Distributed non-disclosive validation of predictive models by a modified ROC-GLM.

BACKGROUND: Distributed statistical analyses provide a promising approach for privacy protection when analyzing data distributed over several databases. Instead of directly operating on data, the analyst receives anonymous summary statistics, which are combined into an aggregated result. Further, in discrimination model (prognosis, diagnosis, etc.) development, it is key to evaluate a trained model w.r.t. to its prognostic or predictive performance on new independent data. For binary classification, quantifying discrimination uses the receiver operating characteristics (ROC) and its area under the curve (AUC) as aggregation measure. We are interested to calculate both as well as basic indicators of calibration-in-the-large for a binary classification task using a distributed and privacy-preserving approach. METHODS: We employ DataSHIELD as the technology to carry out distributed analyses, and we use a newly developed algorithm to validate the prediction score by conducting distributed and privacy-preserving ROC analysis. Calibration curves are constructed from mean values over sites. The determination of ROC and its AUC is based on a generalized linear model (GLM) approximation of the true ROC curve, the ROC-GLM, as well as on ideas of differential privacy (DP). DP adds noise (quantified by the ℓ 2 sensitivity Δ 2 ( f ^ ) ) to the data and enables a global handling of placement numbers. The impact of DP parameters was studied by simulations. RESULTS: In our simulation scenario, the true and distributed AUC measures differ by Δ AUC < 0.01 depending heavily on the choice of the differential privacy parameters. It is recommended to check the accuracy of the distributed AUC estimator in specific simulation scenarios along with a reasonable choice of DP parameters. Here, the accuracy of the distributed AUC estimator may be impaired by too much artificial noise added from DP. CONCLUSIONS: The applicability of our algorithms depends on the ℓ 2 sensitivity Δ 2 ( f ^ ) of the underlying statistical/predictive model. The simulations carried out have shown that the approximation error is acceptable for the majority of simulated cases. For models with high Δ 2 ( f ^ ) , the privacy parameters must be set accordingly higher to ensure sufficient privacy protection, which affects the approximation error. This work shows that complex measures, as the AUC, are applicable for validation in distributed setups while preserving an individual's privacy.

An alternative parameterization for the binormal ROC curve, with applications to sizing and simulation studies.

Because the conventional binormal ROC curve parameters are in terms of the underlying normal diseased and nondiseased rating distributions, transformations of these values are required for the user to understand what the corresponding ROC curve looks like in terms of its shape and size. In this paper I propose an alternative parameterization in terms of parameters that explicitly describe the shape and size of the ROC curve. The proposed two parameters are the mean-to-sigma ratio and the familiar area under the ROC curve (AUC), which are easily interpreted in terms of the shape and size of the ROC curve, respectively. In addition, the mean-to-sigma ratio describes the degree of improperness of the ROC curve and the AUC describes the ability of the corresponding diagnostic test to discriminate between diseased and nondiseased cases. The proposed parameterization simplifies the sizing of diagnostic studies when conjectured variance components are used and simplifies choosing the binormal a and b parameter values needed for simulation studies.

Comparing multi-class classifier performance by multi-class ROC analysis: A nonparametric approach.

The area under the Receiver Operating Characteristic (ROC) curve (AUC) is a standard metric for quantifying and comparing binary classifiers. Real world applications often require classification into multiple (more than two) classes. For multi-class classifiers that produce class membership scores, a popular multi-class AUC (MAUC) variant is to average the pairwise AUC values [1]. Due to the complicated correlation patterns, the variance of MAUC is often estimated numerically using resampling techniques. This work is a generalization of DeLong's nonparameteric approach for binary AUC analysis [2] to MAUC. We first derive the closed-form expression of the covariance matrix of the pairwise AUCs within a single MAUC. Then by dropping higher order terms, we obtain an approximate covariance matrix with a compact, matrix factorization form, which then serves as the basis for variance estimation of a single MAUC. We further extend this approach to estimate the covariance of correlated MAUCs that arise from multiple competing classifiers. For the special case of binary correlated AUCs, our results coincide with that of DeLong. Our numerical studies confirm the accuracy of the variance and covariance estimates. We provide the source code of the proposed covariance estimation of correlated MAUCs on GitHub (https://tinyurl.com/euj6wvsz) for its easy adoption by machine learning and statistical analysis packages to quantify and compare multi-class classifiers.

Data mining approaches to pneumothorax detection: Integrating mask-RCNN and medical transfer learning techniques.

With the medical condition of pneumothorax, also known as collapsed lung, air builds up in the pleural cavity and causes the lung to collapse. It is a critical disorder that needs to be identified and treated right as it can cause breathing difficulties, low blood oxygen levels, and, in extreme circumstances, death. Chest X-rays are frequently used to diagnose pneumothorax. Using the Mask R-CNN model and medical transfer learning, the proposed work offers•A novel method for pneumothorax segmentation from chest X-rays.•A method that takes advantage of the Mask R-CNN architecture's for object recognition and segmentation.•A modified model to address the issue of segmenting pneumothoraxes and then polish it using a sizable dataset of chest X-rays. The proposed method is tested against other pneumothorax segmentation techniques using a dataset of 'chest X-rays' with 'pneumothorax annotations. The test findings demonstrate that proposed method outperforms other cutting-edge techniques in terms of segmentation accuracy and speed. The proposed method could lead to better patient outcomes by increasing the precision and effectiveness of pneumothorax diagnosis and therapy. Proposed method also benefits other medical imaging activities by using the medical transfer learning approaches which increases the precision of computer-aided diagnosis and treatment planning.

Uncovering co-regulatory modules and gene regulatory networks in the heart through machine learning-based analysis of large-scale epigenomic data.

The availability of large-scale epigenomic data from various cell types and conditions has yielded valuable insights for evaluating and learning features predicting the co-binding of transcription factors (TF). However, prior attempts to develop models predicting motif co-occurrence lacked scalability for globally analyzing any motif combination or making cross-species predictions. Moreover, mapping co-regulatory modules (CRM) to gene regulatory networks (GRN) is crucial for understanding underlying function. Currently, no comprehensive pipeline exists for large-scale, rapid, and accurate CRM and GRN identification. In this study, we analyzed and evaluated different TF binding characteristics facilitating biologically significant co-binding to identify all potential clusters of co-binding TFs. We curated the UniBind database, containing ChIP-Seq data from over 1983 samples and 232 TFs, and implemented two machine learning models to predict CRMs and the potential regulatory networks they operate on. Two machine learning models, Convolution Neural Networks (CNN) and Random Forest Classifier(RFC), used to predict co-binding between TFs, were compared using precision-recall Receiver Operating Characteristic (ROC) curves. CNN outperformed RFC (AUC 0.94 vs. 0.88) and achieved higher F1 scores (0.938 vs. 0.872). The CRMs generated by the clustering algorithm were validated against ChipAtlas and MCOT, revealing additional motifs forming CRMs. We predicted 200k CRMs for 50k+ human genes, validated against recent CRM prediction methods with 100% overlap. Further, we narrowed our focus to study heart-related regulatory motifs, filtering the generated CRMs to report 1784 Cardiac CRMs containing at least four cardiac TFs. Identified cardiac CRMs revealed potential novel regulators like ARID3A and RXRB for SCAD, including known TFs like PPARG for F11R. Our findings highlight the importance of the NKX family of transcription factors in cardiac development and provide potential targets for further investigation in cardiac disease.

On the bias in the AUC variance estimate.

The area under the Receiver Operating Characteristic (ROC) curve (AUC) is a standard metric for quantifying and comparing binary classifiers. A popular approach to estimating the AUCs and the associated variabilities - the variance of the AUC or the full covariance matrix of multiple correlated AUCs - is the one proposed by DeLong et al [1], which is based on the Mann Whitney two-sample U-statistics. The bias of a variance estimator is an important factor in applications such as hypothesis testing and construction of confidence intervals - a negatively biased variance estimator may lead to incorrect conclusions, and a positive bias is conservative hence preferable. In this work, we show that the (co-)variance estimate in DeLong's approach is always positively biased. More specifically, the difference matrix between the expectation of the estimated covariance and the true covariance is a positive semi-definite matrix. This bias is non-negligible when the sample size is small, and quickly diminishes as the sample size increases. Our method relies on constructing, from the AUC kernel, a random variable whose (co-)variance matrix coincides with the bias, thereby establishing the claim. We also discuss alternative approaches to AUC variance estimation that may potentially reduce the bias.

Relationship between Obuchowski-Rockette-Hillis and Gallas methods for analyzing multi-reader diagnostic imaging data with empirical AUC as the reader performance measure.

For analyzing multireader multicase (MRMC) diagnostic imaging data when the reader performance measure of interest is the area under the receiver-operating-characteristic curve (AUC), two popular methods of analysis that allow conclusions to generalize to both the reader and case populations are the method developed by Obuchowski, Rockette and Hillis (ORH) and the method primarily developed by Gallas (Gallas). While the ORH method is a general method that is applicable to most reader performance metrics, the Gallas method is limited to those metrics for which an unbiased variance estimate exists. Previously it was not known if the ORH method could be adapted so as to produce the same variance estimate as the Gallas method. In this paper I show that a recently proposed version of the OR method produces the same unconstrained variance statistic as the Gallas method. However, the two methods differ in their approaches to constraining the variance estimate to be nonnegative and in their degrees-of-freedom estimates. These two differences are discussed and recommendations given. In addition, several contributions to the development of the ORH method are made, including determining sufficient conditions for unbiased variance estimates and providing justification for the ORH variance constraints and covariance estimation method.

Statistical inference for the two-sample problem under likelihood ratio ordering, with application to the ROC curve estimation.

The receiver operating characteristic (ROC) curve is a powerful statistical tool and has been widely applied in medical research. In the ROC curve estimation, a commonly used assumption is that larger the biomarker value, greater severity the disease. In this article, we mathematically interpret "greater severity of the disease" as "larger probability of being diseased." This in turn is equivalent to assume the likelihood ratio ordering of the biomarker between the diseased and healthy individuals. With this assumption, we first propose a Bernstein polynomial method to model the distributions of both samples; we then estimate the distributions by the maximum empirical likelihood principle. The ROC curve estimate and the associated summary statistics are obtained subsequently. Theoretically, we establish the asymptotic consistency of our estimators. Via extensive numerical studies, we compare the performance of our method with competitive methods. The application of our method is illustrated by a real-data example.

Prognostic utility of albumin-bilirubin grade in Japanese patients with primary biliary cholangitis.

Background & Aims: The albumin-bilirubin (ALBI) score is calculated using serum levels of total bilirubin and albumin as a simple method to assess liver function. This study investigated the ability of baseline ALBI score/grade measurements to assess histological stage and disease progression in individuals with primary biliary cholangitis (PBC) in a large Japanese nationwide cohort. Methods: A total of 8,768 Japanese patients with PBC were enrolled between 1980 and 2016 from 469 institutions, among whom 83% received ursodeoxycholic acid (UDCA) only, 9% received UDCA and bezafibrate, and 8% were given neither drug. Baseline clinical and laboratory parameters were retrospectively retrieved and reviewed from a central database. Associations of ALBI score/grade with histological stage, mortality, and need for liver transplantation (LT) were evaluated using Cox proportional hazards models. Results: During the median follow-up period of 5.3 years, 1,227 patients died (including 789 from liver-related causes) and 113 underwent LT. ALBI score and ALBI grade were significantly associated with Scheuer's classification (both p <0.0001). ALBI grade 2 or 3 had significant associations with all-cause mortality or need for LT as well as liver-related mortality or need for LT according to Cox proportional hazards regression analysis (hazard ratio 3.453, 95% CI 2.942-4.052 and hazard ratio 4.242, 95% CI 3.421-5.260, respectively; both p <0.0001). Cumulative LT-free survival rates at 5 years in the ALBI grade 1, 2, and 3 groups were 97.2%, 82.4%, and 38.8%, respectively, while respective non-liver-related survival rates were 98.1%, 86.0%, and 42.0% (both p <0.0001, log-rank test). Conclusions: This large nationwide study of patients with PBC suggested that baseline measurements of ALBI grade were a simple non-invasive predictor of prognosis in PBC. Impact and implications: Primary biliary cholangitis (PBC) is an autoimmune liver disease characterized by progressive destruction of intrahepatic bile ducts. This study examined the ability of albumin-bilirubin (ALBI) score/grade to estimate histological findings and disease progression in PBC by means of a large-scale nationwide cohort in Japan. ALBI score/grade were significantly associated with Scheuer's classification stage. Baseline ALBI grade measurements may be a simple non-invasive predictor of prognosis in PBC.

Artificial intelligence to predict outcomes of head and neck radiotherapy.

Head and neck radiotherapy induces important toxicity, and its efficacy and tolerance vary widely across patients. Advancements in radiotherapy delivery techniques, along with the increased quality and frequency of image guidance, offer a unique opportunity to individualize radiotherapy based on imaging biomarkers, with the aim of improving radiation efficacy while reducing its toxicity. Various artificial intelligence models integrating clinical data and radiomics have shown encouraging results for toxicity and cancer control outcomes prediction in head and neck cancer radiotherapy. Clinical implementation of these models could lead to individualized risk-based therapeutic decision making, but the reliability of the current studies is limited. Understanding, validating and expanding these models to larger multi-institutional data sets and testing them in the context of clinical trials is needed to ensure safe clinical implementation. This review summarizes the current state of the art of machine learning models for prediction of head and neck cancer radiotherapy outcomes.

Nomogram for predicting the probability of permanent stoma in patients with acute obstructive colorectal cancer.

BACKGROUND: Acute obstructive colorectal cancer is a high-risk emergency among colorectal cancer (CRC). Approximately 20% of CRC patients are associated with a permanent stoma, which greatly affects the lifestyle of patients. This study aimed to investigate risk factors for predicting permanent stoma (PS) in patients with acute obstructive colorectal cancer. METHODS: We retrospectively analyzed the clinical-pathological features of patients with acute obstructive colorectal cancer who underwent treatments from our hospital between January 2015 and December 2020. Univariate and multivariate logistic regression analyses were used to evaluate the risk factors for predicting PS chances of CRC patients using a nomogram method. Furthermore, the operating characteristic (ROC) curve and area under the ROC curve (AUC) were used to assess the discrimination power of the nomogram. Calibration plot was used to evaluate nomogram's calibration. RESULTS: A total of 98 patients with acute obstructive colorectal cancer were enrolled in this study, including 24 PS patients with permanent stoma and 74 non-PS patients. Multivariate analysis showed that age [odds ratio (OR): 1.068, 95% confidence interval (CI): 1.006 ~ 1.135, P = 0.032], carcinoembryonic antigen (CEA) [OR: 1.015, 95% CI: 1.003 ~ 1.028, P = 0.013], and surgical method [emergency group vs. stent group, OR: 14.066, 95% CI: 3.625 ~ 54.572, p < 0.001] were independent risk factors for PS. These risk factors were incorporated into a nomogram and showed that the AUC of the nomogram was 0.867 (95% CI: 0.782-0.951). The calibration plot got consistent with prediction for PS in the nomogram. CONCLUSION: Age, CEA, and surgical method were independent risk factors for PS in patients with acute obstructive colorectal cancer. Our nomogram has favorable predictive power for PS in CRC patients.

Deep-learning based approach to identify substrates of human E3 ubiquitin ligases and deubiquitinases.

E3 ubiquitin ligases (E3s) and deubiquitinating enzymes (DUBs) play key roles in protein degradation. However, a large number of E3 substrate interactions (ESIs) and DUB substrate interactions (DSIs) remain elusive. Here, we present DeepUSI, a deep learning-based framework to identify ESIs and DSIs using the rich information present in protein sequences. Utilizing the collected golden standard dataset, key hyperparameters in the process of model training, including the ones relevant to data sampling and number of epochs, have been systematically assessed. The performance of DeepUSI was thoroughly evaluated by multiple metrics, based on internal and external validation. Application of DeepUSI to cancer-associated E3 and DUB genes identified a list of druggable substrates with functional implications, warranting further investigation. Together, DeepUSI presents a new framework for predicting substrates of E3 ubiquitin ligases and deubiquitinates.

CT-based radiomics signature of visceral adipose tissue for prediction of disease progression in patients with Crohn's disease: A multicentre cohort study.

Background: Visceral adipose tissue (VAT) is involved in the pathogenesis of Crohn's disease (CD). However, data describing its effects on CD progression remain scarce. We developed and validated a VAT-radiomics model (RM) using computed tomography (CT) images to predict disease progression in patients with CD and compared it with a subcutaneous adipose tissue (SAT)-RM. Methods: This retrospective study included 256 patients with CD (training, n = 156; test, n = 100) who underwent baseline CT examinations from June 19, 2015 to June 14, 2020 at three tertiary referral centres (The First Affiliated Hospital of Sun Yat-Sen University, The First Affiliated Hospital of Shantou University Medical College, and The First People's Hospital of Foshan City) in China. Disease progression referred to the development of penetrating or stricturing diseases or the requirement for CD-related surgeries during follow-up. A total of 1130 radiomics features were extracted from VAT on CT in the training cohort, and a machine-learning-based VAT-RM was developed to predict disease progression using selected reproducible features and validated in an external test cohort. Using the same modeling methodology, a SAT-RM was developed and compared with the VAT-RM. Findings: The VAT-RM exhibited satisfactory performance for predicting disease progression in total test cohort (the area under the ROC curve [AUC] = 0.850, 95% confidence Interval [CI] 0.764-0.913, P < 0.001) and in test cohorts 1 (AUC = 0.820, 95% CI 0.687-0.914, P < 0.001) and 2 (AUC = 0.871, 95% CI 0.744-0.949, P < 0.001). No significant differences in AUC were observed between test cohorts 1 and 2 (P = 0.673), suggesting considerable efficacy and robustness of the VAT-RM. In the total test cohort, the AUC of the VAT-RM for predicting disease progression was higher than that of SAT-RM (AUC = 0.786, 95% CI 0.692-0.861, P < 0.001). On multivariate Cox regression analysis, the VAT-RM (hazard ratio [HR] = 9.285, P = 0.005) was the most important independent predictor, followed by the SAT-RM (HR = 3.280, P = 0.060). Decision curve analysis further confirmed the better net benefit of the VAT-RM than the SAT-RM. Moreover, the SAT-RM failed to significantly improve predictive efficacy after it was added to the VAT-RM (integrated discrimination improvement = 0.031, P = 0.102). Interpretation: Our results suggest that VAT is an important determinant of disease progression in patients with CD. Our VAT-RM allows the accurate identification of high-risk patients prone to disease progression and offers notable advantages over SAT-RM. Funding: This study was supported by the National Natural Science Foundation of China, Guangdong Basic and Applied Basic Research Foundation, Shenzhen-Hong Kong Institute of Brain Science-Shenzhen Fundamental Research Institutions, Nature Science Foundation of Shenzhen, and Young S&T Talent Training Program of Guangdong Provincial Association for S&T. Translation: For the Chinese translation of the abstract see Supplementary Materials section.

Deep learning-based automatic left atrial appendage filling defects assessment on cardiac computed tomography for clinical and subclinical atrial fibrillation patients.

Rationale and objectives: Selecting region of interest (ROI) for left atrial appendage (LAA) filling defects assessment can be time consuming and prone to subjectivity. This study aimed to develop and validate a novel artificial intelligence (AI), deep learning (DL) based framework for automatic filling defects assessment on CT images for clinical and subclinical atrial fibrillation (AF) patients. Materials and methods: A total of 443,053 CT images were used for DL model development and testing. Images were analyzed by the AI framework and expert cardiologists/radiologists. The LAA segmentation performance was evaluated using Dice coefficient. The agreement between manual and automatic LAA ROI selections was evaluated using intraclass correlation coefficient (ICC) analysis. Receiver operating characteristic (ROC) curve analysis was used to assess filling defects based on the computed LAA to ascending aorta Hounsfield unit (HU) ratios. Results: A total of 210 patients (Group 1: subclinical AF, n = 105; Group 2: clinical AF with stroke, n = 35; Group 3: AF for catheter ablation, n = 70) were enrolled. The LAA volume segmentation achieved 0.931-0.945 Dice scores. The LAA ROI selection demonstrated excellent agreement (ICC ≥0.895, p < 0.001) with manual selection on the test sets. The automatic framework achieved an excellent AUC score of 0.979 in filling defects assessment. The ROC-derived optimal HU ratio threshold for filling defects detection was 0.561. Conclusion: The novel AI-based framework could accurately segment the LAA region and select ROIs while effectively avoiding trabeculae for filling defects assessment, achieving close-to-expert performance. This technique may help preemptively detect the potential thromboembolic risk for AF patients.

A rank-based approach to design and analysis of pretest-posttest randomized trials, with application to COVID-19 ordinal scale data.

Randomized controlled trials with a pretest-posttest design frequently yield ordered categorical outcome data. Focusing on the estimation of the win probability that a treated participant would have a better score than (or win over) a control participant, we developed methods for analysis and sample size planning for such trials. We exploited the analysis of covariance framework with the dependent variable being individual participants' win fractions at posttest and the covariate being the win fractions at pretest. The win fractions were obtained using the mid-ranks of the ordinal data. Simulation evaluation based on a recent randomized trial on COVID-19 suggests that the methods perform very well. A sample SAS code for data analysis is presented.

SeqCP: A sequence-based algorithm for searching circularly permuted proteins.

Circular permutation (CP) is a protein sequence rearrangement in which the amino- and carboxyl-termini of a protein can be created in different positions along the imaginary circularized sequence. Circularly permutated proteins usually exhibit conserved three-dimensional structures and functions. By comparing the structures of circular permutants (CPMs), protein research and bioengineering applications can be approached in ways that are difficult to achieve by traditional mutagenesis. Most current CP detection algorithms depend on structural information. Because there is a vast number of proteins with unknown structures, many CP pairs may remain unidentified. An efficient sequence-based CP detector will help identify more CP pairs and advance many protein studies. For instance, some hypothetical proteins may have CPMs with known functions and structures that are informative for functional annotation, but existing structure-based CP search methods cannot be applied when those hypothetical proteins lack structural information. Despite the considerable potential for applications, sequence-based CP search methods have not been well developed. We present a sequence-based method, SeqCP, which analyzes normal and duplicated sequence alignments to identify CPMs and determine candidate CP sites for proteins. SeqCP was trained by data obtained from the Circular Permutation Database and tested with nonredundant datasets from the Protein Data Bank. It shows high reliability in CP identification and achieves an AUC of 0.9. SeqCP has been implemented into a web server available at: http://pcnas.life.nthu.edu.tw/SeqCP/.

Transcriptomic Dysregulation of Inflammation-Related Genes in Leukocytes of Patients with Gestational Diabetes Mellitus (GDM) during and after Pregnancy: Identifying Potential Biomarkers Relevant to Glycemic Abnormality.

Although the immune system has been implicated in the pathophysiology of gestational diabetes mellitus (GDM) and postpartum abnormal glucose tolerance (AGT), little is known about the transcriptional response of inflammation-related genes linked to metabolic phenotypes of GDM women during and after pregnancy, which may be potential diagnostic classifiers for GDM and biomarkers for predicting AGT. To address these questions, gene expression of IL6, IL8, IL10, IL13, IL18, TNFA, and the nuclear factor κB (NFκB)/RELA transcription factor were quantified in leukocytes of 28 diabetic women at GDM diagnosis (GDM group) and 1-year postpartum (pGDM group: 10 women with AGT and 18 normoglycemic women), using a nested RT-PCR method. Control pregnancies with normal glucose tolerance (NGT group; n = 31) were closely matched for maternal age, gestational age, pre-pregnancy BMI, pregnancy weight, and gestational weight gain. Compared with the NGT group, IL8 was downregulated in the GDM group, and IL13 and RELA were upregulated in the pGDM group, whereas IL6, IL10, and IL18 were upregulated in the GDM and pGDM groups. The TNFA level did not change from pregnancy to postpartum. Associations of some cytokines with glycemic measures were detected in pregnancy (IL6 and RELA) and postpartum (IL10) (p < 0.05). Receiver operating characteristic (ROC) curves showed that IL6, IL8, and IL18, if employed alone, can discriminate GDM patients from NGT individuals at GDM diagnosis, with the area under the ROC curves (AUCs) of 0.844, (95% CI 0.736−0.953), 0.771 (95% CI 0.651−0.890), and 0.714 (95% CI 0.582−0.846), respectively. By the logistic regression method, we also identified a three-gene panel (IL8, IL13, and TNFA) for postpartum AGT prediction. This study demonstrates a different transcriptional response of the studied genes in clinically well-characterized women with GDM at GDM diagnosis and 1-year postpartum, and provides novel transcriptomic biomarkers for future efforts aimed at diagnosing GDM and identifying the high risk of postpartum AGT groups.

Clinical and virological features of chronic hepatitis B in the French national surveillance program, 2008-2012: A cross-sectional study.

Background & Aims: Among people living with HBV, only a subset of individuals with chronic hepatitis is in need of treatment, and this proportion varies according to the population, region, and setting. No estimates of the proportion of people who are infected with HBV and meet the treatment eligibility criteria in France are available. Methods: 552 treatment-naïve individuals with chronic HBV infection referred for the first time to a hepatology reference centre between 2008 and 2012 were prospectively included. Demographic, clinical, and laboratory data were analysed. Results: In total, 61.1% of patients were males, with a median age of 37.5 years. Moreover, 64% were born in an intermediate- or high-HBV endemicity country, and 90% were HBeAg-negative. At referral, median HBV DNA and HBsAg levels were 3.3 and 3.6 log IU/ml, respectively; 37.8% of patients had alanine aminotransferase >40 U/L, and 29.0% had moderate or severe fibrosis (≥F2), including 9.4% with cirrhosis. The most prevalent genotypes were D (34.7%), E (27.4%), and A (25.7%). Coinfections were rare: 2.4% were HIV-positive, 4.0% were HCV-positive, and 6.0% were HDV-positive. According to the 2017 EASL Clinical Practice Guidelines, using a single time point analysis, 2.7% of patients were classified as HBeAg-positive chronic infection, 6.1% as HBeAg-positive chronic hepatitis B, 26.5% as HBeAg-negative chronic hepatitis B, and 61.1% as HBeAg-negative chronic infection, whereas 3.6% patients could not be classified. The performance of HBsAg level quantification to identify individuals with HBeAg-negative chronic hepatitis B was poor. A total of 29.1% met the criteria for initiation of antiviral treatment, whereas 66.5% remained under routine clinical surveillance. Most eligible patients initiated recommended first-line therapies, including tenofovir (45.3%), entecavir (36.8%), or pegylated interferon alpha (11.6%). Conclusions: Of all cases, 9.4% had cirrhosis at presentation and 29.1% met the 2017 EASL Clinical Practice Guidelines treatment criteria. HBsAg levels failed to accurately identify individuals with HBeAg-negative chronic infection. Lay summary: Among French adults chronically infected with HBV referred for the first time to hepatology reference centres, about one-third had a significant liver disease. Approximately one-third of individuals met criteria for initiation of antiviral treatment based on entecavir or tenofovir or, occasionally, pegylated interferon alpha.

MSPJ: Discovering potential biomarkers in small gene expression datasets via ensemble learning.

In transcriptomics, differentially expressed genes (DEGs) provide fine-grained phenotypic resolution for comparisons between groups and insights into molecular mechanisms underlying the pathogenesis of complex diseases or phenotypes. The robust detection of DEGs from large datasets is well-established. However, owing to various limitations (e.g., the low availability of samples for some diseases or limited research funding), small sample size is frequently used in experiments. Therefore, methods to screen reliable and stable features are urgently needed for analyses with limited sample size. In this study, MSPJ, a new machine learning approach for identifying DEGs was proposed to mitigate the reduced power and improve the stability of DEG identification in small gene expression datasets. This ensemble learning-based method consists of three algorithms: an improved multiple random sampling with meta-analysis, SVM-RFE (support vector machines-recursive feature elimination), and permutation test. MSPJ was compared with ten classical methods by 94 simulated datasets and large-scale benchmarking with 165 real datasets. The results showed that, among these methods MSPJ had the best performance in most small gene expression datasets, especially those with sample size below 30. In summary, the MSPJ method enables effective feature selection for robust DEG identification in small transcriptome datasets and is expected to expand research on the molecular mechanisms underlying complex diseases or phenotypes.

Detecting differentially expressed circular RNAs from multiple quantification methods using a generalized linear mixed model.

Finding differentially expressed circular RNAs (circRNAs) is instrumental to understanding the molecular basis of phenotypic variation between conditions linked to circRNA-involving mechanisms. To date, several methods have been developed to identify circRNAs, and combining multiple tools is becoming an established approach to improve the detection rate and robustness of results in circRNA studies. However, when using a consensus strategy, it is unclear how circRNA expression estimates should be considered and integrated into downstream analysis, such as differential expression assessment. This work presents a novel solution to test circRNA differential expression using quantifications of multiple algorithms simultaneously. Our approach analyzes multiple tools' circRNA abundance count data within a single framework by leveraging generalized linear mixed models (GLMM), which account for the sample correlation structure within and between the quantification tools. We compared the GLMM approach with three widely used differential expression models, showing its higher sensitivity in detecting and efficiently ranking significant differentially expressed circRNAs. Our strategy is the first to consider combined estimates of multiple circRNA quantification methods, and we propose it as a powerful model to improve circRNA differential expression analysis.