Machine learning reveals distinct gene signature profiles in lesional and nonlesional regions of inflammatory skin diseases

Navigating the evolving landscape of atopic dermatitis: Challenges and future opportunities: The 4th Davos declaration

The 4th Davos Declaration was developed during the Global Allergy Forum in Davos which aimed to elevate the care of patients with atopic dermatitis (AD) by uniting experts and stakeholders. The forum addressed the high prevalence of AD, with a strategic focus on advancing research, treatment, and management to meet the evolving challenges in the field. This multidisciplinary forum brought together top leaders from research, clinical practice, policy, and patient advocacy to discuss the critical aspects of AD, including neuroimmunology, environmental factors, comorbidities, and breakthroughs in prevention, diagnosis, and treatment. The discussions were geared towards fostering a collaborative approach to integrate these advancements into practical, patient-centric care. The forum underlined the mounting burden of AD, attributing it to significant environmental and lifestyle changes. It acknowledged the progress in understanding AD and in developing targeted therapies but recognized a gap in translating these innovations into clinical practice. Emphasis was placed on the need for enhanced awareness, education, and stakeholder engagement to address this gap effectively and to consider environmental and lifestyle factors in a comprehensive disease management strategy. The 4th Davos Declaration marks a significant milestone in the journey to improve care for people with AD. By promoting a holistic approach that combines research, education, and clinical application, the Forum sets a roadmap for stakeholders to collaborate to improve patient outcomes in AD, reflecting a commitment to adapt and respond to the dynamic challenges of AD in a changing world.

Langerhans cells regulate immunity in adulthood by regulating postnatal dermal lymphatic development

The communication between skin and draining lymph nodes is crucial for well-regulated immune responses to skin insults. The skin sends antigen and other signals via lymphatic vessels to regulate lymph node activity, and regulating dermal lymphatic function is another means to control immunity. Here, we show that Langerhans cells (LCs), epidermis-derived antigen-presenting cells, mediate dermal lymphatic expansion and phenotype acquisition postnatally, a function is independent of LC entry into lymphatic vessels. This postnatal LC-lymphatic axis serves in part to control inflammatory systemic T cell responses in adulthood. Our data provide a tissue-based mechanism by which LCs regulate T cells remotely across time and space and raise the possibility that immune diseases in adulthood could reflect compromise of the LC-lymphatic axis in childhood.

The interferon-rich skin environment regulates Langerhans cell ADAM17 to promote photosensitivity in lupus

The autoimmune disease lupus erythematosus (lupus) is characterized by photosensitivity, where even ambient ultraviolet radiation (UVR) exposure can lead to development of inflammatory skin lesions. We have previously shown that Langerhans cells (LCs) limit keratinocyte apoptosis and photosensitivity via a disintegrin and metalloprotease 17 (ADAM17)-mediated release of epidermal growth factor receptor (EGFR) ligands and that LC ADAM17 sheddase activity is reduced in lupus. Here, we sought to understand how the lupus skin environment contributes to LC ADAM17 dysfunction and, in the process, differentiate between effects on LC ADAM17 sheddase function, LC ADAM17 expression, and LC numbers. We show through transcriptomic analysis a shared IFN-rich environment in non-lesional skin across human lupus and three murine models: MRL/lpr, B6.Sle1yaa, and imiquimod (IMQ) mice. IFN-I inhibits LC ADAM17 sheddase activity in murine and human LCs, and IFNAR blockade in lupus model mice restores LC ADAM17 sheddase activity, all without consistent effects on LC ADAM17 protein expression or LC numbers. Anti-IFNAR-mediated LC ADAM17 sheddase function restoration is associated with reduced photosensitive responses that are dependent on EGFR signaling and LC ADAM17. Reactive oxygen species (ROS) is a known mediator of ADAM17 activity; we show that UVR-induced LC ROS production is reduced in lupus model mice, restored by anti-IFNAR, and is cytoplasmic in origin. Our findings suggest that IFN-I promotes photosensitivity at least in part by inhibiting UVR-induced LC ADAM17 sheddase function and raise the possibility that anifrolumab ameliorates lupus skin disease in part by restoring this function. This work provides insight into IFN-I-mediated disease mechanisms, LC regulation, and a potential mechanism of action for anifrolumab in lupus.

Unraveling transcriptomic signatures and dysregulated pathways in systemic lupus erythematosus across disease states

OBJECTIVES

This study aims to elucidate the transcriptomic signatures and dysregulated pathways in patients with Systemic Lupus Erythematosus (SLE), with a particular focus on those persisting during disease remission.

METHODS

We conducted bulk RNA-sequencing of peripheral blood mononuclear cells (PBMCs) from a well-defined cohort comprising 26 remission patients meeting the Low Lupus Disease Activity State (LLDAS) criteria, 76 patients experiencing disease flares, and 15 healthy controls. To elucidate immune signature changes associated with varying disease states, we performed extensive analyses, including the identification of differentially expressed genes and pathways, as well as the construction of protein-protein interaction networks.

RESULTS

Several transcriptomic features recovered during remission compared to the active disease state, including down-regulation of plasma and cell cycle signatures, as well as up-regulation of lymphocytes. However, specific innate immune response signatures, such as the interferon (IFN) signature, and gene modules involved in chromatin structure modification, persisted across different disease states. Drug repurposing analysis revealed certain drug classes that can target these persistent signatures, potentially preventing disease relapse.

CONCLUSION

Our comprehensive transcriptomic study revealed gene expression signatures for SLE in both active and remission states. The discovery of gene expression modules persisting in the remission stage may shed light on the underlying mechanisms of vulnerability to relapse in these patients, providing valuable insights for their treatment.

Artificial Intelligence: A Snapshot of Its Application in Chronic Inflammatory and Autoimmune Skin Diseases

Immuno-correlated dermatological pathologies refer to skin disorders that are closely associated with immune system dysfunction or abnormal immune responses. Advancements in the field of artificial intelligence (AI) have shown promise in enhancing the diagnosis, management, and assessment of immuno-correlated dermatological pathologies. This intersection of dermatology and immunology plays a pivotal role in comprehending and addressing complex skin disorders with immune system involvement. The paper explores the knowledge known so far and the evolution and achievements of AI in diagnosis; discusses segmentation and the classification of medical images; and reviews existing challenges, in immunological-related skin diseases. From our review, the role of AI has emerged, especially in the analysis of images for both diagnostic and severity assessment purposes. Furthermore, the possibility of predicting patients' response to therapies is emerging, in order to create tailored therapies.

Systemic lupus in the era of machine learning medicine

Artificial intelligence and machine learning applications are emerging as transformative technologies in medicine. With greater access to a diverse range of big datasets, researchers are turning to these powerful techniques for data analysis. Machine learning can reveal patterns and interactions between variables in large and complex datasets more accurately and efficiently than traditional statistical methods. Machine learning approaches open new possibilities for studying SLE, a multifactorial, highly heterogeneous and complex disease. Here, we discuss how machine learning methods are rapidly being integrated into the field of SLE research. Recent reports have focused on building prediction models and/or identifying novel biomarkers using both supervised and unsupervised techniques for understanding disease pathogenesis, early diagnosis and prognosis of disease. In this review, we will provide an overview of machine learning techniques to discuss current gaps, challenges and opportunities for SLE studies. External validation of most prediction models is still needed before clinical adoption. Utilisation of deep learning models, access to alternative sources of health data and increased awareness of the ethics, governance and regulations surrounding the use of artificial intelligence in medicine will help propel this exciting field forward.

Identification and validation of a glycolysis-related taxonomy for improving outcomes in glioma

BACKGROUND

Reprogramming of glucose metabolism is a prominent abnormal energy metabolism in glioma. However, the efficacy of treatments targeting glycolysis varies among patients. The present study aimed to classify distinct glycolysis subtypes (GS) of glioma, which may help to improve the therapy response.

METHODS

The expression profiles of glioma were downloaded from public datasets to perform an enhanced clustering analysis to determine the GS. A total of 101 combinations based on 10 machine learning algorithms were performed to screen out the most valuable glycolysis-related glioma signature (GGS). Through RSF and plsRcox algorithms, adrenomedullin (ADM) was eventually obtained as the most significant glycolysis-related gene for prognostic prediction in glioma. Furthermore, drug sensitivity analysis, molecular docking, and in vitro experiments were utilized to verify the efficacy of ADM and ingenol mebutate (IM).

RESULTS

Glioma patients were classified into five distinct GS (GS1-GS5), characterized by varying glycolytic metabolism levels, molecular expression, immune cell infiltration, immunogenic modulators, and clinical features. Anti-CTLA4 and anti-PD-L1 antibodies significantly improved the prognosis for GS2 and GS5, respectively. ADM has been identified as a potential biomarker for targeted glycolytic therapy in glioma patients. In vitro experiments demonstrated that IM inhibited glioma cell progression by inhibiting ADM.

CONCLUSION

This study elucidates that evaluating GS is essential for comprehending the heterogeneity of glioma, which is pivotal for predicting immune cell infiltration (ICI) characterization, prognosis, and personalized immunotherapy regimens. We also explored the glycolysis-related genes ADM and IM to develop a theoretical framework for anti-tumor strategies targeting glycolysis.

Machine Learning Made Easy (MLme): a comprehensive toolkit for machine learning-driven data analysis

BACKGROUND

Machine learning (ML) has emerged as a vital asset for researchers to analyze and extract valuable information from complex datasets. However, developing an effective and robust ML pipeline can present a real challenge, demanding considerable time and effort, thereby impeding research progress. Existing tools in this landscape require a profound understanding of ML principles and programming skills. Furthermore, users are required to engage in the comprehensive configuration of their ML pipeline to obtain optimal performance.

RESULTS

To address these challenges, we have developed a novel tool called Machine Learning Made Easy (MLme) that streamlines the use of ML in research, specifically focusing on classification problems at present. By integrating 4 essential functionalities-namely, Data Exploration, AutoML, CustomML, and Visualization-MLme fulfills the diverse requirements of researchers while eliminating the need for extensive coding efforts. To demonstrate the applicability of MLme, we conducted rigorous testing on 6 distinct datasets, each presenting unique characteristics and challenges. Our results consistently showed promising performance across different datasets, reaffirming the versatility and effectiveness of the tool. Additionally, by utilizing MLme's feature selection functionality, we successfully identified significant markers for CD8+ naive (BACH2), CD16+ (CD16), and CD14+ (VCAN) cell populations.

CONCLUSION

MLme serves as a valuable resource for leveraging ML to facilitate insightful data analysis and enhance research outcomes, while alleviating concerns related to complex coding scripts. The source code and a detailed tutorial for MLme are available at https://github.com/FunctionalUrology/MLme.

Analysis of clinical significance and molecular characteristics of methionine metabolism and macrophage-related patterns in hepatocellular carcinoma based on machine learning

BACKGROUND

Increasing evidence has indicated that abnormal methionine metabolic activity and tumour-associated macrophage infiltration are correlated with hepatocarcinogenesis. However, the relationship between methionine metabolic activity and tumour-associated macrophage infiltration is unclear in hepatocellular carcinoma, and it contributes to the occurrence and clinical outcome of hepatocellular carcinoma (HCC). Thus, we systematically analysed the expression patterns of methionine metabolism and macrophage infiltration in hepatocellular carcinoma using bioinformatics and machine learning methods and constructed novel diagnostic and prognostic models of HCC.

METHODS

In this study, we first mined the four largest HCC mRNA microarray datasets with patient clinical data in the GEO database, including 880 tissue mRNA expression datasets. Using GSVA analysis and the CIBERSORT and EPIC algorithms, we quantified the methionine metabolic activity and macrophage infiltration degree of each sample. WGCNA was used to identify the gene modules most related to methionine metabolism and tumour-associated macrophage infiltration in HCC. The KNN algorithm was used to cluster gene expression patterns in HCC. Random forest, logistic regression, Cox regression analysis and other algorithms were used to construct the diagnosis and prognosis model of HCC. The above bioinformatics analysis results were also verified by independent datasets (TCGA-LIHC, ICGC-JP and CPTAC datasets) and immunohistochemical fluorescence based on our external HCC panel. Furthermore, we carried out pancancer analysis to verify the specificity of the above model and screened a wide range of drug candidates.

RESULTS

We identified two methionine metabolism and macrophage infiltration expression patterns, and their prognoses were different in hepatocellular carcinoma. We constructed novel diagnostic and prognostic models of hepatocellular carcinoma with good diagnostic efficacy and differentiation ability.

CONCLUSIONS

Methionine metabolism is closely related to tumour-associated macrophage infiltration in hepatocellular carcinoma and can help in the clinical diagnosis and prognosis of HCC.

Identification of ADAM23 as a Potential Signature for Psoriasis Using Integrative Machine-Learning and Experimental Verification

Background

Psoriasis is a common chronic, recurrent, and inflammatory skin disease. Identifying novel and potential biomarkers is valuable in the treatment and diagnosis of psoriasis. The goal of this study was to identify novel key biomarkers of psoriasis and analyze the potential underlying mechanisms.

Methods

Psoriasis-related datasets were downloaded from the Gene Expression Omnibus database to screen differential genes in the datasets. Functional and pathway enrichment analyses were performed on the differentially expressed genes (DEGs). Candidate biomarkers for psoriasis were identified from the GSE30999 and GSE6710 datasets using four machine learning algorithms, namely, random forest (RF), least absolute shrinkage and selection operator (LASSO) logistic regression, weighted gene co-expression network analysis (WGCNA), and support vector machine recursive feature elimination (SVM-RFE), and were validated using the GSE41662 dataset. Next, we used CIBERSORT and single-cell RNA analysis to explore the relationship between ADAM23 and immune cells. Finally, we validated the expression of the identified biomarkers expressions in human and mouse experiments.

Results

A total of 709 overlapping DEGs were identified, including 426 upregulated and 283 downregulated genes. Enhanced by enrichment analysis, the differentially expressed genes (DEGs) were spatially arranged in relation to immune cell involvement, immune-activating processes, and inflammatory signals. Based on the enrichment analysis, the DEGs were mapped to immune cell involvement, immune-activating processes, and inflammatory signals. Four machine learning strategies and single-cell RNA sequencing analysis showed that ADAM23, a disintegrin and metalloprotease, may be a unique, critical biomarker with high diagnostic accuracy for psoriasis. Based on CIBERSORT analysis, ADAM23 was found to be associated with a variety of immune cells, such as macrophages and mast cells, and it was upregulated in the macrophages of psoriatic lesions in patients and mice.

Conclusion

ADAM23 may be a potential biomarker in the diagnosis of psoriasis and may contribute to the pathogenesis by regulating immunological activity in psoriatic lesions.

Analysis of transcriptomic features reveals molecular endotypes of SLE with clinical implications

BACKGROUND

Systemic lupus erythematosus (SLE) is known to be clinically heterogeneous. Previous efforts to characterize subsets of SLE patients based on gene expression analysis have not been reproduced because of small sample sizes or technical problems. The aim of this study was to develop a robust patient stratification system using gene expression profiling to characterize individual lupus patients.

METHODS

We employed gene set variation analysis (GSVA) of informative gene modules to identify molecular endotypes of SLE patients, machine learning (ML) to classify individual patients into molecular subsets, and logistic regression to develop a composite metric estimating the scope of immunologic perturbations. SHapley Additive ExPlanations (SHAP) revealed the impact of specific features on patient sub-setting.

RESULTS

Using five datasets comprising 2183 patients, eight SLE endotypes were identified. Expanded analysis of 3166 samples in 17 datasets revealed that each endotype had unique gene enrichment patterns, but not all endotypes were observed in all datasets. ML algorithms trained on 2183 patients and tested on 983 patients not used to develop the model demonstrated effective classification into one of eight endotypes. SHAP indicated a unique array of features influential in sorting individual samples into each of the endotypes. A composite molecular score was calculated for each patient and significantly correlated with standard laboratory measures. Significant differences in clinical characteristics were associated with different endotypes, with those with the least perturbed transcriptional profile manifesting lower disease severity. The more abnormal endotypes were significantly more likely to experience a severe flare over the subsequent 52 weeks while on standard-of-care medication and specific endotypes were more likely to be clinical responders to the investigational product tested in one clinical trial analyzed (tabalumab).

CONCLUSIONS

Transcriptomic profiling and ML reproducibly separated lupus patients into molecular endotypes with significant differences in clinical features, outcomes, and responsiveness to therapy. Our classification approach using a composite scoring system based on underlying molecular abnormalities has both staging and prognostic relevance.

Skin Gene Expression Profiles in Systemic Sclerosis: From Clinical Stratification to Precision Medicine

Systemic sclerosis, also known as scleroderma or SSc, is a condition characterized by significant heterogeneity in clinical presentation, disease progression, and response to treatment. Consequently, the design of clinical trials to successfully identify effective therapeutic interventions poses a major challenge. Recent advancements in skin molecular profiling technologies and stratification techniques have enabled the identification of patient subgroups that may be relevant for personalized treatment approaches. This narrative review aims at providing an overview of the current status of skin gene expression analysis using computational biology approaches and highlights the benefits of stratifying patients upon their skin gene signatures. Such stratification has the potential to lead toward a precision medicine approach in the management of SSc.

Machine Learning Made Easy (MLme): A Comprehensive Toolkit for Machine Learning-Driven Data Analysis

Background

Machine learning (ML) has emerged as a vital asset for researchers to analyze and extract valuable information from complex datasets. However, developing an effective and robust ML pipeline can present a real challenge, demanding considerable time and effort, thereby impeding research progress. Existing tools in this landscape require a profound understanding of ML principles and programming skills. Furthermore, users are required to engage in the comprehensive configuration of their ML pipeline to obtain optimal performance.

Results

To address these challenges, we have developed a novel tool called Machine Learning Made Easy (MLme) that streamlines the use of ML in research, specifically focusing on classification problems at present. By integrating four essential functionalities, namely Data Exploration, AutoML, CustomML, and Visualization, MLme fulfills the diverse requirements of researchers while eliminating the need for extensive coding efforts. To demonstrate the applicability of MLme, we conducted rigorous testing on six distinct datasets, each presenting unique characteristics and challenges. Our results consistently showed promising performance across different datasets, reaffirming the versatility and effectiveness of the tool. Additionally, by utilizing MLme's feature selection functionality, we successfully identified significant markers for CD8+ naive (BACH2), CD16+ (CD16), and CD14+ (VCAN) cell populations.

Conclusion

MLme serves as a valuable resource for leveraging machine learning (ML) to facilitate insightful data analysis and enhance research outcomes, while alleviating concerns related to complex coding scripts. The source code and a detailed tutorial for MLme are available at https://github.com/FunctionalUrology/MLme.

Convolutional neural network models for automatic diagnosis and graduation in skin frostbite

The study aimed to develop and validate a convolutional neural network (CNN)-based deep learning method for automatic diagnosis and graduation of skin frostbite. A dataset of 71 annotated images was used for the training, the validation, and the testing based on ResNet-50 model. The performances were evaluated with the test set. The diagnosis and graduation performance of our approach was compared with two residents from burns department. The approach correctly identified all the frostbite of IV (18/18, 100%), but with respectively 1 mistake in the diagnosis of degree I (29/30, 96.67%), II (28/29, 96.55%) and III (37/38, 97.37%). The accuracy of the approach on the whole test set was 97.39% (112/115). The accuracy of the two residents were respectively 77.39% and 73.04%. Weighted Kappa of 0.583 indicates good reliability between the two residents (P = .445). Kendall's coefficient of concordance is 0.326 (P = .548), indicating differences in accuracy between the approach and the two residents. Our approach based on CNNs demonstrated an encouraging performance for the automatic diagnosis and graduation of skin frostbite, with higher accuracy and efficiency.

The Hidradenitis Suppurativa 'Omics Database (HS-OmicsDB)

Large scale meta-analyses of genomics and genetics have spurred research in a number of fields, such as cancer, genetics and immunology. Publicly available 'omics databases provide valuable hypothesis generating and validation tools. To date, no such initiative has been undertaken for Hidradenitis Suppurativa (HS), an inflammatory skin disease of unknown etiology. We present here, a longitudinal initiative seeking to aggregate publicly available 'omics data to enhance research efforts in HS. In its first iteration, we include bulk and single-cell RNA sequencing data from untreated HS patients. Our data, aggregated from publicly available GEO datasets provides a tool to profile gene expression in specific tissue types (i.e. lesional, perilesional, nonlesional and healthy skin) as well as map cell-specific gene expression on single-cell data from HS lesions.

Hepatitis B virus pathogenesis relevant immunosignals uncovering amino acids utilization related risk factors guide artificial intelligence-based precision medicine

Background: Although immune microenvironment-related chemokines, extracellular matrix (ECM), and intrahepatic immune cells are reported to be highly involved in hepatitis B virus (HBV)-related diseases, their roles in diagnosis, prognosis, and drug sensitivity evaluation remain unclear. Here, we aimed to study their clinical use to provide a basis for precision medicine in hepatocellular carcinoma (HCC) via the amalgamation of artificial intelligence. Methods: High-throughput liver transcriptomes from Gene Expression Omnibus (GEO), NODE (https://www.bio.sino.org/node), the Cancer Genome Atlas (TCGA), and our in-house hepatocellular carcinoma patients were collected in this study. Core immunosignals that participated in the entire diseases course of hepatitis B were explored using the "Gene set variation analysis" R package. Using ROC curve analysis, the impact of core immunosignals and amino acid utilization related gene on hepatocellular carcinoma patient's clinical outcome were calculated. The utility of core immunosignals as a classifier for hepatocellular carcinoma tumor tissue was evaluated using explainable machine-learning methods. A novel deep residual neural network model based on immunosignals was constructed for the long-term overall survival (LS) analysis. In vivo drug sensitivity was calculated by the "oncoPredict" R package. Results: We identified nine genes comprising chemokines and ECM related to hepatitis B virus-induced inflammation and fibrosis as CLST signals. Moreover, CLST was co-enriched with activated CD4+ T cells bearing harmful factors (aCD4) during all stages of hepatitis B virus pathogenesis, which was also verified by our hepatocellular carcinoma data. Unexpectedly, we found that hepatitis B virus-hepatocellular carcinoma patients in the CLST^highaCD4^high subgroup had the shortest overall survival (OS) and were characterized by a risk gene signature associated with amino acids utilization. Importantly, characteristic genes specific to CLST/aCD4 showed promising clinical relevance in identifying patients with early-stage hepatocellular carcinoma via explainable machine learning. In addition, the 5-year long-term overall survival of hepatocellular carcinoma patients can be effectively classified by CLST/aCD4 based GeneSet-ResNet model. Subgroups defined by CLST and aCD4 were significantly involved in the sensitivity of hepatitis B virus-hepatocellular carcinoma patients to chemotherapy treatments. Conclusion: CLST and aCD4 are hepatitis B virus pathogenesis-relevant immunosignals that are highly involved in hepatitis B virus-induced inflammation, fibrosis, and hepatocellular carcinoma. Gene set variation analysis derived immunogenomic signatures enabled efficient diagnostic and prognostic model construction. The clinical application of CLST and aCD4 as indicators would be beneficial for the precision management of hepatocellular carcinoma.

Recent advances in the use of machine learning and artificial intelligence to improve diagnosis, predict flares, and enrich clinical trials in lupus

PURPOSE OF REVIEW

Machine learning is a computational tool that is increasingly used for the analysis of medical data and has provided the promise of more personalized care.

RECENT FINDINGS

The frequency with which machine learning analytics are reported in lupus research is comparable with that of rheumatoid arthritis and cancer, yet the clinical application of these computational tools has yet to be translated into better care. Considerable work has been applied to the development of machine learning models for lupus diagnosis, flare prediction, and classification of disease using histology or other medical images, yet few models have been tested in external datasets and independent centers. Application of machine learning has yet to be reported for lupus clinical trial enrichment and automated identification of eligible patients. Integration of machine learning into lupus clinical care and clinical trials would benefit from collaborative development between clinicians and data scientists.

SUMMARY

Although the application of machine learning to lupus data is at a nascent stage, initial results suggest a promising future.