Glycomics meets artificial intelligence - Potential of glycan analysis for identification of seropositive and seronegative rheumatoid arthritis patients revealed

High-throughput N-glycan analysis in aging and inflammaging: State of the art and future directions

As the global population ages at an unprecedented rate, the prevalence of age-related diseases is increasing, making inflammaging - a phenomenon characterized by a chronic, low-grade inflammatory state that follows aging - a significant concern. Understanding the mechanisms of inflammaging and its impact on health is critical for developing strategies to improve the quality of life and manage health in the aging population. Despite their crucial roles in various biological processes, including immune response modulation, N-glycans, oligosaccharides covalently attached to many proteins, are often overlooked in clinical and research studies. This repeated oversight is largely due to their inherent complexity and the complexity of the analysis methods. High-throughput N-glycan analysis has emerged as a transformative tool in N-glycosylation research, enabling cost- and time-effective, detailed, and large-scale examination of N-glycan profiles. This paper is the first to explore the application of high-throughput N-glycomics techniques to investigate the complex interplay between N-glycosylation and the immune system in aging. Technological advancements have significantly improved Nglycan detection and characterization, providing insights into age-related changes in Nglycosylation. Key findings highlight consistent shifts in immunoglobulin G (IgG) and plasma/serum glycoprotein glycosylation with age, with a pronounced rise in agalactosylated structures bound to IgG that also affect the composition of the total plasma N-glycome. These N-glycan modifications seem to be strongly associated with inflammaging and have been identified as valuable biomarkers for biological age, predictors of disease risk, and proxy biomarkers for monitoring intervention efficacy at the individual level. Despite current challenges related to data complexity and methodological limitations, ongoing technological innovations and interdisciplinary research are expected tofurther advance our knowledge of glycan biology, improve diagnostic and therapeutic strategies, and promote healthier aging. The integration of glycomics with other omics approaches holds promise for a more comprehensive understanding of the aging immune system, paving the way for personalized medicine and targeted interventions to mitigate inflammaging. In conclusion, this paper underscores the transformative impact of high-throughput Nglycan analysis in aging and inflammaging.

Machine learning-based clinical decision support using laboratory data

Artificial intelligence (AI) and machine learning (ML) are becoming vital in laboratory medicine and the broader context of healthcare. In this review article, we summarized the development of ML models and how they contribute to clinical laboratory workflow and improve patient outcomes. The process of ML model development involves data collection, data cleansing, feature engineering, model development, and optimization. These models, once finalized, are subjected to thorough performance assessments and validations. Recently, due to the complexity inherent in model development, automated ML tools were also introduced to streamline the process, enabling non-experts to create models. Clinical Decision Support Systems (CDSS) use ML techniques on large datasets to aid healthcare professionals in test result interpretation. They are revolutionizing laboratory medicine, enabling labs to work more efficiently with less human supervision across pre-analytical, analytical, and post-analytical phases. Despite contributions of the ML tools at all analytical phases, their integration presents challenges like potential model uncertainties, black-box algorithms, and deskilling of professionals. Additionally, acquiring diverse datasets is hard, and models' complexity can limit clinical use. In conclusion, ML-based CDSS in healthcare can greatly enhance clinical decision-making. However, successful adoption demands collaboration among professionals and stakeholders, utilizing hybrid intelligence, external validation, and performance assessments.

Host glycosylation of immunoglobulins impairs the immune response to acute Lyme disease

BACKGROUND

Lyme disease is caused by the bacteria Borreliella burgdorferi sensu lato (Bb) transmitted to humans from the bite of an infected Ixodes tick. Current diagnostics for Lyme disease are insensitive at the early disease stage and they cannot differentiate between active infections and people with a recent history of antibiotic-treated Lyme disease.

METHODS

Machine learning technology was utilized to improve the prediction of acute Lyme disease and identify sialic acid and galactose sugar structures (N-glycans) on immunoglobulins associated specifically at time points during acute Lyme disease time. A plate-based approach was developed to analyze sialylated N-glycans associated with anti-Bb immunoglobulins. This multiplexed approach quantitates the abundance of Bb-specific IgG and the associated sialic acid, yielding an accuracy of 90% in a powered study.

FINDINGS

It was demonstrated that immunoglobulin sialic acid levels increase during acute Lyme disease and following antibiotic therapy and a 3-month convalescence, the sialic acid level returned to that found in healthy control subjects (p < 0.001). Furthermore, the abundance of sialic acid on Bb-specific IgG during acute Lyme disease impaired the host's ability to combat Lyme disease via lymphocytic receptor FcγRIIIa signaling. After enzymatically removing the sialic acid present on Bb-specific antibodies, the induction of cytotoxicity from acute Lyme disease patient antigen-specific IgG was significantly improved.

INTERPRETATION

Taken together, Bb-specific immunoglobulins contain increased sialylation which impairs the host immune response during acute Lyme disease. Furthermore, this Bb-specific immunoglobulin sialyation found in acute Lyme disease begins to resolve following antibiotic therapy and convalescence.

FUNDING

Funding for this study was provided by the Coulter-Drexel Translational Research Partnership Program as well as from a Faculty Development Award from the Drexel University College of Medicine Institute for Molecular Medicine and Infectious Disease and the Department of Microbiology and Immunology.

Medimatrix: innovative pre-training of grayscale images for rheumatoid arthritis diagnosis revolutionises medical image classification

Efficient and accurate medical image classification (MIC) methods face two major challenges: (1) high similarity between images of different disease classes; and (2) generating large medical image datasets for training deep neural networks is challenging due to privacy restrictions and the need for expert ground truth annotations. In this paper, we introduce a novel deep learning method called pre-training grayscale images with supervised learning for MIC (MediMatrix). Instead of pre-training on color ImageNet, our approach uses MediMatrix on grayscale ImageNet. To improve the performance of the network, we introduce ShuffleAttention (SA), a self-attention mechanism. By combining SA with the multiple residual structure (ResSA block) and replacing short-cut connections with dense residual connections between corresponding layers (densepath), our network can dynamically adjust channel attention weights and receive image inputs of different sizes, resulting in improved feature representation and better discrimination of similarities between different categories. MediMatrix effectively classifies X-ray images of rheumatoid arthritis (RA), enabling efficient screening without the need for expert analysis or invasive testing. Through extensive experiments, we demonstrate the superiority of MediMatrix over state-of-the-art methods and that color is not critical for rich natural image classification. Our results highlight the potential of computer-aided diagnosis combined with MediMatrix as a valuable screening tool for early detection and intervention in RA.

Artificial intelligence in the analysis of glycosylation data

Glycans are complex, yet ubiquitous across biological systems. They are involved in diverse essential organismal functions. Aberrant glycosylation may lead to disease development, such as cancer, autoimmune diseases, and inflammatory diseases. Glycans, both normal and aberrant, are synthesized using extensive glycosylation machinery, and understanding this machinery can provide invaluable insights for diagnosis, prognosis, and treatment of various diseases. Increasing amounts of glycomics data are being generated thanks to advances in glycoanalytics technologies, but to maximize the value of such data, innovations are needed for analyzing and interpreting large-scale glycomics data. Artificial intelligence (AI) provides a powerful analysis toolbox in many scientific fields, and here we review state-of-the-art AI approaches on glycosylation analysis. We further discuss how models can be analyzed to gain mechanistic insights into glycosylation machinery and how the machinery shapes glycans under different scenarios. Finally, we propose how to leverage the gained knowledge for developing predictive AI-based models of glycosylation. Thus, guiding future research of AI-based glycosylation model development will provide valuable insights into glycosylation and glycan machinery.

Artificial Intelligence in Rheumatoid Arthritis: Current Status and Future Perspectives: A State-of-the-Art Review

Investigation of the potential applications of artificial intelligence (AI), including machine learning (ML) and deep learning (DL) techniques, is an exponentially growing field in medicine and healthcare. These methods can be critical in providing high-quality care to patients with chronic rheumatological diseases lacking an optimal treatment, like rheumatoid arthritis (RA), which is the second most prevalent autoimmune disease. Herein, following reviewing the basic concepts of AI, we summarize the advances in its applications in RA clinical practice and research. We provide directions for future investigations in this field after reviewing the current knowledge gaps and technical and ethical challenges in applying AI. Automated models have been largely used to improve RA diagnosis since the early 2000s, and they have used a wide variety of techniques, e.g., support vector machine, random forest, and artificial neural networks. AI algorithms can facilitate screening and identification of susceptible groups, diagnosis using omics, imaging, clinical, and sensor data, patient detection within electronic health record (EHR), i.e., phenotyping, treatment response assessment, monitoring disease course, determining prognosis, novel drug discovery, and enhancing basic science research. They can also aid in risk assessment for incidence of comorbidities, e.g., cardiovascular diseases, in patients with RA. However, the proposed models may vary significantly in their performance and reliability. Despite the promising results achieved by AI models in enhancing early diagnosis and management of patients with RA, they are not fully ready to be incorporated into clinical practice. Future investigations are required to ensure development of reliable and generalizable algorithms while they carefully look for any potential source of bias or misconduct. We showed that a growing body of evidence supports the potential role of AI in revolutionizing screening, diagnosis, and management of patients with RA. However, multiple obstacles hinder clinical applications of AI models. Incorporating the machine and/or deep learning algorithms into real-world settings would be a key step in the progress of AI in medicine.

Practical Implementation of Artificial Intelligence-Based Deep Learning and Cloud Computing on the Application of Traditional Medicine and Western Medicine in the Diagnosis and Treatment of Rheumatoid Arthritis

Rheumatoid arthritis (RA), an autoimmune disease of unknown etiology, is a serious threat to the health of middle-aged and elderly people. Although western medicine, traditional medicine such as traditional Chinese medicine, Tibetan medicine and other ethnic medicine have shown certain advantages in the diagnosis and treatment of RA, there are still some practical shortcomings, such as delayed diagnosis, improper treatment scheme and unclear drug mechanism. At present, the applications of artificial intelligence (AI)-based deep learning and cloud computing has aroused wide attention in the medical and health field, especially in screening potential active ingredients, targets and action pathways of single drugs or prescriptions in traditional medicine and optimizing disease diagnosis and treatment models. Integrated information and analysis of RA patients based on AI and medical big data will unquestionably benefit more RA patients worldwide. In this review, we mainly elaborated the application status and prospect of AI-assisted deep learning and cloud computation-oriented western medicine and traditional medicine on the diagnosis and treatment of RA in different stages. It can be predicted that with the help of AI, more pharmacological mechanisms of effective ethnic drugs against RA will be elucidated and more accurate solutions will be provided for the treatment and diagnosis of RA in the future.

Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

INTRODUCTION

Although the management of rheumatoid arthritis (RA) has improved in major way over the last decades, this disease still leads to an important burden for patients and society, and there is a need to develop more personalized approaches. Machine learning (ML) methods are more and more used in health-related studies and can be applied to different sorts of data (clinical, radiological, or 'omics' data). Such approaches may improve the management of patients with RA.

AREAS COVERED

In this paper, we propose a review regarding ML approaches applied to RA. A scoping literature search was performed in PubMed, in September 2021 using the following MeSH terms: 'arthritis, rheumatoid' and 'machine learning'. Based on this search, the usefulness of ML methods for RA diagnosis, monitoring, and prediction of response to treatment and RA outcomes, is discussed.

EXPERT OPINION

ML methods have the potential to revolutionize RA-related research and improve disease management and patient care. Nevertheless, these models are not yet ready to contribute fully to rheumatologists' daily practice. Indeed, these methods raise technical, methodological, and ethical issues, which should be addressed properly to allow their implementation. Collaboration between data scientists, clinical researchers, and physicians is therefore required to move this field forward.

iTRAQ-based proteomic analysis of differentially expressed proteins in sera of seronegative and seropositive rheumatoid arthritis patients

Objective

The diagnosis of seronegative rheumatoid arthritis (SNRA) is often difficult due to the unavailability of reliable laboratory markers. The aim of this study was to identify differentially expressed proteins in sera of SNRA, seropositive RA (SPRA), and healthy donors (HD).

Methods

A total of 32 seropositive RA patients, 32 SNRA patients, and 35 HD were enrolled in our study. Differentially expressed proteins between 3 groups were identified via isobaric tags for relative and absolute quantitation (iTRAQ)‐based proteomic analysis, and an ELISA test was used for the validation test. Correlation analysis was conducted by GraphPad Prism.

Results

Using iTRAQ quantitative proteomics, we identified 14 proteins were significantly different between SPRA and SNRA, including 4 upregulated proteins and 10 downregulated proteins. Four differentially expressed proteins were validated by ELISA test, and the results showed that SAA1 protein was significantly higher in SPRA and SNRA patients compared with HD, and PSME1 was elevated in SPRA patients. What's more, SAA1 was increased in the anti‐CCP or RF high‐level group in RA patients, and PSME1 was increased in the RF high‐level group. Alternatively, SAA1 was positively correlated with inflammation indicators in RA patients, while PSME1 showed no correlation with inflammation indicators.

Conclusions

iTRAQ proteomic approaches revealed variations in serum protein composition among SPRA patients, SNRA patients, and HD and provided new idea for advanced diagnostic methods and precision treatment of RA.

Validating fPSA Glycoprofile as a Prostate Cancer Biomarker to Avoid Unnecessary Biopsies and Re-Biopsies

BACKGROUND

To compare the clinical performance of a new PCa serum biomarker based on fPSA glycoprofiling to fPSA% and PHI.

METHODS

Serum samples from men who underwent prostate biopsy due to increased PSA were used. A comparison between two equal groups (with histologically confirmed PCa or benign, non-cancer condition) was used for the clinical validation of a new glycan-based PCa oncomarker. SPSS and R software packages were used for the multiparametric analyses of the receiver operating curve (ROC) and for genetic algorithm metaheuristics.

RESULTS

When comparing the non-cancer and PCa cohorts, the combination of four fPSA glycoforms with two clinical parameters (PGI, prostate glycan index (PGI)) showed an area under receiver operating curve (AUC) value of 0.821 (95% CI 0.754-0.890). AUC values were 0.517 for PSA, 0.683 for fPSA%, and 0.737 for PHI. A glycan analysis was also applied to discriminate low-grade tumors (GS = 6) from significant tumors (GS ≥ 7).

CONCLUSIONS

Compared to PSA on its own, or fPSA% and the PHI, PGI showed improved discrimination between presence and absence of PCa and in predicting clinically significant PCa. In addition, the use of PGI would help practitioners avoid 63.5% of unnecessary biopsies, while the use of fPSA% and PHI would help avoid 17.5% and 33.3% of biopsies, respectively, while missing four significant tumors (9.5%).

Precision medicine in the care of rheumatoid arthritis: Focus on prediction and prevention of future clinically-apparent disease

There is an emerging understanding that an individual's risk for future rheumatoid arthritis (RA) can be determined using a combination of factors while they are still in a state where clinically-apparent inflammatory arthritis (IA) is not yet present. Indeed, this concept has underpinned several completed and ongoing prevention trials in RA. Importantly, risk factors can be divided into modifiable (e.g. smoking, exercise, dental care and diet) and non-modifiable factors (e.g. genetics, sex, age). In addition, there are now several biomarkers including autoantibodies, inflammatory markers and imaging techniques that are highly predictive of future clinically-apparent IA/RA. Although none of the prevention studies have yet provided major breakthroughs, several of them have provided valuable insights that can help to improve the design of future clinical trials and enable RA prevention. In aggregate, these findings suggest that the most accurate disease prediction models will require the combination of demographic and clinical information, biomarkers and potentially medical imaging data to identify individuals for intervention. This review summarizes some of the key aspects around precision medicine in RA with special focus on disease prediction and prevention.

A systematic review of the applications of artificial intelligence and machine learning in autoimmune diseases

Autoimmune diseases are chronic, multifactorial conditions. Through machine learning (ML), a branch of the wider field of artificial intelligence, it is possible to extract patterns within patient data, and exploit these patterns to predict patient outcomes for improved clinical management. Here, we surveyed the use of ML methods to address clinical problems in autoimmune disease. A systematic review was conducted using MEDLINE, embase and computers and applied sciences complete databases. Relevant papers included "machine learning" or "artificial intelligence" and the autoimmune diseases search term(s) in their title, abstract or key words. Exclusion criteria: studies not written in English, no real human patient data included, publication prior to 2001, studies that were not peer reviewed, non-autoimmune disease comorbidity research and review papers. 169 (of 702) studies met the criteria for inclusion. Support vector machines and random forests were the most popular ML methods used. ML models using data on multiple sclerosis, rheumatoid arthritis and inflammatory bowel disease were most common. A small proportion of studies (7.7% or 13/169) combined different data types in the modelling process. Cross-validation, combined with a separate testing set for more robust model evaluation occurred in 8.3% of papers (14/169). The field may benefit from adopting a best practice of validation, cross-validation and independent testing of ML models. Many models achieved good predictive results in simple scenarios (e.g. classification of cases and controls). Progression to more complex predictive models may be achievable in future through integration of multiple data types.

Mass Spectrometry as a Clinical Integrative Tool to Evaluate Hepatocellular Carcinoma: Moving to the Mainstream

Introduction: Since the pioneering work of J. J. Thomson on magnetic deflection of charged particles, mass spectrometry (MS) has become the most progressive clinical tool by continuously providing new applications in medical research. In hepatocellular carcinoma (HCC), MS can be used from surveillance in early stages of the disease to constant evaluation of effective treatments. Areas covered: This Special Report highlights the groundbreaking possibilities of mass spectrometry clinical application in the mainstream to evaluate HCC development and progression. Expert opinon: MS has been employed to understand a myriad of liver diseases, such as the identification of early biomarkers in cirrhosis and HVB and HVC, as well as metabolic alterations of lipidic imbalance in HCC due to fatty liver disease. In an integrative point-of-view, researchers worldwide are looking for molecular signatures that may represent more faithfully the complex scenario of the onset and progression of HCC. Following the steps of MELD score (Model of End-stage Liver Disease), which evaluates biochemical dysfunction of end-stage liver diseases, the necessity to use innovative attempts to pursue a molecular-MEaLD (mMEaLD - molecular Model for Early Liver Disease), shifting MS to the upstream and from the lab facilities into the mainstream, inside the surgery room.

Advanced impedimetric biosensor configuration and assay protocol for glycoprofiling of a prostate oncomarker using Au nanoshells with a magnetic core

In this paper several advances were implemented for glycoprofiling of prostate specific antigen (PSA), what can be applied for better prostate cancer (PCa) diagnostics in the future: 1) application of Au nanoshells with a magnetic core (MP@silica@Au); 2) use of surface plasmons of Au nanoshells with a magnetic core for spontaneous immobilization of zwitterionic molecules via diazonium salt grafting; 3) a double anti-fouling strategy with integration of zwitterionic molecules on Au surface and on MP@silica@Au particles was implemented to resist non-specific protein binding; 4) application of anti-PSA antibody modified Au nanoshells with a magnetic core for enrichment of PSA from a complex matrix of a human serum; 5) direct incubation of anti-PSA modified MP@silica@Au with affinity bound PSA to the lectin modified electrode surface. The electrochemical impedance spectroscopy (EIS) signal was enhanced 43 times integrating Au nanoshells with a magnetic core compared to the biosensor without them. This proof-of-concept study shows that the biosensor could detect PSA down to 1.2 fM and at the same time to glycoprofile such low PSA concentration using a lectin patterned biosensor device. The biosensor offers a recovery index of 108%, when serum sample was spiked with a physiological concentration of PSA (3.5 ng mL-1).

Prostate-specific antigen glycoprofiling as diagnostic and prognostic biomarker of prostate cancer

The initial part of this review details the controversy behind the use of a serological level of prostate-specific antigen (PSA) for the diagnostics of prostate cancer (PCa). Novel biomarkers are in demand for PCa diagnostics, outperforming traditional PSA tests. The review provides a detailed and comprehensive summary that PSA glycoprofiling can effectively solve this problem, thereby considerably reducing the number of unnecessary biopsies. In addition, PSA glycoprofiling can serve as a prognostic PCa biomarker to identify PCa patients with an aggressive form of PCa, avoiding unnecessary further treatments which are significantly life altering (incontinence or impotence).

Glycomics of prostate cancer: updates

Introduction: Prostate cancer (PCa) is a life-threatening disease affecting millions of men. The current best PCa biomarker (level of prostate-specific antigen in serum) lacks specificity for PCa diagnostics and this is why novel PCa biomarkers in addition to the conventional ones based on biomolecules such as DNA, RNA and proteins need to be identified. Areas covered: This review details the potential of glycans-based biomarkers to become diagnostic, prognostic, predictive and therapeutic PCa biomarkers with a brief description of the innovative approaches applied to glycan analysis to date. Finally, the review covers the possibility to use exosomes as a rich source of glycans for future innovative and advanced diagnostics of PCa. The review covers updates in the field since 2016. Expert commentary: The summary provided in this review paper suggests that glycan-based biomarkers can offer high-assay accuracy not only for diagnostic purposes but also for monitoring/surveillance of the PCa disease.