tools. Furthermore, DeepFLR estimates FLR accurately for both synthetic and biological datasets, and localizes more phosphosites than probability-based methods. DeepFLR is compatible with data from different organisms, instruments types, and both data-dependent and data-independent acquisition approaches, thus enabling FLR estimation for a broad range of phosphoproteomics experiments.

Synchro-PASEF Allows Precursor-Specific Fragment Ion Extraction and Interference Removal in Data-Independent Acquisition

Data-independent acquisition (DIA) methods have become increasingly popular in mass spectrometry-based proteomics because they enable continuous acquisition of fragment spectra for all precursors simultaneously. However, these advantages come with the challenge of correctly reconstructing the precursor-fragment relationships in these highly convoluted spectra for reliable identification and quantification. Here, we introduce a scan mode for the combination of trapped ion mobility spectrometry with parallel accumulation-serial fragmentation (PASEF) that seamlessly and continuously follows the natural shape of the ion cloud in ion mobility and peptide precursor mass dimensions. Termed synchro-PASEF, it increases the detected fragment ion current several-fold at sub-second cycle times. Consecutive quadrupole selection windows move synchronously through the mass and ion mobility range. In this process, the quadrupole slices through the peptide precursors, which separates fragment ion signals of each precursor into adjacent synchro-PASEF scans. This precisely defines precursor-fragment relationships in ion mobility and mass dimensions and effectively deconvolutes the DIA fragment space. Importantly, the partitioned parts of the fragment ion transitions provide a further dimension of specificity via a lock-and-key mechanism. This is also advantageous for quantification, where signals from interfering precursors in the DIA selection window do not affect all partitions of the fragment ion, allowing to retain only the specific parts for quantification. Overall, we establish the defining features of synchro-PASEF and explore its potential for proteomic analyses.

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.

An Optimized Proteomics Approach Reveals Novel Alternative Proteins in Mouse Liver Development

Alternative ORFs (AltORFs) are unannotated sequences in genome that encode novel peptides or proteins named alternative proteins (AltProts). Although ribosome profiling and bioinformatics predict a large number of AltProts, mass spectrometry as the only direct way of identification is hampered by the short lengths and relative low abundance of AltProts. There is an urgent need for improvement of mass spectrometry methodologies for AltProt identification. Here, we report an approach based on size-exclusion chromatography for simultaneous enrichment and fractionation of AltProts from complex proteome. This method greatly simplifies the variance of AltProts discovery by enriching small proteins smaller than 40 kDa. In a systematic comparison between 10 methods, the approach we reported enabled the discovery of more AltProts with overall higher intensities, with less cost of time and effort compared to other workflows. We applied this approach to identify 89 novel AltProts from mouse liver, 39 of which were differentially expressed between embryonic and adult mice. During embryonic development, the upregulated AltProts were mainly involved in biological pathways on RNA splicing and processing, whereas the AltProts involved in metabolisms were more active in adult livers. Our study not only provides an effective approach for identifying AltProts but also novel AltProts that are potentially important in developmental biology.

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

Machine learning and in particular deep learning (DL) are increasingly important in mass spectrometry (MS)-based proteomics. Recent DL models can predict the retention time, ion mobility and fragment intensities of a peptide just from the amino acid sequence with good accuracy. However, DL is a very rapidly developing field with new neural network architectures frequently appearing, which are challenging to incorporate for proteomics researchers. Here we introduce AlphaPeptDeep, a modular Python framework built on the PyTorch DL library that learns and predicts the properties of peptides ( https://github.com/MannLabs/alphapeptdeep ). It features a model shop that enables non-specialists to create models in just a few lines of code. AlphaPeptDeep represents post-translational modifications in a generic manner, even if only the chemical composition is known. Extensive use of transfer learning obviates the need for large data sets to refine models for particular experimental conditions. The AlphaPeptDeep models for predicting retention time, collisional cross sections and fragment intensities are at least on par with existing tools. Additional sequence-based properties can also be predicted by AlphaPeptDeep, as demonstrated with a HLA peptide prediction model to improve HLA peptide identification for data-independent acquisition ( https://github.com/MannLabs/PeptDeep-HLA ).

Data-independent acquisition proteomics methods for analyzing post-translational modifications

Protein post-translational modifications (PTMs) increase the functional diversity of the cellular proteome. Accurate and high throughput identification and quantification of protein PTMs is a key task in proteomics research. Recent advancements in data-independent acquisition (DIA) mass spectrometry (MS) technology have achieved deep coverage and accurate quantification of proteins and PTMs. This review provides an overview of DIA data processing methods that cover three aspects of PTMs analysis, i.e., detection of PTMs, site localization, and characterization of complex modification moieties, such as glycosylation. In addition, a survey of deep learning methods that boost DIA-based PTMs analysis is presented, including in silico spectral library generation, as well as feature scoring and error rate control. The limitations and future directions of DIA methods for PTMs analysis are also discussed. Novel data analysis methods will take advantage of advanced MS instrumentation techniques to empower DIA MS for in-depth and accurate PTMs measurements. This article is protected by copyright. All rights reserved.

Easy and Rapid Approach to Obtaining the Binding Affinity of Biomolecular Interactions Based on the Deep Learning Boost

Recently, the deep learning (DL) dimension of artificial intelligence has received much attention from biochemical researchers and thus has gradually become the key approach adopted in the area of biosensing applications. Studies have shown that the use of DL techniques for sensing can not only shorten the time of data analysis but also significantly increase the accuracy of data analysis and prediction, resulting in the performance improvement of biosensing systems in comparison to conventional methods. However, obtaining reliable equilibrium and rate constants of biomolecular interactions during the detection process remains difficult and time-consuming to date. In this study, we propose a transformed model based on the deep transfer learning and sequence-to-sequence autoencoder that can successfully transfer the SPR sensorgram to the protein-binding constants, that is, the association rate constant (k_a) and dissociation rate constant (k_d), which provide crucial information to understand the mechanisms of drug action and the functional structures of biomolecules. Experimentally, we first trained and tested the pre-trained model using the Langmuir model which generated ideal SPR sensorgrams and then we fine-tuned the pre-trained model through the augmented SPR sensorgrams which were synthesized by using the synthesized minority oversampling technique (SMOTE) through the moderate-scale experiment. Next, the fine-tuned model was inputted with a short experimental SPR sensorgram that only needs 110 s, and the sensorgram was directly transformed into a reconstructed ideal sensorgram. Finally, the binding kinetic constants, that is, k_a and k_d, as outputs, were obtained through fitting the reconstructed ideal sensorgram. The results showed that the prediction errors of k_a and k_d obtained by our model were less than 12 and 24%, respectively. Based on the convenience, accuracy, and reliability of the proposed DL approach, we believe our strategy significantly boosts the feasibility to monitor the binding affinity of antibodies online during production.

Rational Discovery of Antimicrobial Peptides by Means of Artificial Intelligence

Antibiotic resistance is a worldwide public health problem due to the costs and mortality rates it generates. However, the large pharmaceutical industries have stopped searching for new antibiotics because of their low profitability, given the rapid replacement rates imposed by the increasingly observed resistance acquired by microorganisms. Alternatively, antimicrobial peptides (AMPs) have emerged as potent molecules with a much lower rate of resistance generation. The discovery of these peptides is carried out through extensive in vitro screenings of either rational or non-rational libraries. These processes are tedious and expensive and generate only a few AMP candidates, most of which fail to show the required activity and physicochemical properties for practical applications. This work proposes implementing an artificial intelligence algorithm to reduce the required experimentation and increase the efficiency of high-activity AMP discovery. Our deep learning (DL) model, called AMPs-Net, outperforms the state-of-the-art method by 8.8% in average precision. Furthermore, it is highly accurate to predict the antibacterial and antiviral capacity of a large number of AMPs. Our search led to identifying two unreported antimicrobial motifs and two novel antimicrobial peptides related to them. Moreover, by coupling DL with molecular dynamics (MD) simulations, we were able to find a multifunctional peptide with promising therapeutic effects. Our work validates our previously proposed pipeline for a more efficient rational discovery of novel AMPs.

Accurate Prediction of y Ions in Beam-Type Collision-Induced Dissociation Using Deep Learning

Peptide fragmentation spectra contain critical information for the identification of peptides by mass spectrometry. In this study, we developed an algorithm that more accurately predicts the high-intensity peaks among the peptide spectra. The training data are composed of 180,833 peptides from the National Institute of Standards and Technology and Proteomics Identification database, which were fragmented by either quadrupole time-of-flight or triple-quadrupole collision-induced dissociation methods. Exploratory analysis of the peptide fragmentation pattern was focused on the highest intensity peaks that showed proline, peptide length, and a sliding window of four amino acid combination that can be exploited as key features. The amino acid sequence of each peptide and each of the key features were allocated to different layers of the model, where recurrent neural network, convolutional neural network, and fully connected neural network were used. The trained model, PrAI-frag, accurately predicts the fragmentation spectra compared to previous machine learning-based prediction algorithms. The model excels at high-intensity peak prediction, which is advantageous to selective/multiple reaction monitoring application. PrAI-frag is provided via a Web server which can be used for peptides of length 6-15.

Deephos: Predicted spectral database search for TMT-labeled phosphopeptides and its false discovery rate estimation

MOTIVATION

Tandem mass tag (TMT)-based tandem mass spectrometry (MS/MS) has become the method of choice for the quantification of post-translational modifications in complex mixtures. Many cancer proteogenomic studies have highlighted the importance of large-scale phosphopeptide quantification coupled with TMT labeling. Herein, we propose a predicted Spectral DataBase (pSDB) search strategy called Deephos that can improve both sensitivity and specificity in identifying MS/MS spectra of TMT-labeled phosphopeptides.

RESULTS

With deep learning-based fragment ion prediction, we compiled a pSDB of TMT-labeled phosphopeptides generated from ∼8,000 human phosphoproteins annotated in UniProt. Deep learning could successfully recognize the fragmentation patterns altered by both TMT labeling and phosphorylation. In addition, we discuss the decoy spectra for false discovery rate (FDR) estimation in the pSDB search. We show that FDR could be inaccurately estimated by the existing decoy spectra generation methods and propose an innovative method to generate decoy spectra for more accurate FDR estimation. The utilities of Deephos were demonstrated in multi-stage analyses (coupled with database searches) of glioblastoma, acute myeloid leukemia, and breast cancer phosphoproteomes.

AVAILABILITY

Deephos pSDB and the search software are available at https://github.com/seungjinna/deephos.

pValid 2: A deep learning based validation method for peptide identification in shotgun proteomics with increased discriminating power

Tandem mass spectrometry has been the principal method in shotgun proteomics for peptide and protein identification. However, incorrect identifications reported by proteome search engines are still unknown, and further validation methods are needed. We have proposed a validation method pValid before, but its scope of application is limited because two features used in pValid are related to open database search and sub-optimal peptide candidates for tandem mass spectra, and the performance on complex datasets still has room for improvement. In this study, we developed a more comprehensive validation method, pValid 2, to break these limitations by removing the two features and bringing in a new feature related to the retention time predicted by a deep learning-based method pPredRT. pValid 2 yielded an average false positive rate of 0.03% and an average false negative rate of 1.37% on three testing datasets, better than those of pValid, and flagged 8.47% to 11.31% more incorrect identifications than pValid on two complex datasets. Moreover, pValid 2 flagged almost all decoy identifications in validating the open-search datasets. In addition, the function of validating identifications given by MaxQuant and MS-GF+ was implemented in pValid 2, and the validation results showed that pValid 2 performed dramatically better than three metabolic labeling validation methods. Further considering its cost-effectiveness as a pure computational approach, pValid 2 has the potential to be a widely used validation tool for peptide identifications of any proteome search engines in shotgun proteomics. SIGNIFICANCE: Identification results given by shotgun proteomics are vital to life science research. The correctness of identifications deeply affects the precision of the subsequent studies about protein structures and functions, protein-protein interactions, pathogenic mechanism, and targeted drugs. Thus, validating the correctness of identifications is crucial and urgent. In 2019, we developed an identification credibility validation method named pValid, whose false positive rate (FPR) is 0.03% and false negative rate (FNR) is 1.79%, comparable to those of the gold standard, i.e., the Synthetic-peptide validation method. However, pValid can only be used for validating the results from pFind, and its validation performance on a few complex datasets still has room for improvement. So, in this submission, we proposed pValid 2, a more comprehensive computational validation method that can validate identifications from any proteome search engines with increased discriminating power.

MS2AI: automated repurposing of public peptide LC-MS data for machine learning applications

MOTIVATION

Liquid-chromatography mass-spectrometry (LC-MS) is the established standard for analyzing the proteome in biological samples by identification and quantification of thousands of proteins. Machine learning (ML) promises to considerably improve the analysis of the resulting data, however, there is yet to be any tool that mediates the path from raw data to modern ML applications. More specifically, ML applications are currently hampered by three major limitations: (i) absence of balanced training data with large sample size; (ii) unclear definition of sufficiently information-rich data representations for e.g. peptide identification; (iii) lack of benchmarking of ML methods on specific LC-MS problems.

RESULTS

We created the MS2AI pipeline that automates the process of gathering vast quantities of MS data for large-scale ML applications. The software retrieves raw data from either in-house sources or from the proteomics identifications database, PRIDE. Subsequently, the raw data are stored in a standardized format amenable for ML, encompassing MS1/MS2 spectra and peptide identifications. This tool bridges the gap between MS and AI, and to this effect we also present an ML application in the form of a convolutional neural network for the identification of oxidized peptides.

AVAILABILITY AND IMPLEMENTATION

An open-source implementation of the software can be found at https://gitlab.com/roettgerlab/ms2ai.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Potential Use of Serum Proteomics for Monitoring COVID-19 Progression to Complement RT-PCR Detection

RT-PCR is the primary method to diagnose COVID-19 and is also used to monitor the disease course. This approach, however, suffers from false negatives due to RNA instability and poses a high risk to medical practitioners. Here, we investigated the potential of using serum proteomics to predict viral nucleic acid positivity during COVID-19. We analyzed the proteome of 275 inactivated serum samples from 54 out of 144 COVID-19 patients and shortlisted 42 regulated proteins in the severe group and 12 in the non-severe group. Using these regulated proteins and several key clinical indexes, including days after symptoms onset, platelet counts, and magnesium, we developed two machine learning models to predict nucleic acid positivity, with an AUC of 0.94 in severe cases and 0.89 in non-severe cases, respectively. Our data suggest the potential of using a serum protein-based machine learning model to monitor COVID-19 progression, thus complementing swab RT-PCR tests. More efforts are required to promote this approach into clinical practice since mass spectrometry-based protein measurement is not currently widely accessible in clinic.

Deep learning approaches for data-independent acquisition proteomics

INTRODUCTION

Data-independent acquisition (DIA) is an emerging technology for large-scale proteomic studies. DIA data analysis methods are evolving rapidly, and deep learning has cut a conspicuous figure in this field.

AREAS COVERED

This review discusses and provides an overview of the deep learning methods that are used for DIA data analysis, including spectral library prediction, feature scoring, and statistical control in peptide-centric analysis, as well as de novo peptide sequencing. Literature searches were performed for articles, including preprints, up to December 2021 from PubMed, Scopus, and Web of Science databases.

EXPERT OPINION

While spectral library prediction has broken through the limitation on proteome coverage of experimental libraries, the statistical burden due to the large query space is the remaining challenge of utilizing proteome-wide predicted libraries. Analysis of post-translational modifications is another promising direction of deep learning-based DIA methods.

DeepPhospho accelerates DIA phosphoproteome profiling through in silico library generation

Phosphoproteomics integrating data-independent acquisition (DIA) enables deep phosphoproteome profiling with improved quantification reproducibility and accuracy compared to data-dependent acquisition (DDA)-based phosphoproteomics. DIA data mining heavily relies on a spectral library that in most cases is built on DDA analysis of the same sample. Construction of this project-specific DDA library impairs the analytical throughput, limits the proteome coverage, and increases the sample size for DIA phosphoproteomics. Herein we introduce a deep neural network, DeepPhospho, which conceptually differs from previous deep learning models to achieve accurate predictions of LC-MS/MS data for phosphopeptides. By leveraging in silico libraries generated by DeepPhospho, we establish a DIA workflow for phosphoproteome profiling which involves DIA data acquisition and data mining with DeepPhospho predicted libraries, thus circumventing the need of DDA library construction. Our DeepPhospho-empowered workflow substantially expands the phosphoproteome coverage while maintaining high quantification performance, which leads to the discovery of more signaling pathways and regulated kinases in an EGF signaling study than the DDA library-based approach. DeepPhospho is provided as a web server as well as an offline app to facilitate user access to model training, predictions and library generation.

Deep-Learning-Derived Evaluation Metrics Enable Effective Benchmarking of Computational Tools for Phosphopeptide Identification

Tandem mass spectrometry (MS/MS)-based phosphoproteomics is a powerful technology for global phosphorylation analysis. However, applying four computational pipelines to a typical mass spectrometry (MS)-based phosphoproteomic dataset from a human cancer study, we observed a large discrepancy among the reported phosphopeptide identification and phosphosite localization results, underscoring a critical need for benchmarking. While efforts have been made to compare performance of computational pipelines using data from synthetic phosphopeptides, evaluations involving real application data have been largely limited to comparing the numbers of phosphopeptide identifications due to the lack of appropriate evaluation metrics. We investigated three deep-learning-derived features as potential evaluation metrics: phosphosite probability, Delta RT, and spectral similarity. Predicted phosphosite probability is computed by MusiteDeep, which provides high accuracy as previously reported; Delta RT is defined as the absolute retention time (RT) difference between RTs observed and predicted by AutoRT; and spectral similarity is defined as the Pearson’s correlation coefficient between spectra observed and predicted by pDeep2. Using a synthetic peptide dataset, we found that both Delta RT and spectral similarity provided excellent discrimination between correct and incorrect peptide-spectrum matches (PSMs) both when incorrect PSMs involved wrong peptide sequences and even when incorrect PSMs were caused by only incorrect phosphosite localization. Based on these results, we used all the three deep-learning-derived features as evaluation metrics to compare different computational pipelines on diverse set of phosphoproteomic datasets and showed their utility in benchmarking performance of the pipelines. The benchmark metrics demonstrated in this study will enable users to select computational pipelines and parameters for routine analysis of phosphoproteomics data and will offer guidance for developers to improve computational methods.

•

Computational method selection substantially affects phosphopeptide identification.

•

Deep-learning-derived metrics effectively discriminate correct and incorrect PSMs.

•

Novel metrics enable computational method comparison on real application data.

Artificial intelligence for proteomics and biomarker discovery

There is an avalanche of biomedical data generation and a parallel expansion in computational capabilities to analyze and make sense of these data. Starting with genome sequencing and widely employed deep sequencing technologies, these trends have now taken hold in all omics disciplines and increasingly call for multi-omics integration as well as data interpretation by artificial intelligence technologies. Here, we focus on mass spectrometry (MS)-based proteomics and describe how machine learning and, in particular, deep learning now predicts experimental peptide measurements from amino acid sequences alone. This will dramatically improve the quality and reliability of analytical workflows because experimental results should agree with predictions in a multi-dimensional data landscape. Machine learning has also become central to biomarker discovery from proteomics data, which now starts to outperform existing best-in-class assays. Finally, we discuss model transparency and explainability and data privacy that are required to deploy MS-based biomarkers in clinical settings.

Diving Deep into the Data: A Review of Deep Learning Approaches and Potential Applications in Foodomics

Deep learning is a trending field in bioinformatics; so far, mostly known for image processing and speech recognition, but it also shows promising possibilities for data processing in food analysis, especially, foodomics. Thus, more and more deep learning approaches are used. This review presents an introduction into deep learning in the context of metabolomics and proteomics, focusing on the prediction of shelf-life, food authenticity, and food quality. Apart from the direct food-related applications, this review summarizes deep learning for peptide sequencing and its context to food analysis. The review’s focus further lays on MS (mass spectrometry)-based approaches. As a result of the constant development and improvement of analytical devices, as well as more complex holistic research questions, especially with the diverse and complex matrix food, there is a need for more effective methods for data processing. Deep learning might offer meeting this need and gives prospect to deal with the vast amount and complexity of data.

Deep learning neural network tools for proteomics

Mass-spectrometry-based proteomics enables quantitative analysis of thousands of human proteins. However, experimental and computational challenges restrict progress in the field. This review summarizes the recent flurry of machine-learning strategies using artificial deep neural networks (or "deep learning") that have started to break barriers and accelerate progress in the field of shotgun proteomics. Deep learning now accurately predicts physicochemical properties of peptides from their sequence, including tandem mass spectra and retention time. Furthermore, deep learning methods exist for nearly every aspect of the modern proteomics workflow, enabling improved feature selection, peptide identification, and protein inference.

INFERYS rescoring: Boosting peptide identifications and scoring confidence of database search results

Database search engines for bottom-up proteomics largely ignore peptide fragment ion intensities during the automated scoring of tandem mass spectra against protein databases. Recent advances in deep learning allow the accurate prediction of peptide fragment ion intensities. Using these predictions to calculate additional intensity-based scores helps to overcome this drawback. Here, we describe a processing workflow termed INFERYS™ rescoring for the intensity-based rescoring of Sequest HT search engine results in Thermo Scientific™ Proteome Discoverer™ 2.5 software. The workflow is based on the deep learning platform INFERYS capable of predicting fragment ion intensities, which runs on personal computers without the need for graphics processing units. This workflow calculates intensity-based scores comparing peptide spectrum matches from Sequest HT and predicted spectra. Resulting scores are combined with classical search engine scores for input to the false discovery rate estimation tool Percolator. We demonstrate the merits of this approach by analyzing a classical HeLa standard sample and exemplify how this workflow leads to a better separation of target and decoy identifications, in turn resulting in increased peptide spectrum match, peptide and protein identification numbers. On an immunopeptidome dataset, this workflow leads to a 50% increase in identified peptides, emphasizing the advantage of intensity-based scores when analyzing low-intensity spectra or analytes with very similar physicochemical properties that require vast search spaces. Overall, the end-to-end integration of INFERYS rescoring enables simple and easy access to a powerful enhancement to classical database search engines, promising a deeper, more confident and more comprehensive analysis of proteomic data from any organism by unlocking the intensity dimension of tandem mass spectra for identification and more confident scoring.

pDeep3: Toward More Accurate Spectrum Prediction with Fast Few-Shot Learning

Spectrum prediction using deep learning has attracted a lot of attention in recent years. Although existing deep learning methods have dramatically increased the prediction accuracy, there is still considerable space for improvement, which is presently limited by the difference of fragmentation types or instrument settings. In this work, we use the few-shot learning method to fit the data online to make up for the shortcoming. The method is evaluated using ten data sets, where the instruments includes Velos, QE, Lumos, and Sciex, with collision energies being differently set. Experimental results show that few-shot learning can achieve higher prediction accuracy with almost negligible computing resources. For example, on the data set from a untrained instrument Sciex-6600, within about 10 s, the prediction accuracy is increased from 69.7% to 86.4%; on the CID (collision-induced dissociation) data set, the prediction accuracy of the model trained by HCD (higher energy collision dissociation) spectra is increased from 48.0% to 83.9%. It is also shown that, the method is not critical to data quality and is sufficiently efficient to fill the accuracy gap. The source code of pDeep3 is available at http://pfind.ict.ac.cn/software/pdeep3.

pDeepXL: MS/MS Spectrum Prediction for Cross-Linked Peptide Pairs by Deep Learning

In cross-linking mass spectrometry, the identification of cross-linked peptide pairs heavily relies on the ability of a database search engine to measure the similarities between experimental and theoretical MS/MS spectra. However, the lack of accurate ion intensities in theoretical spectra impairs the performance of search engines, in particular, on proteome scales. Here we introduce pDeepXL, a deep neural network to predict MS/MS spectra of cross-linked peptide pairs. To train pDeepXL, we used the transfer-learning technique because it facilitated the training with limited benchmark data of cross-linked peptide pairs. Test results on more than ten data sets showed that pDeepXL accurately predicted the spectra of both noncleavable DSS/BS3/Leiker cross-linked peptide pairs (>80% of predicted spectra have Pearson's r values higher than 0.9) and cleavable DSSO/DSBU cross-linked peptide pairs (>75% of predicted spectra have Pearson's r values higher than 0.9). pDeepXL also achieved the accurate prediction on unseen data sets using an online fine-tuning technique. Lastly, integrating pDeepXL into a database search engine increased the number of identified cross-link spectra by 18% on average.

Application of spectral library prediction for parallel reaction monitoring of viral peptides

A major part of the analysis of parallel reaction monitoring (PRM) data is the comparison of observed fragment ion intensities to a library spectrum. Classically, these libraries are generated by data‐dependent acquisition (DDA). Here, we test Prosit, a published deep neural network algorithm, for its applicability in predicting spectral libraries for PRM. For this purpose, we targeted 1529 precursors derived from synthetic viral peptides and analyzed the data with Prosit and DDA‐derived libraries. Viral peptides were chosen as an example, because virology is an area where in silico library generation could significantly improve PRM assay design. With both libraries a total of 1174 precursors were identified. Notably, compared to the DDA‐derived library, we could identify 101 more precursors by using the Prosit‐derived library. Additionally, we show that Prosit can be applied to predict tandem mass spectra of synthetic viral peptides with different collision energies. Finally, we used a spectral library predicted by Prosit and a DDA library to identify SARS‐CoV‐2 peptides from a simulated oropharyngeal swab demonstrating that both libraries are suited for peptide identification by PRM. Summarized, Prosit‐derived viral spectral libraries predicted in silico can be used for PRM data analysis, making DDA analysis for library generation partially redundant in the future.

CIDer: A Statistical Framework for Interpreting Differences in CID and HCD Fragmentation

Library searching is a powerful technique for detecting peptides using either data independent or data dependent acquisition. While both large-scale spectrum library curators and deep learning prediction approaches have focused on beam-type CID fragmentation (HCD), resonance CID fragmentation remains a popular technique. Here we demonstrate an approach to model the differences between HCD and CID spectra, and present a software tool, CIDer, for converting libraries between the two fragmentation methods. We demonstrate that just using a combination of simple linear models and basic principles of peptide fragmentation, we can explain up to 43% of the variation between ions fragmented by HCD and CID across an array of collision energy settings. We further show that in some circumstances, searching converted CID libraries can detect more peptides than searching existing CID libraries or libraries of machine learning predictions from FASTA databases. These results suggest that leveraging information in existing libraries by converting between HCD and CID libraries may be an effective interim solution while large-scale CID libraries are being developed.

Ionic liquid modification of metal-organic framework endows high selectivity for phosphoproteins adsorption

Zr-based metal-organic framework, UiO-66-NH₂, provides favorable adsorption capacity to phosphoproteins, however, it exhibits obvious nonspecific adsorption to other proteins. In the present work, we report a facile strategy to reduce the nonspecific adsorption of nonphosphoproteins by modifying UiO-66-NH₂ with imidazolium ionic liquids (ILs). With respect to bare UiO-66-NH₂, the modified counterpart, UiO@IL, exhibits much improved selectivity to phosphoproteins while maintains comparable adsorption performance. The surface of UiO@IL presents a strong hydrophilicity due to the modification of ILs. Hydrophobic and electrostatic interaction between the absorbent and nonphosphoprotein is significantly reduced. In addition, the interaction between imidazole group of ILs moiety and phosphate group in phosphoprotein ensures the favorable adsorption capacity of UiO@IL for phosphoproteins. Anionic moieties of ILs, i.e., Cl^-, Br^-, BF₄^-, CF₃SO₃^-, play negligible effect in the adsorption process. As a representative, phosphoprotein β-casein (β-ca) is selectively enriched at a mass ratio of BSA:β-ca = 100:1. UiO@IL was further applied for the selective enrichment of phosphoprotein in milk.

Recent Developments in Data Independent Acquisition (DIA) Mass Spectrometry: Application of Quantitative Analysis of the Brain Proteome

Mass spectrometry is the driving force behind current brain proteome analysis. In a typical proteomics approach, a protein isolate is digested into tryptic peptides and then analyzed by liquid chromatography–mass spectrometry. The recent advancements in data independent acquisition (DIA) mass spectrometry provide higher sensitivity and protein coverage than the classic data dependent acquisition. DIA cycles through a pre-defined set of peptide precursor isolation windows stepping through 400–1,200 m/z across the whole liquid chromatography gradient. All peptides within an isolation window are fragmented simultaneously and detected by tandem mass spectrometry. Peptides are identified by matching the ion peaks in a mass spectrum to a spectral library that contains information of the peptide fragment ions' pattern and its chromatography elution time. Currently, there are several reports on DIA in brain research, in particular the quantitative analysis of cellular and synaptic proteomes to reveal the spatial and/or temporal changes of proteins that underlie neuronal plasticity and disease mechanisms. Protocols in DIA are continuously improving in both acquisition and data analysis. The depth of analysis is currently approaching proteome-wide coverage, while maintaining high reproducibility in a stable and standardisable MS environment. DIA can be positioned as the method of choice for routine proteome analysis in basic brain research and clinical applications.

Deep learning embedder method and tool for mass spectra similarity search

Spectral similarity calculation is widely used in protein identification tools and mass spectra clustering algorithms while comparing theoretical or experimental spectra. The performance of the spectral similarity calculation plays an important role in these tools and algorithms especially in the analysis of large-scale datasets. Recently, deep learning methods have been proposed to improve the performance of clustering algorithms and protein identification by training the algorithms with existing data and the use of multiple spectra and identified peptide features. While the efficiency of these algorithms is still under study in comparison with traditional approaches, their application in proteomics data analysis is becoming more common. Here, we propose the use of deep learning to improve spectral similarity comparison. We assessed the performance of deep learning for spectral similarity, with GLEAMS and a newly trained embedder model (DLEAMSE), which uses high-quality spectra from PRIDE Cluster. Also, we developed a new bioinformatics tool (mslookup - https://github.com/bigbio/DLEAMSE/) that allows users to quickly search for spectra in previously identified mass spectra publish in public repositories and spectral libraries. Finally, we released a human database to enable bioinformaticians and biologists to search for identified spectra in their machines. SIGNIFICANCE STATEMENT: Spectral similarity calculation plays an important role in proteomics data analysis. With deep learning's ability to learn the implicit and effective features from large-scale training datasets, deep learning-based MS/MS spectra embedding models has emerged as a solution to improve mass spectral clustering similarity calculation algorithms. We compare multiple similarity scoring and deep learning methods in terms of accuracy (compute the similarity for a pair of the mass spectrum) and computing-time performance. The benchmark results showed no major differences in accuracy between DLEAMSE and normalized dot product for spectrum similarity calculations. The DLEAMSE GPU implementation is faster than NDP in preprocessing on the GPU server and the similarity calculation of DLEAMSE (Euclidean distance on 32-D vectors) takes about 1/3 of dot product calculations. The deep learning model (DLEAMSE) encoding and embedding steps needed to run once for each spectrum and the embedded 32-D points can be persisted in the repository for future comparison, which is faster for future comparisons and large-scale data. Based on these, we proposed a new tool mslookup that enables the researcher to find spectra previously identified in public data. The tool can be also used to generate in-house databases of previously identified spectra to share with other laboratories and consortiums.

Improving the generative performance of chemical autoencoders through transfer learning

Generative models are a sub-class of machine learning models that are capable of generating new samples with a target set of properties. In chemical and materials applications, these new samples might be drug targets, novel semiconductors, or catalysts constrained to exhibit an application-specific set of properties. Given their potential to yield high-value targets from otherwise intractable design spaces, generative models are currently under intense study with respect to how predictions can be improved through changes in model architecture and data representation. Here we explore the potential of multi-task transfer learning as a complementary approach to improving the validity and property specificity of molecules generated by such models. We have compared baseline generative models trained on a single property prediction task against models trained on additional ancillary prediction tasks and observe a generic positive impact on the validity and specificity of the multi-task models. In particular, we observe that the validity of generated structures is strongly affected by whether or not the models have chemical property data, as opposed to only syntactic structural data, supplied during learning. We demonstrate this effect in both interpolative and extrapolative scenarios (i.e., where the generative targets are poorly represented in training data) for models trained to generate high energy structures and models trained to generated structures with targeted bandgaps within certain ranges. In both instances, the inclusion of additional chemical property data improves the ability of models to generate valid, unique structures with increased property specificity. This approach requires only minor alterations to existing generative models, in many cases leveraging prediction frameworks already native to these models. Additionally, the transfer learning strategy is complementary to ongoing efforts to improve model architectures and data representation and can foreseeably be stacked on top of these developments.

Design, Screening, and Testing of Non-Rational Peptide Libraries with Antimicrobial Activity: In Silico and Experimental Approaches

One of the challenges of modern biotechnology is to find new routes to mitigate the resistance to conventional antibiotics. Antimicrobial peptides (AMPs) are an alternative type of biomolecules, naturally present in a wide variety of organisms, with the capacity to overcome the current microorganism resistance threat. Here, we reviewed our recent efforts to develop a new library of non-rationally produced AMPs that relies on bacterial genome inherent diversity and compared it with rationally designed libraries. Our approach is based on a four-stage workflow process that incorporates the interplay of recent developments in four major emerging technologies: artificial intelligence, molecular dynamics, surface-display in microorganisms, and microfluidics. Implementing this framework is challenging because to obtain reliable results, the in silico algorithms to search for candidate AMPs need to overcome issues of the state-of-the-art approaches that limit the possibilities for multi-space data distribution analyses in extremely large databases. We expect to tackle this challenge by using a recently developed classification algorithm based on deep learning models that rely on convolutional layers and gated recurrent units. This will be complemented by carefully tailored molecular dynamics simulations to elucidate specific interactions with lipid bilayers. Candidate AMPs will be recombinantly-expressed on the surface of microorganisms for further screening via different droplet-based microfluidic-based strategies to identify AMPs with the desired lytic abilities. We believe that the proposed approach opens opportunities for searching and screening bioactive peptides for other applications.

DeepRescore: Leveraging Deep Learning to Improve Peptide Identification in Immunopeptidomics

The identification of major histocompatibility complex (MHC)-binding peptides in mass spectrometry (MS)-based immunopeptideomics relies largely on database search engines developed for proteomics data analysis. However, because immunopeptidomics experiments do not involve enzymatic digestion at specific residues, an inflated search space leads to a high false positive rate and low sensitivity in peptide identification. In order to improve the sensitivity and reliability of peptide identification, a post-processing tool named DeepRescore is developed. DeepRescore combines peptide features derived from deep learning predictions, namely accurate retention timeand MS/MS spectra predictions, with previously used features to rescore peptide-spectrum matches. Using two public immunopeptidomics datasets, it is shown that rescoring by DeepRescore increases both the sensitivity and reliability of MHC-binding peptide and neoantigen identifications compared to existing methods. It is also shown that the performance improvement is, to a large extent, driven by the deep learning-derived features. DeepRescore is developed using NextFlow and Docker and is available at https://github.com/bzhanglab/DeepRescore.

Deep Learning in Proteomics

Proteomics, the study of all the proteins in biological systems, is becoming a data-rich science. Protein sequences and structures are comprehensively catalogued in online databases. With recent advancements in tandem mass spectrometry (MS) technology, protein expression and post-translational modifications (PTMs) can be studied in a variety of biological systems at the global scale. Sophisticated computational algorithms are needed to translate the vast amount of data into novel biological insights. Deep learning automatically extracts data representations at high levels of abstraction from data, and it thrives in data-rich scientific research domains. Here, a comprehensive overview of deep learning applications in proteomics, including retention time prediction, MS/MS spectrum prediction, de novo peptide sequencing, PTM prediction, major histocompatibility complex-peptide binding prediction, and protein structure prediction, is provided. Limitations and the future directions of deep learning in proteomics are also discussed. This review will provide readers an overview of deep learning and how it can be used to analyze proteomics data.