Characteristics of myasthenia gravis in elderly patients: a retrospective study.

BACKGROUND: The incidence of myasthenia gravis (MG) is increasing, and its characteristics in elderly patients are believed to differ from those in younger patients. However, only a few studies have focused on elderly patients with MG. OBJECTIVE: To review the characteristics of MG in elderly patients and evaluate whether older age is an independent factor associated with achieving minimal manifestation status (MMS). METHODS: This retrospective cohort study included 367 patients (319 non-elderly and 48 elderly patients) with MG enrolled at Xiangya Hospital from September 1, 2016, to December 31, 2018. We collected demographic data and information regarding comorbidities, antibody status, Myasthenia Gravis Foundation of America classification, affected muscle groups, thymoma, and treatment. MMS was defined as the primary outcome. RESULTS: Comorbidities were more common in elderly than in younger patients with MG. Anti-acetylcholine receptor antibody was the dominant subtype, whereas anti-muscle-specific tyrosine kinase antibody was rare and detected only in non-elderly patients. Elderly patients were more likely than younger patients to have generalized MG, but the frequency of thymoma was lower (28.5% vs. 10.4%, p = 0.0078). MMS or better was achieved in 154 (48.3%) and 13 (27.1%) non-elderly and elderly patients, respectively. Older age did not appear to be an independent factor associated with MMS (hazard ratio = 0.625; 95% confidence interval, 0.345-1.131). CONCLUSIONS: Older age was not an independent factor for a worse prognosis in patients with MG. The treatment of elderly patients with MG should be individually tailored.

Carbon Nanotube Field-Effect Transistor-Based Chemical and Biological Sensors.

Chemical and biological sensors have attracted great interest due to their importance in applications of healthcare, food quality monitoring, environmental monitoring, etc. Carbon nanotube (CNT)-based field-effect transistors (FETs) are novel sensing device configurations and are very promising for their potential to drive many technological advancements in this field due to the extraordinary electrical properties of CNTs. This review focuses on the implementation of CNT-based FETs (CNTFETs) in chemical and biological sensors. It begins with the introduction of properties, and surface functionalization of CNTs for sensing. Then, configurations and sensing mechanisms for CNT FETs are introduced. Next, recent progresses of CNTFET-based chemical sensors, and biological sensors are summarized. Finally, we end the review with an overview about the current application status and the remaining challenges for the CNTFET-based chemical and biological sensors.

Multiple genetic factors affecting the pharmacokinetic and pharmacodynamic processes of tacrolimus in Chinese myasthenia gravis patients.

PURPOSE: Tacrolimus is a novel effective immunosuppressant for myasthenia gravis (MG) patients. However, the narrow therapeutic window, and high inter- and intrapatient variation in bioavailability largely limited its clinical application. This article intended to find the SNPs influencing clinical outcome and discover the possible mechanisms. METHODS: Based on the tagSNPs genotyped by Improved Multiple Ligase Detection Reaction, Plink 1.07 was used to find the SNPs having close interaction to tacrolimus serum concentration, QMG score changes or even reasonable drug dose. Then we searched several databases to predict the possible miRNA binding rs15524 sequence. Based on the prediction, dual-luciferase reporter assay and miRNA transfection were used to discover the mechanism of how SNP rs15524 controls tacrolimus serum concentration through influencing CYP3A5 expression. RESULTS: In this article, we found multiple SNPs on CYP3A4, CYP3A5, FKBP1A, NFATC2 genes were predicted closely related to tacrolimus serum concentration, therapeutic effect which reflected by QMG score changes or even reasonable drug dose. After in silico miRNA selection, possible relationship between hsa-miR-500a and rs15524 was found. With the help of dual-luciferase reporter assay, wild-type rs15524 (T allele) was found having a stronger binding affinity for hsa-miR-500a. Higher expression of CYP3A5 may also led by lower hsa-miR-500a level. CONCLUSIONS: SNP rs15524 may control CYP3A5 expression by affecting the binding affinity between CYP3A5 3'UTR and hsa-miR-500a. Wild type (T allele) 3'UTR of CYP3A5 has stronger binding affinity to hsa-miR-500a and cause lower CYP3A5 expression and higher tacrolimus serum concentration.

Clinical decision support systems for 3-month mortality in elderly patients admitted to ICU with ischemic stroke using interpretable machine learning.

Background: Elderly patients are more likely to suffer from severe ischemic stroke (IS) and have worse outcomes, including death and disability. We aimed to develop and validate predictive models using novel machine learning algorithms for the 3-month mortality in elderly patients with IS admitted to the intensive care unit (ICU). Methods: We conducted a retrospective cohort study. Data were extracted from Medical Information Mart for Intensive Care (MIMIC)-IV and International Stroke Perfusion Imaging Registry (INSPIRE) database. Ten machine learning algorithms including Categorical Boosting (CatBoost), Random Forest (RF), Support Vector Machine (SVM), Neural Network (NN), Gradient Boosting Machine (GBM), K-Nearest Neighbors (KNNs), Multi-Layer Perceptron (MLP), Naive Bayes (NB), eXtreme Gradient Boosting (XGBoost) and Logistic Regression (LR) were used to build the models. Performance was measured using area under the curve (AUC) and accuracy. Finally, interpretable machine learning (IML) models presenting as Shapley additive explanation (SHAP) values were applied for mortality risk prediction. Results: A total of 1826 elderly patients with IS admitted to the ICU were included in the analysis, of whom 624 (34.2%) died, and endovascular treatment was performed in 244 patients. After feature selection, a total of eight variables, including minimum Glasgow Coma Scale values, albumin, lactate dehydrogenase, age, alkaline phosphatase, body mass index, platelets, and types of surgery, were finally used for model construction. The AUCs of the CatBoost model were 0.737 in the testing set and 0.709 in the external validation set. The Brier scores in the training set and testing set were 0.12 and 0.21, respectively. The IML of the CatBoost model was performed based on the SHAP value and the Local Interpretable Model-Agnostic Explanations method. Conclusion: The CatBoost model had the best predictive performance for predicting mortality in elderly patients with IS admitted to the ICU. The IML model would further aid in clinical decision-making and timely healthcare services by the early identification of high-risk patients.

Risk factors associated with adverse pregnancy outcomes and postpartum exacerbation in women with myasthenia gravis.

PROBLEM: Myasthenia gravis (MG), an autoimmune neuromuscular disease affecting women of childbearing age, exerts an impact on pregnancy, and vice versa. The purposes of the study were to evaluate adverse pregnancy outcomes and postpartum exacerbation in a cohort of Asian MG women, and to explore the predictors for these outcomes. METHODS OF STUDY: Thirty-seven MG pregnancies of 33 women followed in Xiangya and the second Xiangya hospitals between January 2012 and January 2022, were included in this study. Baseline maternal data, maternal complications, and neonatal outcomes were extracted from medical records. MG courses were evaluated during pregnancy and postpartum 1 year. RESULTS: In 5.4% of cases, MG exacerbation was reported during gestation, mostly in the third trimester, and in 38.9% in the postpartum period. Maternal complications were measured in 59.5% of women, gestational diabetes mellitus (GDM) taking the lead (29.7%) followed by premature rupture of membranes (PROMs) (18.9%). Transient neonatal MG (TNMG) and hyperbilirubinemia (HB) were seen in 24.3% of newborns. Body mass index (BMI) was the only independent predictor for maternal obstetric complications (p = .017), while thyroid disorders for GDM (p = .006). Younger mothers tended to give birth to babies with TNMG (p = .015). Primipara was the only risk factor for HB (p = .015). Higher gestational BMI gain (GBG) (p = .049) and without thyroid disorder (p = .017) were independent risk factors for puerperal exacerbation. Activated partial thromboplastin time and thyroid-stimulating hormone levels could be reliable to predict puerperal exacerbation. CONCLUSIONS: Most MG patients have unaffected courses during pregnancy but face a higher rate of maternal and fetal complications. Risk factors identified in our study aid the management of pregnancy in MG women.

Carbon nanotube-based flexible high-speed circuits with sub-nanosecond stage delays.

High-speed flexible circuits are required in flexible systems to realize real-time information analysis or to construct wireless communication modules for emerging applications. Here, we present scaled carbon nanotube-based thin film transistors (CNT-TFTs) with channel lengths down to 450 nm on 2-µm-thick parylene substrates, achieving state-of-the-art performances of high on-state current (187.6 µA µm-1) and large transconductance (123.3 µS µm-1). Scaling behavior analyses reveal that the enhanced performance introduced by scaling is attributed to channel resistance reduction while the contact resistance (180 ± 50 kΩ per tube) remains unchanged, which is comparable to that achieved in devices on rigid substrates, indicating great potential in ultimate scaled flexible CNT-TFTs with high performance comparable to their counterparts on rigid substrates where contact resistance dominates the performance. Five-stage flexible ring oscillators are built to benchmark the speed of scaled devices, demonstrating a 281 ps stage delay at a low supply voltage of 2.6 V.

Twenty-eight-day in-hospital mortality prediction for elderly patients with ischemic stroke in the intensive care unit: Interpretable machine learning models.

Background: Risk stratification of elderly patients with ischemic stroke (IS) who are admitted to the intensive care unit (ICU) remains a challenging task. This study aims to establish and validate predictive models that are based on novel machine learning (ML) algorithms for 28-day in-hospital mortality in elderly patients with IS who were admitted to the ICU. Methods: Data of elderly patients with IS were extracted from the electronic intensive care unit (eICU) Collaborative Research Database (eICU-CRD) records of those elderly patients admitted between 2014 and 2015. All selected participants were randomly divided into two sets: a training set and a validation set in the ratio of 8:2. ML algorithms, such as Naïve Bayes (NB), eXtreme Gradient Boosting (xgboost), and logistic regression (LR), were applied for model construction utilizing 10-fold cross-validation. The performance of models was measured by the area under the receiver operating characteristic curve (AUC) analysis and accuracy. The present study uses interpretable ML methods to provide insight into the model's prediction and outcome using the SHapley Additive exPlanations (SHAP) method. Results: As regards the population demographics and clinical characteristics, the analysis in the present study included 1,236 elderly patients with IS in the ICU, of whom 164 (13.3%) died during hospitalization. As regards feature selection, a total of eight features were selected for model construction. In the training set, both the xgboost and NB models showed specificity values of 0.989 and 0.767, respectively. In the internal validation set, the xgboost model identified patients who died with an AUC value of 0.733 better than the LR model which identified patients who died with an AUC value of 0.627 or the NB model 0.672. Conclusion: The xgboost model shows the best predictive performance that predicts mortality in elderly patients with IS in the ICU. By making the ML model explainable, physicians would be able to understand better the reasoning behind the outcome.

Factors affecting minimal manifestation status induction in myasthenia gravis.

Background: Minimal manifestation status (MMS) is an important landmark in the treatment of myasthenia gravis (MG), and predictors of MMS induction have rarely been identified in previous studies. Objective: The objective of this study is to evaluate the clinical factors associated with MMS induction among patients with MG. Design: This two-step retrospective cohort study with a single center investigated the factors that may be associated with MMS induction and retested these predictors in a test cohort. Methods: A total of 388 diagnosed MG patients who visited Xiangya Hospital between 1 July 2015 and 1 July 2019 were involved. We performed detailed chart reviews and recorded all cases achieving MMS. Demographics and clinical characteristics were also collected and their relationships to achieving MMS were investigated. Results: MMS was achieved in 124 patients (50.2%), and the median time to achieve MMS was 26 months. Several factors were found to be associated with MMS induction in exploring cohort, including muscle-specific tyrosine-protein kinase receptor (MuSK) antibody positivity (adjusted hazard ratio, HR = 4.333, 95% confidence interval, CI: 1.862-10.082), isolated ocular involvement (adjusted HR = 1.95, 95% CI: 1.284-2.961), and low baseline quantitative myasthenia gravis score (QMG score; adjusted HR = 2.022, 95% CI: 1.086-3.764). These factors were then retested in the test cohort. Isolated ocular involvement or low baseline QMG scores were factors found to be beneficial for MMS induction were confirmed. Conclusion: Isolated ocular involvement and low baseline QMG score are predictors of MMS induction in MG patients.

Tacrolimus as Single-Agent Immunotherapy and Minimal Manifestation Status in Nonthymoma Myasthenia Gravis.

BACKGROUND: Tacrolimus is a second-line immunosuppressant in myasthenia gravis (MG) therapy, which is mainly used in combination with corticosteroids to reduce steroid dose and maintain the effect of immunotherapy. However, few studies have focused on the effect of tacrolimus as single-agent immunotherapy on achieving minimal manifestation status (MMS). Thus, this study is aimed at exploring the efficacy and influencing factors of tacrolimus as single-agent immunotherapy in MG. METHODS: Clinical data of 75 nonthymoma MG patients treated with tacrolimus single-agent as initial immunotherapy were retrospectively analyzed. The therapeutic effect was evaluated by Myasthenia Gravis Foundation of America postintervention status. Clinical factors affecting the achievement of MMS and treatment reactivity of different MG subtypes were determined by Cox regression analysis. RESULTS: Tacrolimus was generally safe, with only two patients (2.7%) switching medications due to side effects. 32% of patients had improved symptoms after 1 month of treatment. 69.2% of patients achieved MMS or better after one year. The age < 39 years old, QMG score < 11 points, and AChR - Ab titer < 8.07 nmol/L were indicative of a favorable response, which was independent of gender, course of the disease. As for MG subtypes, ocular and seronegative MG showed better treatment sensitivity. CONCLUSIONS: Tacrolimus as single-agent immunotherapy takes effect quickly and can effectively enable nonthymoma MG patients to achieve MMS. Tacrolimus can be used alone for the initial immunotherapy of MG patients, especially for young, mild, and low antibody titer patients.

Clinical Evaluation of the Efficacy and Safety of Co-Administration of Wuzhi Capsule and Tacrolimus in Adult Chinese Patients with Myasthenia Gravis.

BACKGROUND: Tacrolimus has been recommended as an effective immunosuppressant for patients with myasthenia gravis (MG), while the high price, variable bioavailability, and narrow therapeutic window restrict its clinical application. Wuzhi capsule (WZC) could improve tacrolimus blood concentration by inhibiting the metabolism of cytochrome P450 3A (CYP3A) and P-glycoprotein (P-gp). There are few studies focused on the coadministration of WZC and tacrolimus in autoimmune diseases. This study was aimed at quantifying the efficacy and safety of coadministration of WZC and tacrolimus in adult Chinese patients with MG. METHODS: In this retrospective study, 122 patients with MG on tacrolimus were enrolled. The initial tacrolimus dose was 2 mg/d. Patients with standard initial tacrolimus concentration were classified into group A (standard-dose group). Those failed to reach target concentration were divided into group B (high-dose group) and group C (co-administering WZC group), according to treatment adjustment of increasing tacrolimus dose and co-administration of WZC, respectively. A logistic analysis was used to identify factors associated with clinical outcome. Adverse drug reactions (ADRs) were recorded for safety analysis. RESULTS: The tacrolimus concentration after coadministration of WZC was remarkably increased. It was higher compared with simply increasing the tacrolimus dose (p<0.001). The multivariate logistic analysis indicated that the baseline quantitative MG score was a predictive factor for clinical outcomes (OR=0.189; 95% CI 0.082-0.436; p<0.001). Fourteen patients (11.5%) reported ADRs after tacrolimus therapy. ADRs incidence was not related to WZC coadministration. CONCLUSION: The coadministration of WZC and tacrolimus can substantially elevate the tacrolimus concentration. It is a safe and economic treatment for adult Chinese patients with MG. Patients with a worse disease condition tend to present a better clinical outcome after tacrolimus therapy.

Evaluation of outcome measures for myasthenia gravis subgroups.

INTRODUCTION: Disease evaluation and long-term follow-up of myasthenia gravis (MG) patients rely on disease-specific measures. We evaluated four widely used MG-specific assessments, and compared the response to disease change in different MG subgroups. METHODS: We used the Cronbach's α coefficient to test reliability, Pearson correlation coefficients to test construct validity, as well as one-way ANOVA and independent-sample t-tests to access discriminant validity. Analyses of similar items between QMG and MG-ADL included paired-sample t-tests and mean score comparisons. Pearson correlation coefficients were used to describe the correlation between changes of QMG, MG-ADL, MG-QOL15r and MGC. The Wilcoxon matched-pairs signed-ranks test was performed to compare the outcomes. RESULTS: 872 MG patients were enrolled. QMG, MG-ADL, MG-QOL15r, and MGC all exhibited high reliability. All four scales displayed good discriminant validity according to the MGFA classification and MGC score. MG-ADL showed significant differences between patients grouped by age and gender, and MG-QOL15r showed significant differences between patients grouped by age. Analyses of similar items showed that MG-ADL achieved higher scores in bulbar items, whereas QMG produced higher scores in limb items. For patients in remission or minimal manifestation status, QMG exhibited significantly greater improvement than MG-QOL15r. In patients of MGFA I, II, III, and IV, QMG showed significantly greater improvement than MG-ADL. CONCLUSIONS: Patient-reported scale is an important supplement for a given period. MG-ADL has a better response to severe disease, and MG-QOL15r is more comprehensive for patients in remission or minimal manifestation status.

Single-cell RNA-Seq reveals transcriptional heterogeneity and immune subtypes associated with disease activity in human myasthenia gravis.

Myasthenia gravis (MG) is a rare autoimmune disease. Although the impact of immune cell disorder in MG has been extensively studied, little is known about the transcriptomes of individual cells. Here, we assessed the transcriptional profiles of 39,243 cells by single-cell sequencing and identified 13 major cell clusters, along with 39 subgroups of cells derived from patients with new-onset myasthenia gravis and healthy controls. We found that B cells, CD4+ T cells, and monocytes exhibited more heterogeneity in MG patients. CD4+ T cells were expanded in MG patients. We reclustered B cells and CD4+ T cells, and predict their essential regulators. Further analyses demonstrated that B cells in MG exhibited higher transcriptional activity towards plasma cell differentiation, CD4+ T cell subsets were unbalanced, and inflammatory pathways of monocytes were highly activated. Notably, we discovered a disease-relevant subgroup, CD180- B cells. Increased CD180- B cells in MG are indicative of a high IgG composition and were associated with disease activity and the anti-AChR antibody. Together, our data further the understanding of the cellular heterogeneity involved in the pathogenesis of MG and provide large cell-type-specific markers for subsequent research.

Peripheral B Cell Subsets in Autoimmune Diseases: Clinical Implications and Effects of B Cell-Targeted Therapies.

Antibody-secreting cells (ASCs) play a fundamental role in humoral immunity. The aberrant function of ASCs is related to a number of disease states, including autoimmune diseases and cancer. Recent insights into activated B cell subsets, including naïve B cell to ASC stages and their resultant cellular disturbances, suggest that aberrant ASC differentiation occurs during autoimmune diseases and is closely related to disease severity. However, the mechanisms underlying highly active ASC differentiation and the B cell subsets in autoimmune patients remain undefined. Here, we first review the processes of ASC generation. From the perspective of novel therapeutic target discovery, prediction of disease progression, and current clinical challenges, we further summarize the aberrant activity of B cell subsets including specialized memory CD11chiT-bet+ B cells that participate in the maintenance of autoreactive ASC populations. An improved understanding of subgroups may also enhance the knowledge of antigen-specific B cell differentiation. We further discuss the influence of current B cell therapies on B cell subsets, specifically focusing on systemic lupus erythematosus, rheumatoid arthritis, and myasthenia gravis.

Recent Advances in Flexible and Stretchable Sensing Systems: From the Perspective of System Integration.

Biological signals generated during various biological processes are critically important for providing insight into the human physiological status. Recently, there have been many great efforts in developing flexible and stretchable sensing systems to provide biological signal monitoring platforms with intimate integration with biological surfaces. Here, this review summarizes the recent advances in flexible and stretchable sensing systems from the perspective of electronic system integration. A comprehensive general sensing system architecture is described, which consists of sensors, sensor interface circuits, memories, and digital processing units. The subsequent content focuses on the integration requirements and highlights some advanced progress for each component. Next, representative examples of flexible and stretchable sensing systems for electrophysiological, physical, and chemical information monitoring are introduced. This review concludes with an outlook on the remaining challenges and opportunities for future fully flexible or stretchable sensing systems.

Prognostic Analysis of Thymoma-Associated Myasthenia Gravis (MG) in Chinese Patients and Its Implication of MG Management: Experiences from a Tertiary Hospital.

BACKGROUND: Myasthenia gravis (MG) is an autoantibody-mediated neuromuscular disorder. Approximately 10-20% of all MG patients experience thymoma (benign tumor arising from thymus tissue). Thymectomy has been the standard of care for thymomatous myasthenia gravis (TMG). However, the clinical outcome of TMG after thymectomy has not been sufficiently studied, especially the long-term prognosis. Therefore, the aim of this study was to analyze the clinical characteristics contributing to the prognostic factors of TMG. METHODS: We reviewed 70 TMG patients in the Xiangya Hospital and classified them into the minimal manifestation (MM) group and No MM group, according to the long-term treatment outcome. MM-or-better status was defined as the goal treatment for TMG patients. We collected and analyzed the demographic data, the WHO classification of thymoma, MG-associated antibody levels, disease severity, treatment-related data as well as clinical outcome at six months. Variables selected by univariate analysis were used in the multivariate logistic regression model to identify the prognostic factors. RESULTS: The differences in clinical outcome at six months and worst QMGS were significant, while the differences in other factors were insignificant between groups. Clinical outcome at six months (OR=23.5 95% CI 2.4-231.5, P=0.007) and dyspnea before thymectomy (OR=0.2, 95% CI 0.03-0.75, P=0.021) were identified as the prognostic factors of long-term treatment. CONCLUSION: Demographic and clinical features were similar in TMG patients treated at our hospital. The early achievement of MM-or-better status may indicate a good outcome in the long term. Dyspnea before thymectomy appears to associate with a poor prognosis.

Glucose metabolism pattern of peripheral blood immune cells in myasthenia gravis patients.

BACKGROUND: We investigated the correlation between glucose metabolism patterns of different immune cells and the metabolic regulatory signaling pathways in myasthenia gravis (MG) and aimed to identify therapeutic targets for MG. METHODS: We isolated peripheral blood mononuclear cells (PBMCs) and sorted CD19+B cells, dendritic cells (DCs), CD4+ T cells, CD8+ T cells, CD4+CD25+ regulatory T cells (Tregs), CD4+CD25-T cells, and T helper (Th) cells such as Th1, Th2, and Th17 cells. Then, we detected the expression levels of PI3K/AKT/mTOR-HIF-1α, GLUT1, hexokinase (HK), phosphofructokinase (PFK), and pyruvate kinase (PK) by RT-PCR, measured the oxygen consumption rate and extracellular acidification rate of ex vivo freshly sorted cells using the Seahorse XFe96 Analyzer. In addition, we compared the glycolysis levels using these cells from the same MG patients. By performing in vitro experiments, we measured, the mRNA expression levels of mTOR, HIF-1α, B cell activating factor receptor (BAFF-R), GLUT1, HK, PFK, and PK, in addition to ECAR profiles, frequency of CD80 and CD86, and IgG levels from the culture supernatant of B cells (isolated from MG patients) treated with rapamycin and PX-478 (selective mTOR and HIF-1α inhibitor, respectively) from. RESULTS: Except PBMCs, Th2 and CD8+ T cells, the expression levels of the key enzymes involved in glycolysis and HIF-1α were significantly higher in B cells, DCs, Tregs, CD4+CD25-T cells, and Th1 and Th17 cells in MG patients, and the measurement of ECAR and OCR confirmed the metabolic status. In MG patients, B cells and DCs showed significantly higher levels of glycolysis and glycolytic capacity than CD8+ T cells, CD4+ T cells and its subsets. In vitro, except IgG levels, the increased glycolysis levels, expression of key glycolytic enzymes, BAFF-R and frequency of CD80 and CD86 of B cells, could be inhibited by rapamycin and PX-478. CONCLUSIONS: Different subtypes of immune cells in MG exhibit different glucose metabolism patterns. The mTOR-HIF-1α signaling pathway might be the immunometabolism reprogramming checkpoint of glycolysis-dependent activated B cells in MG.

The Application of Machine Learning Techniques in Clinical Drug Therapy.

INTRODUCTION: The development of a novel drug is an extremely complicated process that includes the target identification, design and manufacture, and proper therapy of the novel drug, as well as drug dose selection, drug efficacy evaluation, and adverse drug reaction control. Due to the limited resources, high costs, long duration, and low hit-to-lead ratio in the development of pharmacogenetics and computer technology, machine learning techniques have assisted novel drug development and have gradually received more attention by researchers. METHODS: According to current research, machine learning techniques are widely applied in the process of the discovery of new drugs and novel drug targets, the decision surrounding proper therapy and drug dose, and the prediction of drug efficacy and adverse drug reactions. RESULTS AND CONCLUSION: In this article, we discussed the history, workflow, and advantages and disadvantages of machine learning techniques in the processes mentioned above. Although the advantages of machine learning techniques are fairly obvious, the application of machine learning techniques is currently limited. With further research, the application of machine techniques in drug development could be much more widespread and could potentially be one of the major methods used in drug development.

SNPs affecting the clinical outcomes of regularly used immunosuppressants.

Recent studies have suggested that genomic diversity may play a key role in different clinical outcomes, and the importance of SNPs is becoming increasingly clear. In this article, we summarize the bioactivity of SNPs that may affect the sensitivity to or possibility of drug reactions that occur among the signaling pathways of regularly used immunosuppressants, such as glucocorticoids, azathioprine, tacrolimus, mycophenolate mofetil, cyclophosphamide and methotrexate. The development of bioinformatics, including machine learning models, has enabled prediction of the proper immunosuppressant dosage with minimal adverse drug reactions for patients after organ transplantation or for those with autoimmune diseases. This article provides a theoretical basis for the personalized use of immunosuppressants in the future.