Statistical analysis of proteomics data: A review on feature selection

Comprehensive Overview of Bottom-Up Proteomics Using Mass Spectrometry

Proteomics is the large scale study of protein structure and function from biological systems through protein identification and quantification. "Shotgun proteomics" or "bottom-up proteomics" is the prevailing strategy, in which proteins are hydrolyzed into peptides that are analyzed by mass spectrometry. Proteomics studies can be applied to diverse studies ranging from simple protein identification to studies of proteoforms, protein-protein interactions, protein structural alterations, absolute and relative protein quantification, post-translational modifications, and protein stability. To enable this range of different experiments, there are diverse strategies for proteome analysis. The nuances of how proteomic workflows differ may be challenging to understand for new practitioners. Here, we provide a comprehensive overview of different proteomics methods. We cover from biochemistry basics and protein extraction to biological interpretation and orthogonal validation. We expect this Review will serve as a handbook for researchers who are new to the field of bottom-up proteomics.

Groundwater salinity modeling and mapping using machine learning approaches: a case study in Sidi Okba region, Algeria

The groundwater salinization process complexity and the lack of data on its controlling factors are the main challenges for accurate predictions and mapping of aquifer salinity. For this purpose, effective machine learning (ML) methodologies are employed for effective modeling and mapping of groundwater salinity (GWS) in the Mio-Pliocene aquifer in the Sidi Okba region, Algeria, based on limited dataset of electrical conductivity (EC) measurements and readily available digital elevation model (DEM) derivatives. The dataset was randomly split into training (70%) and testing (30%) sets, and three wrapper selection methods, recursive feature elimination (RFE), forward feature selection (FFS), and backward feature selection (BFS) are applied to train the data. The resulting combinations are used as inputs for five ML models, namely random forest (RF), hybrid neuro-fuzzy inference system (HyFIS), K-nearest neighbors (KNN), cubist regression model (CRM), and support vector machine (SVM). The best-performing model is identified and applied to predict and map GWS across the entire study area. It is highlighted that the applied methods yield input variation combinations as critical factors that are often overlocked by many researchers, which substantially impacts the models' accuracy. Among different alternatives the RF model emerged as the most effective for predicting and mapping GWS in the study area, which led to the high performance in both the training (RMSE = 1.016, R = 0.854, and MAE = 0.759) and testing (RMSE = 1.069, R = 0.831, and MAE = 0.921) phases. The generated digital map highlighted the alarming situation regarding excessive GWS levels in the study area, particularly in zones of low elevations and far from the Foum Elgherza dam and Elbiraz wadi. Overall, this study represents a significant advancement over previous approaches, offering enhanced predictive performance for GWS with the minimum number of input variables.

Naïve Inflammatory Proteome Profiles of Glucocorticoid Responsive Polymyalgia Rheumatica and Rheumatic Arthritis Patients-Links to Triggers and Proteomic Manifestations

Polymyalgia rheumatica (PMR) is an inflammatory disorder of unknown etiology, sharing symptoms with giant cell arthritis (GCA) and rheumatoid arthritis (RA). The pathogenic inflammatory roots are still not well understood, and there is a lack of extensive biomarker studies to explain the disease debut and post-acute phase. This study aimed to deeply analyze the serum proteome and inflammatory response of PMR patients before and after glucocorticoid treatment. We included treatment-naïve PMR patients, collecting samples before and after 3 months of treatment. For comparison, disease-modifying antirheumatic drug (DMARD)-naïve RA patients were included and matched to healthy controls (CTL). The serum proteome was examined using label-free quantitative mass spectrometry, while inflammation levels were assessed using multiplex inflammatory cytokine and cell-free DNA assays. The serum proteomes of the four groups comprised acute phase reactants, coagulation factors, complement proteins, immunoglobulins, and apolipoproteins. Serum amyloid A (SAA1) was significantly reduced by active PMR treatment. Cell-free DNA levels in PMR and RA groups were significantly higher than in healthy controls due to acute inflammation. Complement factors had minimal changes post-treatment. The individual serum proteome in PMR patients showed over 100 abundantly variable proteins, emphasizing the systemic impact of PMR disease debut and the effect of treatment. Interleukin (IL)-6 and interferon-gamma (IFN-γ) were significantly impacted by glucocorticoid treatment. Our study defines the PMR serum proteome during glucocorticoid treatment and highlights the role of SAA1, IL-6, and IFN-γ in treatment responses. An involvement of PGLYRP2 in acute PMR could indicate a response to bacterial infection, highlighting its role in the acute phase of the immune response. The results suggest that PMR may be an aberrant response to a bacterial infection with an exacerbated IL-6 and acute phase inflammatory response and molecular attempts to limit the inflammation.

Exploring the mechanism of Lycium barbarum fruit cell wall polysaccharide remodeling reveals potential pectin accumulation contributors

The level of polysaccharides in the mature Lycium barbarum fruit (LBF) cell wall depends on their metabolism, trafficking, and reassembly within the cell. In this study, we examined the composition, content, and ultrastructure of the cell wall polysaccharides of LBF during maturation, and further analyzed cell wall polysaccharide remodeling using isotope tagging with relative and absolute quantification (iTRAQ)-based proteomics. The results showed that the contents of cellulose and hemicellulose tended to increase in the pre-maturation stage and decrease in the later stage, while pectin level increased before fruit maturing. The differential expression of the 54 proteins involved in the metabolic pathways for glucose, fructose, galactose, galacturonic acid and arabinose was found to be responsible for these alterations. The work provides a biological framework for the reorganization of polysaccharides in the LBF cell wall, and supports the hypothesis that pectic polysaccharide glycosyl donors come from starch, cellulose, hemicellulose and isomorphic pectin.

Phenomics and Robust Multiomics Data for Cardiovascular Disease Subtyping

The complex landscape of cardiovascular diseases encompasses a wide range of related pathologies arising from diverse molecular mechanisms and exhibiting heterogeneous phenotypes. This variety of manifestations poses significant challenges in the development of treatment strategies. The increasing availability of precise phenotypic and multiomics data of cardiovascular disease patient populations has spurred the development of a variety of computational disease subtyping techniques to identify distinct subgroups with unique underlying pathogeneses. In this review, we outline the essential components of computational approaches to select, integrate, and cluster omics and clinical data in the context of cardiovascular disease research. We delve into the challenges faced during different stages of the analysis, including feature selection and extraction, data integration, and clustering algorithms. Next, we highlight representative applications of subtyping pipelines in heart failure and coronary artery disease. Finally, we discuss the current challenges and future directions in the development of robust subtyping approaches that can be implemented in clinical workflows, ultimately contributing to the ongoing evolution of precision medicine in health care.

Protein biomarkers for radiation injury and testing of medical countermeasure efficacy: promises, pitfalls, and future directions

INTRODUCTION

Radiological/nuclear accidents, hostile military activity, or terrorist strikes have the potential to expose a large number of civilians and military personnel to high doses of radiation resulting in the development of acute radiation syndrome and delayed effects of exposure. Thus, there is an urgent need for sensitive and specific assays to assess the levels of radiation exposure to individuals. Such radiation exposures are expected to alter primary cellular proteomic processes, resulting in multifaceted biological responses.

AREAS COVERED

This article covers the application of proteomics, a promising and fast developing technology based on quantitative and qualitative measurements of protein molecules for possible rapid measurement of radiation exposure levels. Recent advancements in high-resolution chromatography, mass spectrometry, high-throughput, and bioinformatics have resulted in comprehensive (relative quantitation) and precise (absolute quantitation) approaches for the discovery and accuracy of key protein biomarkers of radiation exposure. Such proteome biomarkers might prove useful for assessing radiation exposure levels as well as for extrapolating the pharmaceutical dose of countermeasures for humans based on efficacy data generated using animal models.

EXPERT OPINION

The field of proteomics promises to be a valuable asset in evaluating levels of radiation exposure and characterizing radiation injury biomarkers.

A Multi-level Random Forest Model-Based Intrusion Detection Using Fuzzy Inference System for Internet of Things Networks

Intrusion detection (ID) methods are security frameworks designed to safeguard network information systems. The strength of an intrusion detection method is dependent on the robustness of the feature selection method. This study developed a multi-level random forest algorithm for intrusion detection using a fuzzy inference system. The strengths of the filter and wrapper approaches are combined in this work to create a more advanced multi-level feature selection technique, which strengthens network security. The first stage of the multi-level feature selection is the filter method using a correlation-based feature selection to select essential features based on the multi-collinearity in the data. The correlation-based feature selection used a genetic search method to choose the best features from the feature set. The genetic search algorithm assesses the merits of each attribute, which then delivers the characteristics with the highest fitness values for selection. A rule assessment has also been used to determine whether two feature subsets have the same fitness value, which ultimately returns the feature subset with the fewest features. The second stage is a wrapper method based on the sequential forward selection method to further select top features based on the accuracy of the baseline classifier. The selected top features serve as input into the random forest algorithm for detecting intrusions. Finally, fuzzy logic was used to classify intrusions as either normal, low, medium, or high to reduce misclassification. When the developed intrusion method was compared to other existing models using the same dataset, the results revealed a higher accuracy, precision, sensitivity, specificity, and F1-score of 99.46%, 99.46%, 99.46%, 93.86%, and 99.46%, respectively. The classification of attacks using the fuzzy inference system also indicates that the developed method can correctly classify attacks with reduced misclassification. The use of a multi-level feature selection method to leverage the advantages of filter and wrapper feature selection methods and fuzzy logic for intrusion classification makes this study unique.

Using Artificial Intelligence to Better Predict and Develop Biomarkers

Advancements in technology have improved biomarker discovery in the field of heart failure (HF). What was once a slow and laborious process has gained efficiency through use of high-throughput omics platforms to phenotype HF at the level of genes, transcripts, proteins, and metabolites. Furthermore, improvements in artificial intelligence (AI) have made the interpretation of large omics data sets easier and improved analysis. Use of omics and AI in biomarker discovery can aid clinicians by identifying markers of risk for developing HF, monitoring care, determining prognosis, and developing druggable targets. Combined, AI has the power to improve HF patient care.

Neurodegenerative disorders: From clinicopathology convergence to systems biology divergence

Neurodegenerative diseases are multifactorial. This means that several genetic, epigenetic, and environmental factors contribute to their emergence. Therefore, for the future management of these highly prevalent diseases, it is necessary to change perspective. If a holistic viewpoint is assumed, the phenotype (the clinicopathological convergence) emerges from the perturbation of a complex system of functional interactions among proteins (systems biology divergence). The systems biology top-down approach starts with the unbiased collection of sets of data generated through one or more -omics techniques and has the aim to identify the networks and the components that participate in the generation of a phenotype (disease), often without any available a priori knowledge. The principle behind the top-down method is that the molecular components that respond similarly to experimental perturbations are somehow functionally related. This allows the study of complex and relatively poorly characterized diseases without requiring extensive knowledge of the processes under investigation. In this chapter, the use of a global approach will be applied to the comprehension of neurodegeneration, with a particular focus on the two most prevalent ones, Alzheimer's and Parkinson's diseases. The final purpose is to distinguish disease subtypes (even with similar clinical manifestations) to launch a future of precision medicine for patients with these disorders.

Crosstalk between CD4+ T Cells and Airway Smooth Muscle in Pediatric Obesity-related Asthma

Rationale: Pediatric obesity-related asthma is a nonatopic asthma phenotype with high disease burden and few effective therapies. RhoGTPase upregulation in peripheral blood T helper (Th) cells is associated with the phenotype, but the mechanisms that underlie this association are not known. Objectives: To investigate the mechanisms by which upregulation of CDC42 (Cell Division Cycle 42), a RhoGTPase, in Th cells is associated with airway smooth muscle (ASM) biology. Methods: Chemotaxis of obese asthma and healthy-weight asthma Th cells, and their adhesion to obese and healthy-weight nonasthmatic ASM, was investigated. Transcriptomics and proteomics were used to determine the differential effect of obese and healthy-weight asthma Th cell adhesion to obese or healthy-weight ASM biology. Measurements and Main Results: Chemotaxis of obese asthma Th cells with CDC42 upregulation was resistant to CDC42 inhibition. Obese asthma Th cells were more adherent to obese ASM compared with healthy-weight asthma Th cells to healthy-weight ASM. Compared with coculture with healthy-weight ASM, obese asthma Th cell coculture with obese ASM was positively enriched for genes and proteins involved in actin cytoskeleton organization, transmembrane receptor protein kinase signaling, and cell mitosis, and negatively enriched for extracellular matrix organization. Targeted gene evaluation revealed upregulation of IFNG, TNF (tumor necrosis factor), and Cluster of Differentiation 247 (CD247) among Th cell genes, and of Ak strain transforming (AKT), Ras homolog family member A (RHOA), and CD38, with downregulation of PRKCA (Protein kinase C-alpha), among smooth muscle genes. Conclusions: Obese asthma Th cells have uninhibited chemotaxis and are more adherent to obese ASM, which is associated with upregulation of genes and proteins associated with smooth muscle proliferation and reciprocal nonatopic Th cell activation.

Proteomics data analysis using multiple statistical approaches identified proteins and metabolic networks associated with sucrose accumulation in sugarcane

Background

Sugarcane is the most important sugar crop, contributing > 80% of global sugar production. High sucrose content is a key target of sugarcane breeding, yet sucrose improvement in sugarcane remains extremely slow for decades. Molecular breeding has the potential to break through the genetic bottleneck of sucrose improvement. Dissecting the molecular mechanism(s) and identifying the key genetic elements controlling sucrose accumulation will accelerate sucrose improvement by molecular breeding. In our previous work, a proteomics dataset based on 12 independent samples from high- and low-sugar genotypes treated with ethephon or water was established. However, in that study, employing conventional analysis, only 25 proteins involved in sugar metabolism were identified .

Results

In this work, the proteomics dataset used in our previous study was reanalyzed by three different statistical approaches, which include a logistic marginal regression, a penalized multiple logistic regression named Elastic net, as well as a Bayesian multiple logistic regression method named Stochastic search variable selection (SSVS) to identify more sugar metabolism-associated proteins. A total of 507 differentially abundant proteins (DAPs) were identified from this dataset, with 5 of them were validated by western blot. Among the DAPs, 49 proteins were found to participate in sugar metabolism-related processes including photosynthesis, carbon fixation as well as carbon, amino sugar, nucleotide sugar, starch and sucrose metabolism. Based on our studies, a putative network of key proteins regulating sucrose accumulation in sugarcane is proposed, with glucose-6-phosphate isomerase, 2-phospho-D-glycerate hydrolyase, malate dehydrogenase and phospho-glycerate kinase, as hub proteins.

Conclusions

The sugar metabolism-related proteins identified in this work are potential candidates for sucrose improvement by molecular breeding. Further, this work provides an alternative solution for omics data processing.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-022-08768-2.

Comparison of grain protein profiles of Brazilian cowpea (Vigna unguiculata) cultivars based on principal component analysis

This study aims to compare the grain protein profile of four Brazilian cowpea cultivars (BRS Aracê, BRS Itaim, BRS Pajeú, and BRS Xiquexique) by two-dimensional electrophoresis (2-DE) and principal component analysis (PCA). 2-DE efficiently separate cowpea protein profiles, showing high homogeneity among the four cultivars. In addition, the principal component analysis indicated that there is a difference in abundance of proteins among the cultivars. The cultivars BRS Aracê and BRS Xiquexique, both biofortified in iron and zinc, were separated from the cultivars BRS Itaim and BRS Pajeú. These results demonstrate that protein profiles can be used to discriminate cowpea varieties.

Graphical Abstract

Proteomics Profiling of Stool Samples from Preterm Neonates with SWATH/DIA Mass Spectrometry for Predicting Necrotizing Enterocolitis

Necrotizing enterocolitis (NEC) is a life-threatening condition for premature infants in neonatal intensive care units. Finding indicators that can predict NEC development before symptoms appear would provide more time to apply targeted interventions. In this study, stools from 132 very-low-birth-weight (VLBW) infants were collected daily in the context of a multi-center prospective study aimed at investigating the potential of fecal biomarkers for NEC prediction using proteomics technology. Eight of the VLBW infants received a stage-3 NEC diagnosis. Stools collected from the NEC infants up to 10 days before their diagnosis were available for seven of them. Their samples were matched with those from seven pairs of non-NEC controls. The samples were processed for liquid chromatography-tandem mass spectrometry analysis using SWATH/DIA acquisition and cross-compatible proteomic software to perform label-free quantification. ROC curve and principal component analyses were used to explore discriminating information and to evaluate candidate protein markers. A series of 36 proteins showed the most efficient capacity with a signature that predicted all seven NEC infants at least a week in advance. Overall, our study demonstrates that multiplexed proteomic signature detection constitutes a promising approach for the early detection of NEC development in premature infants.

Current Insights on Neurodegeneration by the Italian Proteomics Community

The growing number of patients affected by neurodegenerative disorders represents a huge problem for healthcare systems, human society, and economics. In this context, omics strategies are crucial for the identification of molecular factors involved in disease pathobiology, and for the discovery of biomarkers that allow early diagnosis, patients’ stratification, and treatment response prediction. The integration of different omics data is a required step towards the goal of personalized medicine. The Italian proteomics community is actively developing and applying proteomics approaches to the study of neurodegenerative disorders; moreover, it is leading the mitochondria-focused initiative of the Human Proteome Project, which is particularly important given the central role of mitochondrial impairment in neurodegeneration. Here, we describe how Italian research groups in proteomics have contributed to the knowledge of many neurodegenerative diseases, through the elucidation of the pathobiology of these disorders, and through the discovery of disease biomarkers. In particular, we focus on the central role of post-translational modifications analysis, the implementation of network-based approaches in functional proteomics, the integration of different omics in a systems biology view, and the development of novel platforms for biomarker discovery for the high-throughput quantification of thousands of proteins at a time.

Using Artificial Intelligence to Better Predict and Develop Biomarkers

Advancements in technology have improved biomarker discovery in the field of heart failure (HF). What was once a slow and laborious process has gained efficiency through use of high-throughput omics platforms to phenotype HF at the level of genes, transcripts, proteins, and metabolites. Furthermore, improvements in artificial intelligence (AI) have made the interpretation of large omics data sets easier and improved analysis. Use of omics and AI in biomarker discovery can aid clinicians by identifying markers of risk for developing HF, monitoring care, determining prognosis, and developing druggable targets. Combined, AI has the power to improve HF patient care.

POSREG: proteomic signature discovered by simultaneously optimizing its reproducibility and generalizability

Mass spectrometry-based proteomic technique has become indispensable in current exploration of complex and dynamic biological processes. Instrument development has largely ensured the effective production of proteomic data, which necessitates commensurate advances in statistical framework to discover the optimal proteomic signature. Current framework mainly emphasizes the generalizability of the identified signature in predicting the independent data but neglects the reproducibility among signatures identified from independently repeated trials on different sub-dataset. These problems seriously restricted the wide application of the proteomic technique in molecular biology and other related directions. Thus, it is crucial to enable the generalizable and reproducible discovery of the proteomic signature with the subsequent indication of phenotype association. However, no such tool has been developed and available yet. Herein, an online tool, POSREG, was therefore constructed to identify the optimal signature for a set of proteomic data. It works by (i) identifying the proteomic signature of good reproducibility and aggregating them to ensemble feature ranking by ensemble learning, (ii) assessing the generalizability of ensemble feature ranking to acquire the optimal signature and (iii) indicating the phenotype association of discovered signature. POSREG is unique in its capacity of discovering the proteomic signature by simultaneously optimizing its reproducibility and generalizability. It is now accessible free of charge without any registration or login requirement at https://idrblab.org/posreg/.

Looking at COVID-19 from a Systems Biology Perspective

The sudden outbreak and worldwide spread of the SARS-CoV-2 pandemic pushed the scientific community to find fast solutions to cope with the health emergency. COVID-19 complexity, in terms of clinical outcomes, severity, and response to therapy suggested the use of multifactorial strategies, characteristic of the network medicine, to approach the study of the pathobiology. Proteomics and interactomics especially allow to generate datasets that, reduced and represented in the forms of networks, can be analyzed with the tools of systems biology to unveil specific pathways central to virus-human host interaction. Moreover, artificial intelligence tools can be implemented for the identification of druggable targets and drug repurposing. In this review article, we provide an overview of the results obtained so far, from a systems biology perspective, in the understanding of COVID-19 pathobiology and virus-host interactions, and in the development of disease classifiers and tools for drug repurposing.

Mass-Spectrometry-Based Functional Proteomic and Phosphoproteomic Technologies and Their Application for Analyzing Ex Vivo and In Vitro Models of Hypertrophic Cardiomyopathy

Hypertrophic cardiomyopathy (HCM) is an autosomal dominant disease thought to be principally caused by mutations in sarcomeric proteins. Despite extensive genetic analysis, there are no comprehensive molecular frameworks for how single mutations in contractile proteins result in the diverse assortment of cellular, phenotypic, and pathobiological cascades seen in HCM. Molecular profiling and system biology approaches are powerful tools for elucidating, quantifying, and interpreting dynamic signaling pathways and differential macromolecule expression profiles for a wide range of sample types, including cardiomyopathy. Cutting-edge approaches combine high-performance analytical instrumentation (e.g., mass spectrometry) with computational methods (e.g., bioinformatics) to study the comparative activity of biochemical pathways based on relative abundances of functionally linked proteins of interest. Cardiac research is poised to benefit enormously from the application of this toolkit to cardiac tissue models, which recapitulate key aspects of pathogenesis. In this review, we evaluate state-of-the-art mass-spectrometry-based proteomic and phosphoproteomic technologies and their application to in vitro and ex vivo models of HCM for global mapping of macromolecular alterations driving disease progression, emphasizing their potential for defining the components of basic biological systems, the fundamental mechanistic basis of HCM pathogenesis, and treating the ensuing varied clinical outcomes seen among affected patient cohorts.

Looking deeper into ocular surface health: an introduction to clinical tear proteomics analysis

Ocular surface diseases are becoming more prevalent worldwide. Reasons for this include the ongoing population ageing and increasing use of digital displays, although ophthalmologists have a wide selection of tools, which can be implemented in the evaluation of the ocular surface health, methods, which enable the in-depth study of biological functions are gaining more interest. These new approaches are needed, since the individual responses to ocular surface diseases and treatments can vary from person to person, and the correlations between clinical signs and symptoms are often low. Modern mass spectrometry (MS) methods can produce information on hundreds of tear proteins, which in turn can provide valuable information on the biological effects occurring on the ocular surface. In this review article, we will provide an overview of the different aspects, which are part of a successful tear proteomics study design and equip readers with a better understanding of the methods most suited for their MS-based tear proteomics study in the field of ophthalmology and ocular surface.

Effect of Transgenic Rootstock Grafting on the Omics Profiles in Tomato

Grafting of non-transgenic scion onto genetically modified (GM) rootstocks provides superior agronomic traits in the GM rootstock, and excellent fruits can be produced for consumption. In such grafted plants, the scion does not contain any foreign genes, but the fruit itself is likely to be influenced directly or indirectly by the foreign genes in the rootstock. Before market release of such fruit products, the effects of grafting onto GM rootstocks should be determined from the perspective of safety use. Here, we evaluated the effects of a transgene encoding β-glucuronidase (GUS) on the grafted tomato fruits as a model case. An edible tomato cultivar, Stella Mini Tomato, was grafted onto GM Micro-Tom tomato plants that had been transformed with the GUS gene. The grafted plants showed no difference in their fruit development rate and fresh weight regardless of the presence or absence of the GUS gene in the rootstock. The fruit samples were subjected to transcriptome (NGS-illumina), proteome (shotgun LC-MS/MS), metabolome (LC-ESI-MS and GC-EI-MS), and general food ingredient analyses. In addition, differentially detected items were identified between the grafted plants onto rootstocks with or without transgenes (more than two-fold). The transcriptome analysis detected approximately 18,500 expressed genes on average, and only 6 genes were identified as differentially expressed. Principal component analysis of 2,442 peaks for peptides in proteome profiles showed no significant differences. In the LC-ESI-MS and GC-EI-MS analyses, a total of 93 peak groups and 114 peak groups were identified, respectively, and only 2 peak groups showed more than two-fold differences. The general food ingredient analysis showed no significant differences in the fruits of Stella scions between GM and non-GM Micro-Tom rootstocks. These multiple omics data showed that grafting on the rootstock harboring the GUS transgene did not induce any genetic or metabolic variation in the scion.

Proteomics of Cellular Response to Stress: Taking Control of False Positive Results

One of the main goals of quantitative proteomics is molecular profiling of cellular response to stress at the protein level. To perform this profiling, statistical analysis of experimental data involves multiple testing of a hypothesis about the equality of protein concentrations between the cells under normal and stress conditions. This analysis is then associated with the multiple testing problem dealing with the increased chance of obtaining false positive results. A number of solutions to this problem are known, yet, they may lead to the loss of potentially important biological information when applied with commonly accepted thresholds of statistical significance. Using the proteomic data obtained earlier for the yeast samples containing proteins at known concentrations and the biological models of early and late cellular responses to stress, we analyzed dependences of distributions of false positive and false negative rates on the protein fold changes and thresholds of statistical significance. Based on the analysis of the density of data points in the volcano plots, Benjamini-Hochberg method, and gene ontology analysis, visual approach for optimization of the statistical threshold and selection of the differentially regulated proteins has been suggested, which could be useful for researchers working in the field of quantitative proteomics.

Proteomics and plant biology: contributions to date and a look towards the next decade

INTRODUCTION

This review presents the view of the author, that is opinionable and even speculative, on the field of proteomics, its application to plant biology knowledge, and translation to biotechnology. Written in a more academic than scientific style, it is based on past original and review articles by the author´s group, and those published by leading scientists in the last two years.

AREAS COVERED

Starting with a general definition and references to historical milestones, it covers sections devoted to the different platforms employed, the plant biology discourse in the protein language, challenges and future prospects, ending with the author opinion.

EXPERT OPINION

In 25 years, five proteomics platform generations have appeared. We are now moving from proteomics to Systems Biology. While feasible with model organisms, proteomics of orphan species remains challenging. Proteomics, even in its simplest approach, sheds light on plant biological processes, central dogma, and molecular bases of phenotypes of interest, and it can be translated to areas such as food traceability and allergen detection. Proteomics should be validated and optimized to each experimental system, objectives, and hypothesis. It has limitations, artifacts, and biases. We should not blindly accept proteomics data and just create a list of proteins, networks, and avoid speculative biological interpretations. From the hundred to thousand proteins identified and quantified, it is important to obtain a focus and validate some of them, otherwise it is merely. We are starting to have the protein pieces, so let, from now, build the proteomics and biological puzzle.

Susceptibility mapping of groundwater salinity using machine learning models

Increasing groundwater salinity has recently raised severe environmental and health concerns around the world. Advancement of the novel methods for spatial salinity modeling and prediction would be essential for effective management of the resources and planning mitigation policies. The current research presents the application of machine learning (ML) models in groundwater salinity mapping based on the dichotomous predictions. The groundwater salinity is predicted using the essential factors (i.e., identified by the simulated annealing feature selection methodology) through k-fold cross-validation methodology. Six ML models, namely, flexible discriminant analysis (FDA), mixture discriminant analysis (MAD), boosted regression tree (BRT), multivariate adaptive regression spline (MARS), random forest (RF), support vector machine (SVM), were employed to groundwater salinity mapping. The results of the modeling indicated that the SVM model had superior performance than other models. Variables of soil order, groundwater withdrawal, precipitation, land use, and elevation had the most contribute to groundwater salinity mapping. Results highlighted that the southern parts of the region and some parts in the north, northeast, and west have a high groundwater salinity, in which these areas are mostly matched with soil order of Entisols, bareland areas, and low elevations.

Advances in Current Diabetes Proteomics: From the Perspectives of Label- free Quantification and Biomarker Selection

BACKGROUND

Due to its prevalence and negative impacts on both the economy and society, the diabetes mellitus (DM) has emerged as a worldwide concern. In light of this, the label-free quantification (LFQ) proteomics and diabetic marker selection methods have been applied to elucidate the underlying mechanisms associated with insulin resistance, explore novel protein biomarkers, and discover innovative therapeutic protein targets.

OBJECTIVE

The purpose of this manuscript is to review and analyze the recent computational advances and development of label-free quantification and diabetic marker selection in diabetes proteomics.

METHODS

Web of Science database, PubMed database and Google Scholar were utilized for searching label-free quantification, computational advances, feature selection and diabetes proteomics.

RESULTS

In this study, we systematically review the computational advances of label-free quantification and diabetic marker selection methods which were applied to get the understanding of DM pathological mechanisms. Firstly, different popular quantification measurements and proteomic quantification software tools which have been applied to the diabetes studies are comprehensively discussed. Secondly, a number of popular manipulation methods including transformation, pretreatment (centering, scaling, and normalization), missing value imputation methods and a variety of popular feature selection techniques applied to diabetes proteomic data are overviewed with objective evaluation on their advantages and disadvantages. Finally, the guidelines for the efficient use of the computationbased LFQ technology and feature selection methods in diabetes proteomics are proposed.

CONCLUSION

In summary, this review provides guidelines for researchers who will engage in proteomics biomarker discovery and by properly applying these proteomic computational advances, more reliable therapeutic targets will be found in the field of diabetes mellitus.

Integration of Proteomics and Other Omics Data

In recent biomedical studies, multidimensional profiling, which collects proteomics as well as other types of omics data on the same subjects, is getting increasingly popular. Proteomics, transcriptomics, genomics, epigenomics, and other types of data contain overlapping as well as independent information, which suggests the possibility of integrating multiple types of data to generate more reliable findings/models with better classification/prediction performance. In this chapter, a selective review is conducted on recent data integration techniques for both unsupervised and supervised analysis. The main objective is to provide the "big picture" of data integration that involves proteomics data and discuss the "intuition" beneath the recently developed approaches without invoking too many mathematical details. Potential pitfalls and possible directions for future developments are also discussed.

Multivariate Analysis Applications in X-ray Diffraction

Multivariate analysis (MA) is becoming a fundamental tool for processing in an efficient way the large amount of data collected in X-ray diffraction experiments. Multi-wedge data collections can increase the data quality in case of tiny protein crystals; in situ or operando setups allow investigating changes on powder samples occurring during repeated fast measurements; pump and probe experiments at X-ray free-electron laser (XFEL) sources supply structural characterization of fast photo-excitation processes. In all these cases, MA can facilitate the extraction of relevant information hidden in data, disclosing the possibility of automatic data processing even in absence of a priori structural knowledge. MA methods recently used in the field of X-ray diffraction are here reviewed and described, giving hints about theoretical background and possible applications. The use of MA in the framework of the modulated enhanced diffraction technique is described in detail.

Finding a Needle in a Haystack: Proteomics in Heart Failure

•

Proteomics has aided HF biomarker discovery, which allows for greater disease insights.

•

Experiment design can be tailored to HF research to discover novel biomarkers.

•

Primary methods include MS, protein microarray, aptamer, and PEA-based technologies.

•

Proteomics can detect unique low abundance proteins and detect protein modifications.

Circulating protein biomarkers provide information regarding pathways in heart failure (HF) and can add important value to clinicians. Advancements in proteomics allow researchers to measure a multitude of proteins simultaneously with excellent sensitivity and selectivity to detect low abundance proteins. This helps identify previously unrecognized pathways in HF and discover biomarkers and potential targets for HF therapies. Although several proteomic methods exist, including mass spectrometry, protein microarray, aptamer, and proximity extension assay−based techniques, each have their unique advantages. This paper provides an overview of the various proteomic methods, with examples of how each has contributed to understanding the pathways in HF.

The Promises and the Challenges of Integrating Multi-Omics and Systems Biology in Comparative Stress Biology

Comparative stress biology is inherently a systems biology approach with the goal of integrating the molecular, cellular, and physiological responses with fitness outcomes. In this way, the systems biology approach is expected to provide a holistic understanding of how different stressors result in different fitness outcomes, and how different individuals (or populations or species) respond to stressors differently. In this perceptive article, I focus on the use of multiple types of -omics data in stress biology. Targeting students and those researchers who are considering integrating -omics approaches in their comparative stress biology studies, I discuss the promise of the integration of these measures for furthering our holistic understanding of how organisms respond to different stressors. I also discuss the logistical and conceptual challenges encountered when working with -omics data and the current hurdles to fully utilize these data in studies of stress biology in non-model organisms.

Proteostasis and Proteotoxicity in the Network Medicine Era

Neurodegenerative proteinopathies are complex diseases that share some pathogenetic processes. One of these is the failure of the proteostasis network (PN), which includes all components involved in the synthesis, folding, and degradation of proteins, thus leading to the aberrant accumulation of toxic protein aggregates in neurons. The single components that belong to the three main modules of the PN are highly interconnected and can be considered as part of a single giant network. Several pharmacological strategies have been proposed to ameliorate neurodegeneration by targeting PN components. Nevertheless, effective disease-modifying therapies are still lacking. In this review article, after a general description of the PN and its failure in proteinopathies, we will focus on the available pharmacological tools to target proteostasis. In this context, we will discuss the main advantages of systems-based pharmacology in contrast to the classical targeted approach, by focusing on network pharmacology as a strategy to innovate rational drug design.

Proteomic Signatures During Treatment in Different Stages of Heart Failure

Background:

Proteomics have already provided novel insights into the pathophysiology of heart failure (HF) with reduced ejection fraction. Previous studies have evaluated cross-sectional protein signatures of HF, but few have characterized proteomic changes following HF with reduced ejection fraction treatment with ARNI (angiotensin receptor/neprilysin inhibitor) therapy or left ventricular assist devices.

Methods:

In this retrospective omics study, we performed targeted proteomics (N=625) of whole blood sera from patients with American College of Cardiology/American Heart Association stage D (N=29) and stage C (N=12) HF using proximity extension assays. Samples were obtained before and after (median=82 days) left ventricular assist device implantation (stage D; primary analysis) and ARNI therapy initiation (stage C; matched reference). Oblique principal component analysis and point biserial correlations were used for feature extraction and selection; standardized mean differences were used to assess within and between-group differences; and enrichment analysis was used to generate and cluster Gene Ontology terms.

Results:

Core sets of proteins were identified for stage C (N=9 proteins) and stage D (N=18) HF; additionally, a core set of 5 shared HF proteins (NT-proBNP [N-terminal pro-B type natriuretic peptide], ESM [endothelial cell-specific molecule]-1, cathepsin L1, osteopontin, and MCSF-1) was also identified. For patients with stage D HF, moderate (δ, 0.40–0.60) and moderate-to-large (δ, 0.60–0.80) sized differences were observed in 8 of their 18 core proteins after left ventricular assist devices implantation. Additionally, specific protein groups reached concentration levels equivalent ( g <0.10) to stage C HF after initiation on ARNI therapy.

Conclusions:

HF with reduced ejection fraction severity associates with distinct proteomic signatures that reflect underlying disease attributes; these core signatures may be useful for monitoring changes in cardiac function following initiation on ARNI or left ventricular assist device implantation.

Computational advances of tumor marker selection and sample classification in cancer proteomics

Cancer proteomics has become a powerful technique for characterizing the protein markers driving transformation of malignancy, tracing proteome variation triggered by therapeutics, and discovering the novel targets and drugs for the treatment of oncologic diseases. To facilitate cancer diagnosis/prognosis and accelerate drug target discovery, a variety of methods for tumor marker identification and sample classification have been developed and successfully applied to cancer proteomic studies. This review article describes the most recent advances in those various approaches together with their current applications in cancer-related studies. Firstly, a number of popular feature selection methods are overviewed with objective evaluation on their advantages and disadvantages. Secondly, these methods are grouped into three major classes based on their underlying algorithms. Finally, a variety of sample separation algorithms are discussed. This review provides a comprehensive overview of the advances on tumor maker identification and patients/samples/tissues separations, which could be guidance to the researches in cancer proteomics.

Biomarker development for axial spondyloarthritis

The term axial spondyloarthritis (axSpA) encompasses a heterogeneous group of diseases that have variable presentations, extra-articular manifestations and clinical outcomes, and that will respond differently to treatments. The prototypical type of axSpA, ankylosing spondylitis, is thought to be caused by interaction between the genetically primed host immune system and gut microbiota. Currently used biomarkers such as HLA-B27 status, C-reactive protein and erythrocyte sedimentation rate have, at best, moderate diagnostic and predictive value. Improved biomarkers are needed for axSpA to assist with early diagnosis and to better predict treatment responses and long-term outcomes. Advances in a range of 'omics' technologies and statistical approaches, including genomics approaches (such as polygenic risk scores), microbiome profiling and, potentially, transcriptomic, proteomic and metabolomic profiling, are making it possible for more informative biomarker sets to be developed for use in such clinical applications. Future developments in this field will probably involve combinations of biomarkers that require novel statistical approaches to analyse and to produce easy to interpret metrics for clinical application. Large publicly available datasets from well-characterized case-cohort studies that use extensive biological sampling, particularly focusing on early disease and responses to medications, are required to establish successful biomarker discovery and validation programmes.

MetaFS: Performance assessment of biomarker discovery in metaproteomics

Metaproteomics suffers from the issues of dimensionality and sparsity. Data reduction methods can maximally identify the relevant subset of significant differential features and reduce data redundancy. Feature selection (FS) methods were applied to obtain the significant differential subset. So far, a variety of feature selection methods have been developed for metaproteomic study. However, due to FS's performance depended heavily on the data characteristics of a given research, the well-suitable feature selection method must be carefully selected to obtain the reproducible differential proteins. Moreover, it is critical to evaluate the performance of each FS method according to comprehensive criteria, because the single criterion is not sufficient to reflect the overall performance of the FS method. Therefore, we developed an online tool named MetaFS, which provided 13 types of FS methods and conducted the comprehensive evaluation on the complex FS methods using four widely accepted and independent criteria. Furthermore, the function and reliability of MetaFS were systematically tested and validated via two case studies. In sum, MetaFS could be a distinguished tool for discovering the overall well-performed FS method for selecting the potential biomarkers in microbiome studies. The online tool is freely available at https://idrblab.org/metafs/.

Study Protocol for a Prospective Longitudinal Cohort Study to Identify Proteomic Predictors of Pluripotent Risk for Mental Illness: The Seoul Pluripotent Risk for Mental Illness Study

BACKGROUND

The Seoul Pluripotent Risk for Mental Illness (SPRIM) study was designed to identify predictors leading to mental illness in help-seeking individuals by securing sufficient statistical power through transdiagnostic approaches. The SPRIM study aims to examine the clinical characteristics of high-risk individuals for mental illness and to identify proteomic biomarkers that can predict the onset of mental illness.

METHODS

This paper describes the study protocol of the SPRIM study. We aim to recruit 150 participants who meet the criteria for high risk for major mental illness, 150 patients with major psychiatric disorders (schizophrenia, bipolar disorder, and major depressive disorder), and 50 matched healthy control subjects for 2 years. Clinical evaluations, self-report measures, and proteomic analyses will be implemented. The assessment points are at baseline, 6, 12, 18, and 24 months.

CONCLUSIONS

In the present study, we introduced the study protocol of the SPRIM study, which is the first prospective cohort study of transdiagnostic high-risk concepts using proteomic biomarkers. This study has a paradigm that encompasses various diseases without aiming at predicting and preventing the development of a specific mental illness in help-seeking individuals. The transdiagnostic high-risk concept could be extended to provide a perspective for people with various psychopathological tendencies below a threshold, such that they do not meet the existing diagnostic criteria of mental illnesses, to determine what may lead them to a specific disease and help identify appropriate preventative interventions.

Feature engineering applied to intraoperative in vivo Raman spectroscopy sheds light on molecular processes in brain cancer: a retrospective study of 65 patients

Raman spectroscopy is a promising tool for neurosurgical guidance and cancer research. Quantitative analysis of the Raman signal from living tissues is, however, limited. Their molecular composition is convoluted and influenced by clinical factors, and access to data is limited. To ensure acceptance of this technology by clinicians and cancer scientists, we need to adapt the analytical methods to more closely model the Raman-generating process. Our objective is to use feature engineering to develop a new representation for spectral data specifically tailored for brain diagnosis that improves interpretability of the Raman signal while retaining enough information to accurately predict tissue content. The method consists of band fitting of Raman bands which consistently appear in the brain Raman literature, and the generation of new features representing the pairwise interaction between bands and the interaction between bands and patient age. Our technique was applied to a dataset of 547 in situ Raman spectra from 65 patients undergoing glioma resection. It showed superior predictive capacities to a principal component analysis dimensionality reduction. After analysis through a Bayesian framework, we were able to identify the oncogenic processes that characterize glioma: increased nucleic acid content, overexpression of type IV collagen and shift in the primary metabolic engine. Our results demonstrate how this mathematical transformation of the Raman signal allows the first biological, statistically robust analysis of in vivo Raman spectra from brain tissue.

Spatial hazard assessment of the PM10 using machine learning models in Barcelona, Spain

Air pollution, and especially atmospheric particulate matter (PM), has a profound impact on human mortality and morbidity, environment, and ecological system. Accordingly, it is very relevant predicting air quality. Although the application of the machine learning (ML) models for predicting air quality parameters, such as PM concentrations, has been evaluated in previous studies, those on the spatial hazard modeling of them are very limited. Due to the high potential of the ML models, the spatial modeling of PM can help managers to identify the pollution hotspots. Accordingly, this study aims at developing new ML models, such as Random Forest (RF), Bagged Classification and Regression Trees (Bagged CART), and Mixture Discriminate Analysis (MDA) for the hazard prediction of PM10 (particles with a diameter less than 10 µm) in the Barcelona Province, Spain. According to the annual PM10 concentration in 75 stations, the healthy and unhealthy locations are determined, and a ratio 70/30 (53/22 stations) is applied for calibrating and validating the ML models to predict the most hazardous areas for PM10. In order to identify the influential variables of PM modeling, the simulated annealing (SA) feature selection method is used. Seven features, among the thirteen features, are selected as critical features. According to the results, all the three-machine learning (ML) models achieve an excellent performance (Accuracy > 87% and precision > 86%). However, the Bagged CART and RF models have the same performance and higher than the MDA model. Spatial hazard maps predicted by the three models indicate that the high hazardous areas are located in the middle of the Barcelona Province more than in the Barcelona's Metropolitan Area.

Novel Hyperthermophilic Crenarchaeon Thermofilum adornatum sp. nov. Uses GH1, GH3, and Two Novel Glycosidases for Cellulose Hydrolysis

A novel hyperthermophilic, anaerobic filamentous archaeon, Thermofilum adornatum strain 1910bT, is capable of growing with cellulose as its sole carbon and energy source. This strain was isolated from a terrestrial hot spring in Kamchatka, Russia. The isolate 1910bT grew optimally at a temperature of 80°C and a pH of 5.5-6.0, producing cell-bound inducible cellulases. During genome analysis, genes, encoding various glycosidases (GHs) involved in oligo- and polysaccharide hydrolysis and genes for the fermentation of sugars were identified. No homologs of currently known cellulase families were found among the GHs encoded by the 1910bT genome, suggesting that novel proteins are involved. To figure this out, a proteomic analysis of cells grown on cellulose or pyruvate (as a control) was performed. Both in-depth genomic and proteomic analyses revealed four proteins (Cel25, Cel30, Cel40, and Cel45) that were the most likely to be involved in the cellulose hydrolysis in this archaeon. Two of these proteins (Cel30 and Cel45) were hypothetical according to genome analysis, while the other two (Cel25 and Cel40) have GH3 and GH1 domains, respectively. The respective genes were heterologously expressed in Escherichia coli BL21 (DE3), and enzymatic activities of recombinant proteins were measured with carboxymethyl cellulose (CMC), Avicel and cellobiose as substrates. It was revealed that the Cel30 and Cel25 proteins were likely exoglucanases with side beta-glucosidase and endoglucanase activities, that Cel40 was a multifunctional glucanase capable of hydrolyzing beta-1,4-glucosides of various lengths, and that Cel45 was an endoglucanase with side exoglucanase activity. Taking into account that the cellulolytic activity of T. adornatum 1910bT surface protein fractions was inducible, that recombinant Cel25 and Cel30 were much less active than Cel40 and Cel45, and that their gene expressions were (almost) non-induced by CMC, we suggest that Cel40 and Cel45 play a major role in the degradation of cellulose, while Cel25 and Cel30 act only as accessory enzymes.

Proteomic Changes in Human Sperm During Sequential in vitro Capacitation and Acrosome Reaction

The male gamete is not completely mature after ejaculation and requires further events in the female genital tract to acquire fertilizing ability, including the processes of capacitation and acrosome reaction. In order to shed light on protein changes experienced by the sperm cell in preparation for fertilization, a comprehensive quantitative proteomic profiling based on isotopic peptide labeling and liquid chromatography followed by tandem mass spectrometry was performed on spermatozoa from three donors of proven fertility under three sequential conditions: purification with density gradient centrifugation, incubation with capacitation medium, and induction of acrosome reaction by exposure to the calcium ionophore A23187. After applying strict selection criteria for peptide quantification and for statistical analyses, 36 proteins with significant changes in their relative abundance within sperm protein extracts were detected. Moreover, the presence of peptide residues potentially harboring sites for post-translational modification was revealed, suggesting that protein modification may be an important mechanism in sperm maturation. In this regard, increased levels of proteins mainly involved in motility and signaling, both regulated by protein modifiers, were detected in sperm lysates following incubation with capacitation medium. In contrast, less abundant proteins in acrosome-reacted cell lysates did not contain potentially modifiable residues, suggesting the possibility that all those proteins might be relocated or released during the process. Protein-protein interaction analysis revealed a subset of proteins potentially involved in sperm maturation, including the proteins Erlin-2 (ERLIN2), Gamma-glutamyl hydrolase (GGH) and Transmembrane emp24 domain-containing protein 10 (TMED10). These results contribute to the current knowledge of the molecular basis of human fertilization. It should now be possible to further validate the potential role of the detected altered proteins as modulators of male infertility.

Proteomics: a powerful tool to study plant responses to biotic stress

In recent years, mass spectrometry-based proteomics has provided scientists with the tremendous capability to study plants more precisely than previously possible. Currently, proteomics has been transformed from an isolated field into a comprehensive tool for biological research that can be used to explain biological functions. Several studies have successfully used the power of proteomics as a discovery tool to uncover plant resistance mechanisms. There is growing evidence that indicates that the spatial proteome and post-translational modifications (PTMs) of proteins directly participate in the plant immune response. Therefore, understanding the subcellular localization and PTMs of proteins is crucial for a comprehensive understanding of plant responses to biotic stress. In this review, we discuss current approaches to plant proteomics that use mass spectrometry, with particular emphasis on the application of spatial proteomics and PTMs. The purpose of this paper is to investigate the current status of the field, discuss recent research challenges, and encourage the application of proteomics techniques to further research.

Exploring the Mitochondrial Degradome by the TAILS Proteomics Approach in a Cellular Model of Parkinson's Disease

Parkinson’s disease (PD) is the second most frequent neurodegenerative disease worldwide and the availability of early biomarkers and novel biotargets represents an urgent medical need. The main pathogenetic hallmark of PD is the specific loss of nigral dopaminergic neurons, in which mitochondrial dysfunction plays a crucial role. Mitochondrial proteases are central to the maintenance of healthy mitochondria and they have recently emerged as drug targets. However, an exhaustive characterization of these enzymes and their targets is still lacking, due to difficulties in analyzing proteolytic fragments by bottom-up proteomics approaches. Here, we propose the “mitochondrial dimethylation-TAILS” strategy, which combines the isolation of mitochondria with the enrichment of N-terminal peptides to analyze the mitochondrial N-terminome. We applied this method in a cellular model of altered dopamine homeostasis in neuroblastoma SH-SY5Y cells, which recapitulates early steps of PD pathogenesis. The main aim was to identify candidate mitochondrial proteases aberrantly activated by dopamine dysregulation and their cleaved targets. The proposed degradomics workflow was able to improve the identification of mitochondrial proteins if compared to classical shotgun analysis. In detail, 40% coverage of the mitochondrial proteome was obtained, the sequences of the transit peptides of two mitochondrial proteins were unveiled, and a consensus cleavage sequence for proteases involved in the processing of mitochondrial proteins was depicted. Mass spectrometry proteomics data have been submitted to ProteomeXchange with the identifier PXD013900. Moreover, sixty-one N-terminal peptides whose levels were affected by dopamine treatment were identified. By an in-depth analysis of the proteolytic peptides included in this list, eleven mitochondrial proteins showed altered proteolytic processing. One of these proteins (i.e., the 39S ribosomal protein L49 – MRPL49) was cleaved by the neprilysin protease, already exploited in clinics as a biotarget. We eventually demonstrated a mitochondrial subcellular localization of neprilysin in human cells for the first time. Collectively, these results shed new light on mitochondrial dysfunction linked to dopamine imbalance in PD and opened up the possibility to explore the mitochondrial targets of neprilysin as candidate biomarkers.

AI-based applications in hybrid imaging: how to build smart and truly multi-parametric decision models for radiomics

INTRODUCTION

The quantitative imaging features (radiomics) that can be obtained from the different modalities of current-generation hybrid imaging can give complementary information with regard to the tumour environment, as they measure different morphologic and functional imaging properties. These multi-parametric image descriptors can be combined with artificial intelligence applications into predictive models. It is now the time for hybrid PET/CT and PET/MRI to take the advantage offered by radiomics to assess the added clinical benefit of using multi-parametric models for the personalized diagnosis and prognosis of different disease phenotypes.

OBJECTIVE

The aim of the paper is to provide an overview of current challenges and available solutions to translate radiomics into hybrid PET-CT and PET-MRI imaging for a smart and truly multi-parametric decision model.