Deep Learning in Proteomics

Current computational tools for protein lysine acylation site prediction

As a main subtype of post-translational modification (PTM), protein lysine acylations (PLAs) play crucial roles in regulating diverse functions of proteins. With recent advancements in proteomics technology, the identification of PTM is becoming a data-rich field. A large amount of experimentally verified data is urgently required to be translated into valuable biological insights. With computational approaches, PLA can be accurately detected across the whole proteome, even for organisms with small-scale datasets. Herein, a comprehensive summary of 166 in silico PLA prediction methods is presented, including a single type of PLA site and multiple types of PLA sites. This recapitulation covers important aspects that are critical for the development of a robust predictor, including data collection and preparation, sample selection, feature representation, classification algorithm design, model evaluation, and method availability. Notably, we discuss the application of protein language models and transfer learning to solve the small-sample learning issue. We also highlight the prediction methods developed for functionally relevant PLA sites and species/substrate/cell-type-specific PLA sites. In conclusion, this systematic review could potentially facilitate the development of novel PLA predictors and offer useful insights to researchers from various disciplines.

ARID3C Acts as a Regulator of Monocyte-to-Macrophage Differentiation Interacting with NPM1

ARID3C is a protein located on human chromosome 9 and expressed at low levels in various organs, yet its biological function has not been elucidated. In this study, we investigated both the cellular localization and function of ARID3C. Employing a combination of LC-MS/MS and deep learning techniques, we identified NPM1 as a binding partner for ARID3C's nuclear shuttling. ARID3C was found to predominantly localize with the nucleus, where it functioned as a transcription factor for genes STAT3, STAT1, and JUNB, thereby facilitating monocyte-to-macrophage differentiation. The precise binding sites between ARID3C and NPM1 were predicted by AlphaFold2. Mutating this binding site prevented ARID3C from interacting with NPM1, resulting in its retention in the cytoplasm instead of translocation to the nucleus. Consequently, ARID3C lost its ability to bind to the promoters of target genes, leading to a loss of monocyte-to-macrophage differentiation. Collectively, our findings indicate that ARID3C forms a complex with NPM1 to translocate to the nucleus, acting as a transcription factor that promotes the expression of the genes involved in monocyte-to-macrophage differentiation.

Imputation of label-free quantitative mass spectrometry-based proteomics data using self-supervised deep learning

Imputation techniques provide means to replace missing measurements with a value and are used in almost all downstream analysis of mass spectrometry (MS) based proteomics data using label-free quantification (LFQ). Here we demonstrate how collaborative filtering, denoising autoencoders, and variational autoencoders can impute missing values in the context of LFQ at different levels. We applied our method, proteomics imputation modeling mass spectrometry (PIMMS), to an alcohol-related liver disease (ALD) cohort with blood plasma proteomics data available for 358 individuals. Removing 20 percent of the intensities we were able to recover 15 out of 17 significant abundant protein groups using PIMMS-VAE imputations. When analyzing the full dataset we identified 30 additional proteins (+13.2%) that were significantly differentially abundant across disease stages compared to no imputation and found that some of these were predictive of ALD progression in machine learning models. We, therefore, suggest the use of deep learning approaches for imputing missing values in MS-based proteomics on larger datasets and provide workflows for these.

Recent Developments in Machine Learning for Mass Spectrometry

Statistical analysis and modeling of mass spectrometry (MS) data have a long and rich history with several modern MS-based applications using statistical and chemometric methods. Recently, machine learning (ML) has experienced a renaissance due to advents in computational hardware and the development of new algorithms for artificial neural networks (ANN) and deep learning architectures. Moreover, recent successes of new ANN and deep learning architectures in several areas of science, engineering, and society have further strengthened the ML field. Importantly, modern ML methods and architectures have enabled new approaches for tasks related to MS that are now widely adopted in several popular MS-based subdisciplines, such as mass spectrometry imaging and proteomics. Herein, we aim to provide an introductory summary of the practical aspects of ML methodology relevant to MS. Additionally, we seek to provide an up-to-date review of the most recent developments in ML integration with MS-based techniques while also providing critical insights into the future direction of the field.

Proteomics of Penicillium chrysogenum for a Deeper Understanding of Lead (Pb) Metal Bioremediation

Penicillium chrysogenum (P. chrysogenum), a ubiquitous filamentous fungus, has demonstrated remarkable potential in the bioremediation of lead-contaminated environments. Its inherent tolerance and bioaccumulation capacity for lead (Pb), coupled with its relatively rapid growth rate, make it an attractive candidate for bioremediation applications. This study aims to identify the proteomic changes in P. chrysogenuminduced by Pb metal stress and unravel the roles of identified proteins in molecular mechanisms and cellular responses. Untargeted proteomic analysis was carried out using a two-dimensional difference in gel electrophoresis (2D-DIGE) coupled with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS). This study reported the identification of 43 statistically significant proteins (24 upregulated and 19 downregulated, ANOVA, p ≤ 0.05; fold change ≥1.5) in P. chrysogenum as a consequence of Pb treatment. Proteins were grouped according to their function into 18 groups from which 13 proteins were related to metabolism, 11 were related to cellular process and signaling, and 19 proteins were related to information storage and processing. The current study is considered the first report about the proteomics study of P. chrysogenum under Pb stress conditions, where upregulated proteins could better explain the mechanism of tolerance and Pb toxicity removal. Our research has provided a thorough understanding of the molecular and cellular processes involved in fungal-metal interactions, paving the way for the development of innovative molecular markers for heavy metal myco-remediation. To the best of our knowledge, this study of P. chrysogenum provides valuable insights toward growing research in comprehending the metal-microbe interactions. This will facilitate development of novel molecular markers for metal bioremediation.

Artificial intelligence and neoantigens: paving the path for precision cancer immunotherapy

Cancer immunotherapy has witnessed rapid advancement in recent years, with a particular focus on neoantigens as promising targets for personalized treatments. The convergence of immunogenomics, bioinformatics, and artificial intelligence (AI) has propelled the development of innovative neoantigen discovery tools and pipelines. These tools have revolutionized our ability to identify tumor-specific antigens, providing the foundation for precision cancer immunotherapy. AI-driven algorithms can process extensive amounts of data, identify patterns, and make predictions that were once challenging to achieve. However, the integration of AI comes with its own set of challenges, leaving space for further research. With particular focus on the computational approaches, in this article we have explored the current landscape of neoantigen prediction, the fundamental concepts behind, the challenges and their potential solutions providing a comprehensive overview of this rapidly evolving field.

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.

Bayesian and deep-learning models applied to the early detection of ovarian cancer using multiple longitudinal biomarkers

BACKGROUND

Ovarian cancer is the most lethal of all gynecological cancers. Cancer Antigen 125 (CA125) is the best-performing ovarian cancer biomarker which however is still not effective as a screening test in the general population. Recent literature reports additional biomarkers with the potential to improve on CA125 for early detection when using longitudinal multimarker models.

METHODS

Our data comprised 180 controls and 44 cases with serum samples sourced from the multimodal arm of UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Our models were based on Bayesian change-point detection and recurrent neural networks.

RESULTS

We obtained a significantly higher performance for CA125-HE4 model using both methodologies (AUC 0.971, sensitivity 96.7% and AUC 0.987, sensitivity 96.7%) with respect to CA125 (AUC 0.949, sensitivity 90.8% and AUC 0.953, sensitivity 92.1%) for Bayesian change-point model (BCP) and recurrent neural networks (RNN) approaches, respectively. One year before diagnosis, the CA125-HE4 model also ranked as the best, whereas at 2 years before diagnosis no multimarker model outperformed CA125.

CONCLUSIONS

Our study identified and tested different combination of biomarkers using longitudinal multivariable models that outperformed CA125 alone. We showed the potential of multivariable models and candidate biomarkers to increase the detection rate of ovarian cancer.

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.

AlphaPept: a modern and open framework for MS-based proteomics

In common with other omics technologies, mass spectrometry (MS)-based proteomics produces ever-increasing amounts of raw data, making efficient analysis a principal challenge. A plethora of different computational tools can process the MS data to derive peptide and protein identification and quantification. However, during the last years there has been dramatic progress in computer science, including collaboration tools that have transformed research and industry. To leverage these advances, we develop AlphaPept, a Python-based open-source framework for efficient processing of large high-resolution MS data sets. Numba for just-in-time compilation on CPU and GPU achieves hundred-fold speed improvements. AlphaPept uses the Python scientific stack of highly optimized packages, reducing the code base to domain-specific tasks while accessing the latest advances. We provide an easy on-ramp for community contributions through the concept of literate programming, implemented in Jupyter Notebooks. Large datasets can rapidly be processed as shown by the analysis of hundreds of proteomes in minutes per file, many-fold faster than acquisition. AlphaPept can be used to build automated processing pipelines with web-serving functionality and compatibility with downstream analysis tools. It provides easy access via one-click installation, a modular Python library for advanced users, and via an open GitHub repository for developers.

i5mC-DCGA: an improved hybrid network framework based on the CBAM attention mechanism for identifying promoter 5mC sites

BACKGROUND

5-Methylcytosine (5mC) plays a very important role in gene stability, transcription, and development. Therefore, accurate identification of the 5mC site is of key importance in genetic and pathological studies. However, traditional experimental methods for identifying 5mC sites are time-consuming and costly, so there is an urgent need to develop computational methods to automatically detect and identify these 5mC sites.

RESULTS

Deep learning methods have shown great potential in the field of 5mC sites, so we developed a deep learning combinatorial model called i5mC-DCGA. The model innovatively uses the Convolutional Block Attention Module (CBAM) to improve the Dense Convolutional Network (DenseNet), which is improved to extract advanced local feature information. Subsequently, we combined a Bidirectional Gated Recurrent Unit (BiGRU) and a Self-Attention mechanism to extract global feature information. Our model can learn feature representations of abstract and complex from simple sequence coding, while having the ability to solve the sample imbalance problem in benchmark datasets. The experimental results show that the i5mC-DCGA model achieves 97.02%, 96.52%, 96.58% and 85.58% in sensitivity (Sn), specificity (Sp), accuracy (Acc) and matthews correlation coefficient (MCC), respectively.

CONCLUSIONS

The i5mC-DCGA model outperforms other existing prediction tools in predicting 5mC sites, and it is currently the most representative promoter 5mC site prediction tool. The benchmark dataset and source code for the i5mC-DCGA model can be found in https://github.com/leirufeng/i5mC-DCGA .

UniSpec: Deep Learning for Predicting the Full Range of Peptide Fragment Ion Series to Enhance the Proteomics Data Analysis Workflow

We present UniSpec, an attention-driven deep neural network designed to predict comprehensive collision-induced fragmentation spectra, thereby improving peptide identification in shotgun proteomics. Utilizing a training data set of 1.8 million unique high-quality tandem mass spectra (MS2) from 0.8 million unique peptide ions, UniSpec learned with a peptide fragmentation dictionary encompassing 7919 fragment peaks. Among these, 5712 are neutral loss peaks, with 2310 corresponding to modification-specific neutral losses. Remarkably, UniSpec can predict 73%-77% of fragment intensities based on our NIST reference library spectra, a significant leap from the 35%-45% coverage of only b and y ions. Comparative studies with Prosit elucidate that while both models are strong at predicting their respective fragment ion series, UniSpec particularly shines in generating more complex MS2 spectra with diverse ion annotations. The integration of UniSpec's predictions into shotgun proteomics data analysis boosts the identification rate of tryptic peptides by 48% at a 1% false discovery rate (FDR) and 60% at a more confident 0.1% FDR. Using UniSpec's predicted in-silico spectral library, the search results closely matched those from search engines and experimental spectral libraries used in peptide identification, highlighting its potential as a stand-alone identification tool. The source code and Python scripts are available on GitHub (https://github.com/usnistgov/UniSpec) and Zenodo (https://zenodo.org/records/10452792), and all data sets and analysis results generated in this work were deposited in Zenodo (https://zenodo.org/records/10052268).

Steroid profiling in adrenal disease

The measurement of steroid hormones in blood and urine, which reflects steroid biosynthesis and metabolism, has been recognized as a valuable tool for identifying and distinguishing steroidogenic disorders. The application of mass spectrometry enables the reliable and simultaneous analysis of large panels of steroids, ushering in a new era for diagnosing adrenal diseases. However, the interpretation of complex hormone results necessitates the expertise and experience of skilled clinicians. In this scenario, machine learning techniques are gaining worldwide attention within healthcare fields. The clinical values of combining mass spectrometry-based steroid profiles analysis with machine learning models, also known as steroid metabolomics, have been investigated for identifying and discriminating adrenal disorders such as adrenocortical carcinomas, adrenocortical adenomas, and congenital adrenal hyperplasia. This promising approach is expected to lead to enhanced clinical decision-making in the field of adrenal diseases. This review will focus on the clinical performances of steroid profiling, which is measured using mass spectrometry and analyzed by machine learning techniques, in the realm of decision-making for adrenal diseases.

Development of digital health management systems in longitudinal study: The Malaysian cohort experience

Background

The management of extensive longitudinal data in cohort studies presents significant challenges, particularly in middle-income countries like Malaysia where technological resources may be limited. These challenges include ensuring data integrity, security, and scalability of storage solutions over extended periods.

Objective

This article outlines innovative methods developed and implemented by The Malaysian Cohort project to effectively manage and maintain large-scale databases from project inception through the follow-up phase, ensuring robust data privacy and security.

Methods

We describe the comprehensive strategies employed to develop and sustain the database infrastructure necessary for handling large volumes of data collected during the study. This includes the integration of advanced information management systems and adherence to stringent data security protocols.

Outcomes

Key achievements include the establishment of a scalable database architecture and an effective data privacy framework that together support the dynamic requirements of longitudinal healthcare research. The solutions implemented serve as a model for similar cohort studies in resource-limited settings. The article also explores the broader implications of these methodologies for public health and personalized medicine, addressing both the challenges posed by big data in healthcare and the opportunities it offers for enhancing disease prevention and management strategies.

Conclusion

By sharing these insights, we aim to contribute to the global discourse on improving data management practices in cohort studies and to assist other researchers in overcoming the complexities associated with longitudinal health data.

Bacterial tolerance and detoxification of cyanide, arsenic and heavy metals: Holistic approaches applied to bioremediation of industrial complex wastes

Cyanide is a highly toxic compound that is found in wastewaters generated from different industrial activities, such as mining or jewellery. These residues usually contain high concentrations of other toxic pollutants like arsenic and heavy metals that may form different complexes with cyanide. To develop bioremediation strategies, it is necessary to know the metabolic processes involved in the tolerance and detoxification of these pollutants, but most of the current studies are focused on the characterization of the microbial responses to each one of these environmental hazards individually, and the effect of co-contaminated wastes on microbial metabolism has been hardly addressed. This work summarizes the main strategies developed by bacteria to alleviate the effects of cyanide, arsenic and heavy metals, analysing interactions among these toxic chemicals. Additionally, it is discussed the role of systems biology and synthetic biology as tools for the development of bioremediation strategies of complex industrial wastes and co-contaminated sites, emphasizing the importance and progress derived from meta-omic studies.

Small proteins in bacteria - Big challenges in prediction and identification

Proteins with up to 100 amino acids have been largely overlooked due to the challenges associated with predicting and identifying them using traditional methods. Recent advances in bioinformatics and machine learning, DNA sequencing, RNA and Ribo-seq technologies, and mass spectrometry (MS) have greatly facilitated the detection and characterisation of these elusive proteins in recent years. This has revealed their crucial role in various cellular processes including regulation, signalling and transport, as toxins and as folding helpers for protein complexes. Consequently, the systematic identification and characterisation of these proteins in bacteria have emerged as a prominent field of interest within the microbial research community. This review provides an overview of different strategies for predicting and identifying these proteins on a large scale, leveraging the power of these advanced technologies. Furthermore, the review offers insights into the future developments that may be expected in this field.

PepCNN deep learning tool for predicting peptide binding residues in proteins using sequence, structural, and language model features

Protein-peptide interactions play a crucial role in various cellular processes and are implicated in abnormal cellular behaviors leading to diseases such as cancer. Therefore, understanding these interactions is vital for both functional genomics and drug discovery efforts. Despite a significant increase in the availability of protein-peptide complexes, experimental methods for studying these interactions remain laborious, time-consuming, and expensive. Computational methods offer a complementary approach but often fall short in terms of prediction accuracy. To address these challenges, we introduce PepCNN, a deep learning-based prediction model that incorporates structural and sequence-based information from primary protein sequences. By utilizing a combination of half-sphere exposure, position specific scoring matrices from multiple-sequence alignment tool, and embedding from a pre-trained protein language model, PepCNN outperforms state-of-the-art methods in terms of specificity, precision, and AUC. The PepCNN software and datasets are publicly available at https://github.com/abelavit/PepCNN.git .

Mass spectrometry in the discovery of peptides involved in intercellular communication: From targeted to untargeted peptidomics approaches

Endogenous peptide hormones represent an essential class of biomolecules, which regulate cell-cell communications in diverse physiological processes of organisms. Mass spectrometry (MS) has been developed to be a powerful technology for identifying and quantifying peptides in a highly efficient manner. However, it is difficult to directly identify these peptide hormones due to their diverse characteristics, dynamic regulations, low abundance, and existence in a complicated biological matrix. Here, we summarize and discuss the roles of targeted and untargeted MS in discovering peptide hormones using bioassay-guided purification, bioinformatics screening, or the peptidomics-based approach. Although the peptidomics approach is expected to discover novel peptide hormones unbiasedly, only a limited number of successful cases have been reported. The critical challenges and corresponding measures for peptidomics from the steps of sample preparation, peptide extraction, and separation to the MS data acquisition and analysis are also discussed. We also identify emerging technologies and methods that can be integrated into the discovery platform toward the comprehensive study of endogenous peptide hormones.

Advances in data-independent acquisition mass spectrometry towards comprehensive digital proteome landscape

The data-independent acquisition mass spectrometry (DIA-MS) has rapidly evolved as a powerful alternative for highly reproducible proteome profiling with a unique strength of generating permanent digital maps for retrospective analysis of biological systems. Recent advancements in data analysis software tools for the complex DIA-MS/MS spectra coupled to fast MS scanning speed and high mass accuracy have greatly expanded the sensitivity and coverage of DIA-based proteomics profiling. Here, we review the evolution of the DIA-MS techniques, from earlier proof-of-principle of parallel fragmentation of all-ions or ions in selected m/z range, the sequential window acquisition of all theoretical mass spectra (SWATH-MS) to latest innovations, recent development in computation algorithms for data informatics, and auxiliary tools and advanced instrumentation to enhance the performance of DIA-MS. We further summarize recent applications of DIA-MS and experimentally-derived as well as in silico spectra library resources for large-scale profiling to facilitate biomarker discovery and drug development in human diseases with emphasis on the proteomic profiling coverage. Toward next-generation DIA-MS for clinical proteomics, we outline the challenges in processing multi-dimensional DIA data set and large-scale clinical proteomics, and continuing need in higher profiling coverage and sensitivity.

Current perspectives on mass spectrometry-based immunopeptidomics: the computational angle to tumor antigen discovery

Identification of tumor antigens presented by the human leucocyte antigen (HLA) molecules is essential for the design of effective and safe cancer immunotherapies that rely on T cell recognition and killing of tumor cells. Mass spectrometry (MS)-based immunopeptidomics enables high-throughput, direct identification of HLA-bound peptides from a variety of cell lines, tumor tissues, and healthy tissues. It involves immunoaffinity purification of HLA complexes followed by MS profiling of the extracted peptides using data-dependent acquisition, data-independent acquisition, or targeted approaches. By incorporating DNA, RNA, and ribosome sequencing data into immunopeptidomics data analysis, the proteogenomic approach provides a powerful means for identifying tumor antigens encoded within the canonical open reading frames of annotated coding genes and non-canonical tumor antigens derived from presumably non-coding regions of our genome. We discuss emerging computational challenges in immunopeptidomics data analysis and tumor antigen identification, highlighting key considerations in the proteogenomics-based approach, including accurate DNA, RNA and ribosomal sequencing data analysis, careful incorporation of predicted novel protein sequences into reference protein database, special quality control in MS data analysis due to the expanded and heterogeneous search space, cancer-specificity determination, and immunogenicity prediction. The advancements in technology and computation is continually enabling us to identify tumor antigens with higher sensitivity and accuracy, paving the way toward the development of more effective cancer immunotherapies.

Early Diagnosis: End-to-End CNN-LSTM Models for Mass Spectrometry Data Classification

Liquid chromatography-mass spectrometry (LC-MS) is a powerful method for cell profiling. The use of LC-MS technology is a tool of choice for cancer research since it provides molecular fingerprints of analyzed tissues. However, the ubiquitous presence of noise, the peaks shift between acquisitions, and the huge amount of information owing to the high dimensionality of the data make rapid and accurate cancer diagnosis a challenging task. Deep learning (DL) models are not only effective classifiers but are also well suited to jointly learn feature representation and classification tasks. This is particularly relevant when applied to raw LC-MS data and hence avoid the need for costly preprocessing and complicated feature selection. In this study, we propose a new end-to-end DL methodology that addresses all of the above challenges at once, while preserving the high potential of LC-MS data. Our DL model is designed to early discriminate between tumoral and normal tissues. It is a combination of a convolutional neural network (CNN) and a long short-term memory (LSTM) Network. The CNN network allows for significantly reducing the high dimensionality of the data while learning spatially relevant features. The LSTM network enables our model to capture temporal patterns. We show that our model outperforms not only benchmark models but also state-of-the-art models developed on the same data. Our framework is a promising strategy for improving early cancer detection during a diagnostic process.

Insight on physicochemical properties governing peptide MS1 response in HPLC-ESI-MS/MS: A deep learning approach

Accurate and absolute quantification of peptides in complex mixtures using quantitative mass spectrometry (MS)-based methods requires foreground knowledge and isotopically labeled standards, thereby increasing analytical expenses, time consumption, and labor, thus limiting the number of peptides that can be accurately quantified. This originates from differential ionization efficiency between peptides and thus, understanding the physicochemical properties that influence the ionization and response in MS analysis is essential for developing less restrictive label-free quantitative methods. Here, we used equimolar peptide pool repository data to develop a deep learning model capable of identifying amino acids influencing the MS1 response. By using an encoder-decoder with an attention mechanism and correlating attention weights with amino acid physicochemical properties, we obtain insight on properties governing the peptide-level MS1 response within the datasets. While the problem cannot be described by one single set of amino acids and properties, distinct patterns were reproducibly obtained. Properties are grouped in three main categories related to peptide hydrophobicity, charge, and structural propensities. Moreover, our model can predict MS1 intensity output under defined conditions based solely on peptide sequence input. Using a refined training dataset, the model predicted log-transformed peptide MS1 intensities with an average error of 9.7 ± 0.5% based on 5-fold cross validation, and outperformed random forest and ridge regression models on both log-transformed and real scale data. This work demonstrates how deep learning can facilitate identification of physicochemical properties influencing peptide MS1 responses, but also illustrates how sequence-based response prediction and label-free peptide-level quantification may impact future workflows within quantitative proteomics.

A Graph Convolutional Network Based on Univariate Neurodegeneration Biomarker for Alzheimer's Disease Diagnosis

OBJECTIVE

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease that is not easily detectable in the early stage. This study proposed an efficient method of applying a graph convolutional network (GCN) on the early prediction of AD.

METHODS

We proposed a univariate neurodegeneration biomarker (UNB) based GCN semi-supervised classification framework. We generated UNB by comparing the similarity of individual morphological atrophy pattern and the atrophy pattern of [Formula: see text] AD group according to the brain morphological abnormalities induced by AD. For the GCN semi-supervised classification model, we took the UNBs of individuals as the features of nodes and constructed the weight of edges according to the similarity of phenotypic information between individuals, which explored the essential features of individuals through spectral graph convolution. The attention module was constructed and embedded into the GCN framework, which may refine the input morphological features to highlight the main impact of AD on the cerebral cortex and weaken the instability caused by individual diversities, thereby identifying the significant ROIs affected by AD and improving the classification accuracy.

RESULTS

We tested the UNB-GCN framework on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. The estimated minimum sample sizes were 156, 349 and 423 for the longitudinal [Formula: see text] AD, [Formula: see text] mild cognitive impairment (MCI) and [Formula: see text] cognitively unimpaired (CU) groups, respectively. And the proposed UNB-GCN framework combined with the attention module can effectively improve the classification performance with 93.90% classification accuracy for AD vs. CU and 82.05% for AD vs. MCI on the validation set.

CONCLUSION

The proposed UNB measures were superior to the conventional volume measures in describing the AD-induced cerebral cortex morphological changes. And the UNB-GCN framework combined with attention module may effectively improve the classification performance between MCI subjects and AD patients. Clinical and Translational Impact Statement: This study aims to predict the early AD patients, so as to help clinicians develop effective interventions to delay the deterioration of AD symptoms.

Large scale proteomic studies create novel privacy considerations

Privacy protection is a core principle of genomic but not proteomic research. We identified independent single nucleotide polymorphism (SNP) quantitative trait loci (pQTL) from COPDGene and Jackson Heart Study (JHS), calculated continuous protein level genotype probabilities, and then applied a naïve Bayesian approach to link SomaScan 1.3K proteomes to genomes for 2812 independent subjects from COPDGene, JHS, SubPopulations and InteRmediate Outcome Measures In COPD Study (SPIROMICS) and Multi-Ethnic Study of Atherosclerosis (MESA). We correctly linked 90-95% of proteomes to their correct genome and for 95-99% we identify the 1% most likely links. The linking accuracy in subjects with African ancestry was lower (~ 60%) unless training included diverse subjects. With larger profiling (SomaScan 5K) in the Atherosclerosis Risk Communities (ARIC) correct identification was > 99% even in mixed ancestry populations. We also linked proteomes-to-proteomes and used the proteome only to determine features such as sex, ancestry, and first-degree relatives. When serial proteomes are available, the linking algorithm can be used to identify and correct mislabeled samples. This work also demonstrates the importance of including diverse populations in omics research and that large proteomic datasets (> 1000 proteins) can be accurately linked to a specific genome through pQTL knowledge and should not be considered unidentifiable.

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.

SynBioTools: a one-stop facility for searching and selecting synthetic biology tools

BACKGROUND

The rapid development of synthetic biology relies heavily on the use of databases and computational tools, which are also developing rapidly. While many tool registries have been created to facilitate tool retrieval, sharing, and reuse, no relatively comprehensive tool registry or catalog addresses all aspects of synthetic biology.

RESULTS

We constructed SynBioTools, a comprehensive collection of synthetic biology databases, computational tools, and experimental methods, as a one-stop facility for searching and selecting synthetic biology tools. SynBioTools includes databases, computational tools, and methods extracted from reviews via SCIentific Table Extraction, a scientific table-extraction tool that we built. Approximately 57% of the resources that we located and included in SynBioTools are not mentioned in bio.tools, the dominant tool registry. To improve users' understanding of the tools and to enable them to make better choices, the tools are grouped into nine modules (each with subdivisions) based on their potential biosynthetic applications. Detailed comparisons of similar tools in every classification are included. The URLs, descriptions, source references, and the number of citations of the tools are also integrated into the system.

CONCLUSIONS

SynBioTools is freely available at https://synbiotools.lifesynther.com/ . It provides end-users and developers with a useful resource of categorized synthetic biology databases, tools, and methods to facilitate tool retrieval and selection.

Align-gram: Rethinking the Skip-gram Model for Protein Sequence Analysis

The inception of next generations sequencing technologies have exponentially increased the volume of biological sequence data. Protein sequences, being quoted as the 'language of life', has been analyzed for a multitude of applications and inferences. Owing to the rapid development of deep learning, in recent years there have been a number of breakthroughs in the domain of Natural Language Processing. Since these methods are capable of performing different tasks when trained with a sufficient amount of data, off-the-shelf models are used to perform various biological applications. In this study, we investigated the applicability of the popular Skip-gram model for protein sequence analysis and made an attempt to incorporate some biological insights into it. We propose a novel k-mer embedding scheme, Align-gram, which is capable of mapping the similar k-mers close to each other in a vector space. Furthermore, we experiment with other sequence-based protein representations and observe that the embeddings derived from Align-gram aids modeling and training deep learning models better. Our experiments with a simple baseline LSTM model and a much complex CNN model of DeepGoPlus shows the potential of Align-gram in performing different types of deep learning applications for protein sequence analysis.

Proteomic Applications and Considerations: From Research to Patient Care

Despite technological advancements in the field of proteomics, the rate at which serum and plasma biomarkers identified using proteomic approaches are translated into clinical use remains extremely low. In this chapter, we describe recent technological advancements and analytical strategies in proteomic methods. We also describe the progress of proteomic blood-based biomarkers to date and discuss what the future of proteomics might entail with the use of multi-omic approaches and implementing machine learning on large proteomic datasets. Lastly, we provide several key considerations for biomarker studies, ranging from sample type to the use of reference samples, in order to achieve progress from bench to bedside, ultimately improving patient diagnosis, disease, and/or therapeutic monitoring and care.

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.

DNA-MP: a generalized DNA modifications predictor for multiple species based on powerful sequence encoding method

Accurate prediction of deoxyribonucleic acid (DNA) modifications is essential to explore and discern the process of cell differentiation, gene expression and epigenetic regulation. Several computational approaches have been proposed for particular type-specific DNA modification prediction. Two recent generalized computational predictors are capable of detecting three different types of DNA modifications; however, type-specific and generalized modifications predictors produce limited performance across multiple species mainly due to the use of ineffective sequence encoding methods. The paper in hand presents a generalized computational approach "DNA-MP" that is competent to more precisely predict three different DNA modifications across multiple species. Proposed DNA-MP approach makes use of a powerful encoding method "position specific nucleotides occurrence based 117 on modification and non-modification class densities normalized difference" (POCD-ND) to generate the statistical representations of DNA sequences and a deep forest classifier for modifications prediction. POCD-ND encoder generates statistical representations by extracting position specific distributional information of nucleotides in the DNA sequences. We perform a comprehensive intrinsic and extrinsic evaluation of the proposed encoder and compare its performance with 32 most widely used encoding methods on $17$ benchmark DNA modifications prediction datasets of $12$ different species using $10$ different machine learning classifiers. Overall, with all classifiers, the proposed POCD-ND encoder outperforms existing $32$ different encoders. Furthermore, combinedly over 5-fold cross validation benchmark datasets and independent test sets, proposed DNA-MP predictor outperforms state-of-the-art type-specific and generalized modifications predictors by an average accuracy of 7% across 4mc datasets, 1.35% across 5hmc datasets and 10% for 6ma datasets. To facilitate the scientific community, the DNA-MP web application is available at https://sds_genetic_analysis.opendfki.de/DNA_Modifications/.

Grain protein function prediction based on self-attention mechanism and bidirectional LSTM

With the development of genome sequencing technology, using computing technology to predict grain protein function has become one of the important tasks of bioinformatics. The protein data of four grains, soybean, maize, indica and japonica are selected in this experimental dataset. In this paper, a novel neural network algorithm Chemical-SA-BiLSTM is proposed for grain protein function prediction. The Chemical-SA-BiLSTM algorithm fuses the chemical properties of proteins on the basis of amino acid sequences, and combines the self-attention mechanism with the bidirectional Long Short-Term Memory network. The experimental results show that the Chemical-SA-BiLSTM algorithm is superior to other classical neural network algorithms, and can more accurately predict the protein function, which proves the effectiveness of the Chemical-SA-BiLSTM algorithm in the prediction of grain protein function. The source code of our method is available at https://github.com/HwaTong/Chemical-SA-BiLSTM.

Deep learning prediction boosts phosphoproteomics-based discoveries through improved phosphopeptide identification

Shotgun phosphoproteomics enables high-throughput analysis of phosphopeptides in biological samples, but low phosphopeptide identification rate in data analysis limits the potential of this technology. Here we present DeepRescore2, a computational workflow that leverages deep learning-based retention time and fragment ion intensity predictions to improve phosphopeptide identification and phosphosite localization. Using a state-of-the-art computational workflow as a benchmark, DeepRescore2 increases the number of correctly identified peptide-spectrum matches by 17% in a synthetic dataset and identifies 19%-46% more phosphopeptides in biological datasets. In a liver cancer dataset, 30% of the significantly altered phosphosites between tumor and normal tissues and 60% of the prognosis-associated phosphosites identified from DeepRescore2-processed data could not be identified based on the state-of-the-art workflow. Notably, DeepRescore2-processed data uniquely identifies EGFR hyperactivation as a new target in poor-prognosis liver cancer, which is validated experimentally. Integration of deep learning prediction in DeepRescore2 improves phosphopeptide identification and facilitates biological discoveries.

Layer-Wise Relevance Analysis for Motif Recognition in the Activation Pathway of the β2-Adrenergic GPCR Receptor

G-protein-coupled receptors (GPCRs) are cell membrane proteins of relevance as therapeutic targets, and are associated to the development of treatments for illnesses such as diabetes, Alzheimer's, or even cancer. Therefore, comprehending the underlying mechanisms of the receptor functional properties is of particular interest in pharmacoproteomics and in disease therapy at large. Their interaction with ligands elicits multiple molecular rearrangements all along their structure, inducing activation pathways that distinctly influence the cell response. In this work, we studied GPCR signaling pathways from molecular dynamics simulations as they provide rich information about the dynamic nature of the receptors. We focused on studying the molecular properties of the receptors using deep-learning-based methods. In particular, we designed and trained a one-dimensional convolution neural network and illustrated its use in a classification of conformational states: active, intermediate, or inactive, of the β2-adrenergic receptor when bound to the full agonist BI-167107. Through a novel explainability-oriented investigation of the prediction results, we were able to identify and assess the contribution of individual motifs (residues) influencing a particular activation pathway. Consequently, we contribute a methodology that assists in the elucidation of the underlying mechanisms of receptor activation-deactivation.

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.

DeepDN_iGlu: prediction of lysine glutarylation sites based on attention residual learning method and DenseNet

As a key issue in orchestrating various biological processes and functions, protein post-translational modification (PTM) occurs widely in the mechanism of protein's function of animals and plants. Glutarylation is a type of protein-translational modification that occurs at active ε-amino groups of specific lysine residues in proteins, which is associated with various human diseases, including diabetes, cancer, and glutaric aciduria type I. Therefore, the issue of prediction for glutarylation sites is particularly important. This study developed a brand-new deep learning-based prediction model for glutarylation sites named DeepDN_iGlu via adopting attention residual learning method and DenseNet. The focal loss function is utilized in this study in place of the traditional cross-entropy loss function to address the issue of a substantial imbalance in the number of positive and negative samples. It can be noted that DeepDN_iGlu based on the deep learning model offers a greater potential for the glutarylation site prediction after employing the straightforward one hot encoding method, with Sensitivity (Sn), Specificity (Sp), Accuracy (ACC), Mathews Correlation Coefficient (MCC), and Area Under Curve (AUC) of 89.29%, 61.97%, 65.15%, 0.33 and 0.80 accordingly on the independent test set. To the best of the authors' knowledge, this is the first time that DenseNet has been used for the prediction of glutarylation sites. DeepDN_iGlu has been deployed as a web server (https://bioinfo.wugenqiang.top/~smw/DeepDN_iGlu/) that is available to make glutarylation site prediction data more accessible.

Multienzyme deep learning models improve peptide de novo sequencing by mass spectrometry proteomics

Generating and analyzing overlapping peptides through multienzymatic digestion is an efficient procedure for de novo protein using from bottom-up mass spectrometry (MS). Despite improved instrumentation and software, de novo MS data analysis remains challenging. In recent years, deep learning models have represented a performance breakthrough. Incorporating that technology into de novo protein sequencing workflows require machine-learning models capable of handling highly diverse MS data. In this study, we analyzed the requirements for assembling such generalizable deep learning models by systemcally varying the composition and size of the training set. We assessed the generated models' performances using two test sets composed of peptides originating from the multienzyme digestion of samples from various species. The peptide recall values on the test sets showed that the deep learning models generated from a collection of highly N- and C-termini diverse peptides generalized 76% more over the termini-restricted ones. Moreover, expanding the training set's size by adding peptides from the multienzymatic digestion with five proteases of several species samples led to a 2-3 fold generalizability gain. Furthermore, we tested the applicability of these multienzyme deep learning (MEM) models by fully de novo sequencing the heavy and light monomeric chains of five commercial antibodies (mAbs). MEMs extracted over 10000 matching and overlapped peptides across six different proteases mAb samples, achieving a 100% sequence coverage for 8 of the ten polypeptide chains. We foretell that the MEMs' proven improvements to de novo analysis will positively impact several applications, such as analyzing samples of high complexity, unknown nature, or the peptidomics field.

iNSP-GCAAP: Identifying nonclassical secreted proteins using global composition of amino acid properties

Nonclassical secreted proteins (NSPs) refer to a group of proteins released into the extracellular environment under the facilitation of different biological transporting pathways apart from the Sec/Tat system. As experimental determination of NSPs is often costly and requires skilled handling techniques, computational approaches are necessary. In this study, we introduce iNSP-GCAAP, a computational prediction framework, to identify NSPs. We propose using global composition of a customized set of amino acid properties to encode sequence data and use the random forest (RF) algorithm for classification. We used the training dataset introduced by Zhang et al. (Bioinformatics, 36(3), 704-712, 2020) to develop our model and test it with the independent test set in the same study. The area under the receiver operating characteristic curve on that test set was 0.9256, which outperformed other state-of-the-art methods using the same datasets. Our framework is also deployed as a user-friendly web-based application to support the research community to predict NSPs.

Variability analysis of LC-MS experimental factors and their impact on machine learning

BACKGROUND

Machine learning (ML) technologies, especially deep learning (DL), have gained increasing attention in predictive mass spectrometry (MS) for enhancing the data-processing pipeline from raw data analysis to end-user predictions and rescoring. ML models need large-scale datasets for training and repurposing, which can be obtained from a range of public data repositories. However, applying ML to public MS datasets on larger scales is challenging, as they vary widely in terms of data acquisition methods, biological systems, and experimental designs.

RESULTS

We aim to facilitate ML efforts in MS data by conducting a systematic analysis of the potential sources of variability in public MS repositories. We also examine how these factors affect ML performance and perform a comprehensive transfer learning to evaluate the benefits of current best practice methods in the field for transfer learning.

CONCLUSIONS

Our findings show significantly higher levels of homogeneity within a project than between projects, which indicates that it is important to construct datasets most closely resembling future test cases, as transferability is severely limited for unseen datasets. We also found that transfer learning, although it did increase model performance, did not increase model performance compared to a non-pretrained model.

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 ).

ncRNAInter: a novel strategy based on graph neural network to discover interactions between lncRNA and miRNA

In recent years, many studies have illustrated the significant role that non-coding RNA (ncRNA) plays in biological activities, in which lncRNA, miRNA and especially their interactions have been proved to affect many biological processes. Some in silico methods have been proposed and applied to identify novel lncRNA-miRNA interactions (LMIs), but there are still imperfections in their RNA representation and information extraction approaches, which imply there is still room for further improving their performances. Meanwhile, only a few of them are accessible at present, which limits their practical applications. The construction of a new tool for LMI prediction is thus imperative for the better understanding of their relevant biological mechanisms. This study proposed a novel method, ncRNAInter, for LMI prediction. A comprehensive strategy for RNA representation and an optimized deep learning algorithm of graph neural network were utilized in this study. ncRNAInter was robust and showed better performance of 26.7% higher Matthews correlation coefficient than existing reputable methods for human LMI prediction. In addition, ncRNAInter proved its universal applicability in dealing with LMIs from various species and successfully identified novel LMIs associated with various diseases, which further verified its effectiveness and usability. All source code and datasets are freely available at https://github.com/idrblab/ncRNAInter.

A Hybrid Spectral Library and Protein Sequence Database Search Strategy for Bottom-Up and Top-Down Proteomic Data Analysis

Tandem mass spectrometry (MS/MS) is widely employed for the analysis of complex proteomic samples. While protein sequence database searching and spectral library searching are both well-established peptide identification methods, each has shortcomings. Protein sequence databases lack fragment peak intensity information, which can result in poor discrimination between correct and incorrect spectrum assignments. Spectral libraries usually contain fewer peptides than protein sequence databases, which limits the number of peptides that can be identified. Notably, few post-translationally modified peptides are represented in spectral libraries. This is because few search engines can both identify a broad spectrum of PTMs and create corresponding spectral libraries. Also, programs that generate spectral libraries using deep learning approaches are not yet able to accurately predict spectra for the vast majority of PTMs. Here, we address these limitations through use of a hybrid search strategy that combines protein sequence database and spectral library searches to improve identification success rates and sensitivity. This software uses Global PTM Discovery (G-PTM-D) to produce spectral libraries for a wide variety of different PTMs. These features, along with a new spectrum annotation and visualization tool, have been integrated into the freely available and open-source search engine MetaMorpheus.

The rise of the machines in chemistry

The use of artificial intelligence and, more specifically, deep learning methods in chemistry is becoming increasingly common. Applications in informatics fields, such as cheminformatics and proteomics, structural biology, and spectroscopy, including NMR, are on the rise. Recent developments in model architectures, such as graph convolutional neural networks and transformers, have been enabled by advancements in computational hardware and software. However, model architectures with more predictive power often require larger amounts of training data, which can be challenging to acquire, but this requirement can be mitigated through techniques like pretraining and fine-tuning. In spite of these successes, challenges remain, such as normalization and scaling of data, availability of experimentally acquired data, and model explainability.

CNNArginineMe: A CNN structure for training models for predicting arginine methylation sites based on the One-Hot encoding of peptide sequence

Protein arginine methylation (PRme), as one post-translational modification, plays a critical role in numerous cellular processes and regulates critical cellular functions. Though several in silico models for predicting PRme sites have been reported, new models may be required to develop due to the significant increase of identified PRme sites. In this study, we constructed multiple machine-learning and deep-learning models. The deep-learning model CNN combined with the One-Hot coding showed the best performance, dubbed CNNArginineMe. CNNArginineMe performed best in AUC scoring metrics in comparisons with several reported predictors. Additionally, we employed CNNArginineMe to predict arginine methylation proteome and performed functional analysis. The arginine methylated proteome is significantly enriched in the amyotrophic lateral sclerosis (ALS) pathway. CNNArginineMe is freely available at https://github.com/guoyangzou/CNNArginineMe.

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.

Mini-review: Recent advances in post-translational modification site prediction based on deep learning

Post-translational modifications (PTMs) are closely linked to numerous diseases, playing a significant role in regulating protein structures, activities, and functions. Therefore, the identification of PTMs is crucial for understanding the mechanisms of cell biology and diseases therapy. Compared to traditional machine learning methods, the deep learning approaches for PTM prediction provide accurate and rapid screening, guiding the downstream wet experiments to leverage the screened information for focused studies. In this paper, we reviewed the recent works in deep learning to identify phosphorylation, acetylation, ubiquitination, and other PTM types. In addition, we summarized PTM databases and discussed future directions with critical insights.

Leveraging Protein Dynamics to Identify Functional Phosphorylation Sites using Deep Learning Models

Accurate prediction of post-translational modifications (PTMs) is of great significance in understanding cellular processes, by modulating protein structure and dynamics. Nowadays, with the rapid growth of protein data at different "omics" levels, machine learning models largely enriched the prediction of PTMs. However, most machine learning models only rely on protein sequence and little structural information. The lack of the systematic dynamics analysis underlying PTMs largely limits the PTM functional predictions. In this research, we present two dynamics-centric deep learning models, namely, cDL-PAU and cDL-FuncPhos, by incorporating sequence, structure, and dynamics-based features to elucidate the molecular basis and underlying functional landscape of PTMs. cDL-PAU achieved satisfactory area under the curve (AUC) scores of 0.804-0.888 for predicting phosphorylation, acetylation, and ubiquitination (PAU) sites, while cDL-FuncPhos achieved an AUC value of 0.771 for predicting functional phosphorylation (FuncPhos) sites, displaying reliable improvements. Through a feature selection, the dynamics-based coupling and commute ability show large contributions in discovering PAU sites and FuncPhos sites, suggesting the allosteric propensity for important PTMs. The application of cDL-FuncPhos in three oncoproteins not only corroborates its strong performance in FuncPhos prioritization but also gains insight into the physical basis for the functions. The source code and data set of cDL-PAU and cDL-FuncPhos are available at https://github.com/ComputeSuda/PTM_ML.

A Discovery Strategy for Active Compounds of Chinese Medicine Based on the Prediction Model of Compound-Disease Relationship

An accurate characterization of diseases and compounds is the key to predicting the compound-disease relationship (CDR). However, due to the difficulty of a comprehensive description of CDR, the accuracy of traditional drug development models for large-scale CDR prediction is usually unsatisfactory. In order to solve this problem, we propose a new method that integrates the molecular descriptors of compounds and the symptom descriptors of diseases to build a CDR two-dimensional matrix to predict candidate active compounds. The Matlab software draws grayscale images of CDRs, which are used as a benchmark dataset for training convolutional neural network (CNN) models. The trained model is used to predict candidate antitumor active compounds. Among the AlexNet and GoogLeNet models, we selected the GoogLeNet model for the prediction of active compounds in Chinese medicine, and its Acc, Sen, Pre, F-measure, MCC, and AUC are 0.960, 0.956, 0.965, 0.960, 0.920, and 0.964, respectively. In the prediction results of compounds, 1624 candidate CDRs were found in 124 Chinese medicines. Among them, we obtained 31 features of candidate antitumor active compounds. This method provides new insights for the discovery of candidate active compounds in Chinese medicine.

Research progress of reduced amino acid alphabets in protein analysis and prediction

•

A comprehensive summary of the literature on the reduced amino acid alphabets.

•

A systematic review of the development history of reduced amino acid alphabets.

•

Rich application cases of amino acid reduction alphabets are described in the article.

•

A detailed analysis of the properties and uses of the reduced amino acid alphabets.

Proteins are the executors of cellular physiological activities, and accurate structural and function elucidation are crucial for the refined mapping of proteins. As a feature engineering method, the reduction of amino acid composition is not only an important method for protein structure and function analysis, but also opens a broad horizon for the complex field of machine learning. Representing sequences with fewer amino acid types greatly reduces the complexity and noise of traditional feature engineering in dimension, and provides more interpretable predictive models for machine learning to capture key features. In this paper, we systematically reviewed the strategy and method studies of the reduced amino acid (RAA) alphabets, and summarized its main research in protein sequence alignment, functional classification, and prediction of structural properties, respectively. In the end, we gave a comprehensive analysis of 672 RAA alphabets from 74 reduction methods.