Differential gene expression profile in PBMCs from subjects with AERD and ATA: a gene marker for AERD

Aspirin-exacerbated respiratory disease (AERD) is associated with severe asthma and aspirin can cause asthma to worsen, often in the form of a severe and sudden attack. The oral aspirin challenge is the gold standard to confirm the diagnosis of AERD, but it is time consuming and produces serious complications in some cases. Therefore, more efficient and practical method is needed to predict AERD patients. The aim of the present study was to identify AERD-related gene expression in peripheral blood mononuclear cells (PBMCs) and examine the diagnostic potential of these candidate gene(s) for predicting AERD. To do this, RNAs from 24 subjects with AERD and 18 subjects with aspirin-tolerant asthma (ATA) were subjected to microarray analysis of ~34,560 genes. In total, 10 genes were selected as candidate gene markers by applying p ≤ 0.001(t test) and ≥8-fold change, and to correct for multiple comparisons, the false discovery rate analyses were performed. By applying multiple logistic regression analysis, among possible 1,023 models (2¹⁰–1), a model consisting of CNKSR3, SPTBN2, and IMPACT was selected as candidate set, because this set showed the best AUC (0.98) with 88 % sensitivity and 89 % specificity. For validation, mRNA levels by real-time PCR on PBMCs from two population sets in a gene-chip study and another replication sample, 20 AERD, 20 ATA, and 8 normal controls, were significantly different between groups with 100 % sensitivity and 100 % specificity in each of the two population sets. However, IMPACT gene did not differentiate between AERD and normal controls. The set of the two genes (CNKSR3 and SPTBN2) showed the best AUC (0.96) with 88 % sensitivity and 94 % specificity in a gene-chip study sample. In addition, this set showed perfect discriminative power with AUC (1.0, 100 % sensitivity and 100 % specificity) in each of the two population sets: the gene-chip samples and the replication samples. It also showed perfect discrimination for AERD from NC (AUC: 1.0) and ATA from NC (AUC: 1.0). In conclusion, we developed the two gene markers (CNKSR3 and SPTBN2) of PBMC which differentiate between AERD and ATA with a perfect discriminative power. These gene markers may be an efficient and practical method for predicting AERD.     

A Predictive Model Combining Fecal Calgranulin B and Fecal Occult Blood Tests Can Improve the Diagnosis of Colorectal Cancer

Aim

Current fecal screening tools for colorectal cancer (CRC), such as fecal occult blood tests (FOBT), are limited by their low sensitivity. Calgranulin B (CALB) was previously reported as a candidate fecal marker for CRC. This study investigated whether a combination of the FOBT and fecal CALB has increased sensitivity and specificity for a diagnosis of CRC.

Materials and Methods

Patients with CRC (n = 175), and healthy individuals (controls; n = 151) were enrolled into the development (81 cases and 51 controls) and validation (94 cases and 100 controls) sets. Stool samples were collected before bowel preparation. CALB levels were determined by western blotting. FOBT and fecal CALB results were used to develop a predictive model based on logistic regression analysis. The benefit of adding CALB to a model with only FOBT was evaluated as an increased area under the receiver operating curve (AUC), partial AUC, and reclassification improvement (RI) in cases and controls, and net reclassification improvement (NRI).

Results

Mean CALB level was significantly higher in CRC patients than in controls (P<0.001). CALB was not associated with tumor stage or cancer site, but positivity on the FOBT was significantly higher in advanced than in earlier tumor stages. At a specificity of 90%, the cross-validated AUC and sensitivity were 89.81% and 82.72%, respectively, in the development set, and 92.74% and 79.79%, respectively, in the validation set. The incremental benefit of adding CALB to the model, as shown by the increase in AUC, had a p-value of 0.0499. RI in cases and controls and NRI all revealed that adding CALB significantly improved the prediction model.

Conclusion

A predictive model using a combination of FOBT and CALB may have greater sensitivity and specificity and AUC for predicting CRC than models using a single marker.     

How citizen science could improve species distribution models and their independent assessment

1. Species distribution models (SDM) have been increasingly developed in recent years, but their validity is questioned. Their assessment can be improved by the use of independent data, but this can be difficult to obtain and prohibitive to collect. Standardized data from citizen science may be used to establish external evaluation datasets and to improve SDM validation and applicability.
2. We used opportunistic presence‐only data along with presence–absence data from a standardized citizen science program to establish and assess habitat suitability maps for 9 species of amphibian in western France. We assessed Generalized Additive and Random Forest Models’ performance by (1) cross‐validation using 30% of the opportunistic dataset used to calibrate the model or (2) external validation using different independent datasets derived from citizen science monitoring. We tested the effects of applying different combinations of filters to the citizen data and of complementing it with additional standardized fieldwork.
3. Cross‐validation with an internal evaluation dataset resulted in higher AUC (Area Under the receiver operating Curve) than external evaluation causing overestimation of model accuracy and did not select the same models; models integrating sampling effort performed better with external validation. AUC, specificity, and sensitivity of models calculated with different filtered external datasets differed for some species. However, for most species, complementary fieldwork was not necessary to obtain coherent results, as long as the citizen science data were strongly filtered.
4. Since external validation methods using independent data are considered more robust, filtering data from citizen sciences may make a valuable contribution to the assessment of SDM. Limited complementary fieldwork with volunteer''s participation to complete ecological gradients may also possibly enhance citizen involvement and lead to better use of SDM in decision processes for nature conservation.

     

A novel microRNA-based signature predicts prognosis among nasopharyngeal cancer patients

Nasopharyngeal cancer is one of the most common malignant tumors in the head and neck. Identification of promising miRNA biomarkers might benefit a lot to the detection of nasopharyngeal carcinoma. miRNA expression profile and clinical information were obtained from two microarray profiling data sets from the Gene Expression Omnibus (GEO) database. miRNA signature model was constructed via univariate Cox survival analysis, multivariate Cox survival analysis, and least absolute shrinkage and selection operator Cox regression analysis. Kaplan–Meier curve, area under the curve (AUC), decision curve analysis, Box plot, and nomogram were used to evaluate the prognosis of the model to patients. 67 up-regulated and 93 down-regulated miRNAs were identified from GEO microarray data sets (P < 0.05). A three-miRNA signature (has-miR-142-3p, has-miR-29c, and has-miR-30e) was obviously associated with the overall survival of nasopharyngeal carcinoma patients (P  < 0.001). The AUCs for the signature were 0.74, 0.7 for the training set and external validation set. The AUC of disease free survival and distant metastasis-free survival were also high. The model has better clinical independence and has better clinical prediction effect when combined with clinical characteristics (P < 0.0001). Compared with the published models, our model had a higher AUC. Our results revealed that a three-miRNA signature was a potential novel prognostic biomarker for nasopharyngeal carcinoma.Impact statementNasopharyngeal cancer is one of the most common malignant tumors in the head and neck. Identification of promising miRNA biomarkers might benefit a lot to the detection of nasopharyngeal carcinoma. A three-miRNA signature (has-miR-142-3p, has-miR-29c, and has-miR-30e) was obviously associated with the overall survival of nasopharyngeal carcinoma patients. The model has better clinical independence and has better clinical prediction effect when combined with clinical characteristics. Our results revealed that a three-miRNA signature was a potential novel prognostic biomarker for nasopharyngeal carcinoma.     

A deep learning classifier for digital breast tomosynthesis

PurposeTo develop a computerized detection system for the automatic classification of the presence/absence of mass lesions in digital breast tomosynthesis (DBT) annotated exams, based on a deep convolutional neural network (DCNN).Materials and MethodsThree DCNN architectures working at image-level (DBT slice) were compared: two state-of-the-art pre-trained DCNN architectures (AlexNet and VGG19) customized through transfer learning, and one developed from scratch (DBT-DCNN). To evaluate these DCNN-based architectures we analysed their classification performance on two different datasets provided by two hospital radiology departments. DBT slice images were processed following normalization, background correction and data augmentation procedures. The accuracy, sensitivity, and area-under-the-curve (AUC) values were evaluated on both datasets, using receiver operating characteristic curves. A Grad-CAM technique was also implemented providing an indication of the lesion position in the DBT slice.Results Accuracy, sensitivity and AUC for the investigated DCNN are in-line with the best performance reported in the field. The DBT-DCNN network developed in this work showed an accuracy and a sensitivity of (90% ± 4%) and (96% ± 3%), respectively, with an AUC as good as 0.89 ± 0.04. A k-fold cross validation test (with k = 4) showed an accuracy of 94.0% ± 0.2%, and a F1-score test provided a value as good as 0.93 ± 0.03. Grad-CAM maps show high activation in correspondence of pixels within the tumour regions.Conclusions We developed a deep learning-based framework (DBT-DCNN) to classify DBT images from clinical exams. We investigated also a possible application of the Grad-CAM technique to identify the lesion position.     

Human Salivary Micro-RNA in Patients with Parotid Salivary Gland Neoplasms

Background

Currently, clinical examination, ultrasound scanning (with or without fine needle aspiration cytology), preoperative CT-scan and MRI are available for the differential diagnosis of parotid gland swelling. A preliminary non-invasive salivary diagnostic tool may be helpful in the clinical decision making process. Altered salivary micro-RNA (miRNA) expression levels have been observed in saliva from patients with various cancers. Therefore, we investigated miRNA expression levels in saliva samples from patients with a parotid gland neoplasm using Human miRNA cards in comparison to controls.

Results

In the discovery phase, eight miRNAs were identified having different expression levels in patients compared to controls. In the validation phase, the differences in miRNA expression levels between patients and controls were confirmed for seven out of eight discovered miRNAs (p < 0.001). A combination of two miRNAs yielded a receiver-operator-characteristics curve with an AUC of 0.94 (95% CI: 0.87–1.00; sensitivity 91%; specificity 86%). Validation of discovered miRNAs in segregated collected parotid saliva revealed that expression of these miRNAs differ between whole saliva and parotid saliva.

Conclusions

A two miRNA combination can predict the presence of a parotid gland neoplasm. Furthermore, this study suggested that the identified, patient-specific, salivary miRNAs were not derived from the parotid gland itself.     

Development and Validation of a Risk-Score Model for Type 2 Diabetes: A Cohort Study of a Rural Adult Chinese Population

Some global models to predict the risk of diabetes may not be applicable to local populations. We aimed to develop and validate a score to predict type 2 diabetes mellitus (T2DM) in a rural adult Chinese population. Data for a cohort of 12,849 participants were randomly divided into derivation (n = 11,564) and validation (n = 1285) datasets. A questionnaire interview and physical and blood biochemical examinations were performed at baseline (July to August 2007 and July to August 2008) and follow-up (July to August 2013 and July to October 2014). A Cox regression model was used to weigh each variable in the derivation dataset. For each significant variable, a score was calculated by multiplying β by 100 and rounding to the nearest integer. Age, body mass index, triglycerides and fasting plasma glucose (scores 3, 12, 24 and 76, respectively) were predictors of incident T2DM. The model accuracy was assessed by the area under the receiver operating characteristic curve (AUC), with optimal cut-off value 936. With the derivation dataset, sensitivity, specificity and AUC of the model were 66.7%, 74.0% and 0.768 (95% CI 0.760–0.776), respectively. With the validation dataset, the performance of the model was superior to the Chinese (simple), FINDRISC, Oman and IDRS models of T2DM risk but equivalent to the Framingham model, which is widely applicable in a variety of populations. Our model for predicting 6-year risk of T2DM could be used in a rural adult Chinese population.     

Inverse relationship between the level of miRNA 148a-3p and both TGF-β1 and FIB-4 in hepatocellular carcinoma

Background and aimHepatocellular carcinoma (HCC) is a major health burden globally. Dysregulation of miRNA 148a-3p is engaged in carcinogenesis. TGF-β is a profibrogenic cytokine. This study assesses the expression level of miRNA 148a-3p and its relationship with serum TGF-β1 and fibrosis index based on four factors (FIB-4) in Egyptian patients with HCV-associated HCC.Subjectsand Methods: The study included 72 HCC patients with HCV, 48 HCV cirrhotic patients, and 47 healthy controls. Serum TGF-β1 was assessed by ELISA and the expression of miRNA 148a-3p was measured by RT-PCR.ResultsPatients with HCC had lower plasma miRNA 148a-3p, higher serum TGF-β1, and higher FIB-4 levels than patients with cirrhosis and controls. miRNA 148a-3p discriminated HCC either from control (AUC: 0.997, 95.83% sensitivity, 85.11% specificity) or from cirrhosis (AUC: 0.943, 91.67% sensitivity, 81.25% specificity). Moreover, it distinguished metastatic from nonmetastatic patients (AUC: 0.800, 88.89% sensitivity, 60.0% specificity). The decreased miRNA 148a-3p and the increased TGF-β1 levels were related to distant metastasis, multinodular lesions, advanced TNM stage, and BCLC score (C). A negative correlation between miRNA 148a-3p and each of FIB-4 and TGF-β1 was detected. The decreased miRNA 148a-3p was associated with poor overall survival and poor progression-free survival.ConclusionAn inverse relationship between miRNA 148a-3p and both TGF-β1 and FIB-4 was observed, which could be involved in HCC pathogenesis. Moreover, this miRNA is a potential diagnostic and prognostic biomarker for HCC.     

心衰进程中非编码RNA对线粒体功能的调控作用

心衰长久以来一直缺少有效治疗方法,给社会造成了巨大的经济和民生负担,新诊断标志物的确认和治疗方法的研发十分迫切。线粒体功能障碍与心衰发生和发展密切相关,以线粒体为基础的能量供应紊乱、钙失衡、氧化应激和细胞死亡在心衰的发展中起着重要作用,但线粒体调控的具体机制还不十分清楚。非编码RNA被证实在表观调控、转录后修饰、翻译调节等多方面发挥重要调控作用。研究表明,包括miRNA、lncRNA、circRNA在内的大量非编码RNA在心脏发育和心脏疾病发展过程中存在差异表达,并在线粒体蛋白稳态、氧化磷酸化、氧化应激、凋亡与自噬等调控中发挥了重要作用,进而影响心衰等心脏疾病的发生发展,但其详细机制尚未完全阐明。本文就近年心衰发生和发展过程中非编码RNA调控线粒体功能机制的相关研究进行综述,梳理了近年来非编码RNA在调节线粒体结构与功能进而影响心衰发展方面的研究进展,以期为心衰研究与治疗提供新的思路和靶点。     

In silico prediction of potential miRNA‐disease association using an integrative bioinformatics approach based on kernel fusion

Accumulating experimental evidence has demonstrated that microRNAs (miRNAs) have a huge impact on numerous critical biological processes and they are associated with different complex human diseases. Nevertheless, the task to predict potential miRNAs related to diseases remains difficult. In this paper, we developed a Kernel Fusion‐based Regularized Least Squares for MiRNA‐Disease Association prediction model (KFRLSMDA), which applied kernel fusion technique to fuse similarity matrices and then utilized regularized least squares to predict potential miRNA‐disease associations. To prove the effectiveness of KFRLSMDA, we adopted leave‐one‐out cross‐validation (LOOCV) and 5‐fold cross‐validation and then compared KFRLSMDA with 10 previous computational models (MaxFlow, MiRAI, MIDP, RKNNMDA, MCMDA, HGIMDA, RLSMDA, HDMP, WBSMDA and RWRMDA). Outperforming other models, KFRLSMDA achieved AUCs of 0.9246 in global LOOCV, 0.8243 in local LOOCV and average AUC of 0.9175 ± 0.0008 in 5‐fold cross‐validation. In addition, respectively, 96%, 100% and 90% of the top 50 potential miRNAs for breast neoplasms, colon neoplasms and oesophageal neoplasms were confirmed by experimental discoveries. We also predicted potential miRNAs related to hepatocellular cancer by removing all known related miRNAs of this cancer and 98% of the top 50 potential miRNAs were verified. Furthermore, we predicted potential miRNAs related to lymphoma using the data set in the old version of the HMDD database and 80% of the top 50 potential miRNAs were confirmed. Therefore, it can be concluded that KFRLSMDA has reliable prediction performance.     

Early Detection of Malignant Pleural Mesothelioma in Asbestos-Exposed Individuals with a Noninvasive Proteomics-Based Surveillance Tool

Background

Malignant pleural mesothelioma (MM) is an aggressive, asbestos-related pulmonary cancer that is increasing in incidence. Because diagnosis is difficult and the disease is relatively rare, most patients present at a clinically advanced stage where possibility of cure is minimal. To improve surveillance and detection of MM in the high-risk population, we completed a series of clinical studies to develop a noninvasive test for early detection.

Methodology/Principal Findings

We conducted multi-center case-control studies in serum from 117 MM cases and 142 asbestos-exposed control individuals. Biomarker discovery, verification, and validation were performed using SOMAmer proteomic technology, which simultaneously measures over 1000 proteins in unfractionated biologic samples. Using univariate and multivariate approaches we discovered 64 candidate protein biomarkers and derived a 13-marker random forest classifier with an AUC of 0.99±0.01 in training, 0.98±0.04 in independent blinded verification and 0.95±0.04 in blinded validation studies. Sensitivity and specificity at our pre-specified decision threshold were 97%/92% in training and 90%/95% in blinded verification. This classifier accuracy was maintained in a second blinded validation set with a sensitivity/specificity of 90%/89% and combined accuracy of 92%. Sensitivity correlated with pathologic stage; 77% of Stage I, 93% of Stage II, 96% of Stage III and 96% of Stage IV cases were detected. An alternative decision threshold in the validation study yielding 98% specificity would still detect 60% of MM cases. In a paired sample set the classifier AUC of 0.99 and 91%/94% sensitivity/specificity was superior to that of mesothelin with an AUC of 0.82 and 66%/88% sensitivity/specificity. The candidate biomarker panel consists of both inflammatory and proliferative proteins, processes strongly associated with asbestos-induced malignancy.

Significance

The SOMAmer biomarker panel discovered and validated in these studies provides a solid foundation for surveillance and diagnosis of MM in those at highest risk for this disease.     

Genome‐wide identification of urinary cell‐free microRNAs for non‐invasive detection of bladder cancer

Urinary microRNAs (miRNAs) are emerging as clinically useful tool for early and non‐invasive detection of various types of cancer including bladder cancer (BCA). In this study, 205 patients with BCA and 99 healthy controls were prospectively enrolled. Expression profiles of urinary miRNAs were obtained using Affymetrix miRNA microarrays (2578 miRNAs) and candidate miRNAs further validated in independent cohorts using qRT‐PCR. Whole‐genome profiling identified 76 miRNAs with significantly different concentrations in urine of BCA compared to controls (P < 0.01). In the training and independent validation phase of the study, miR‐31‐5p, miR‐93‐5p and miR‐191‐5p were confirmed to have significantly higher levels in urine of patients with BCA in comparison with controls (P < 0.01). We further established 2‐miRNA‐based urinary DxScore (miR‐93‐5p, miR‐31‐5p) enabling sensitive BCA detection with AUC being 0.84 and 0.81 in the training and validation phase, respectively. Moreover, DxScore significantly differed in the various histopathological subgroups of BCA and decreased post‐operatively. In conclusion, we identified and independently validated cell‐free urinary miRNAs as promising biomarkers enabling non‐invasive detection of BCA.     

Validation of chemometric models for the determination of deoxynivalenol on maize by mid-infrared spectroscopy

Validation methods for chemometric models are presented, which are a necessity for the evaluation of model performance and prediction ability. Reference methods with known performance can be employed for comparison studies. Other validation methods include test set and cross validation, where some samples are set aside for testing purposes. The choice of the testing method mainly depends on the size of the original dataset. Test set validation is suitable for large datasets (>50), whereas cross validation is the best method for medium to small datasets (<50). In this study the K-nearest neighbour algorithm (KNN) was used as a reference method for the classification of contaminated and blank corn samples. A Partial least squares (PLS) regression model was evaluated using full cross validation. Mid-Infrared spectra were collected using the attenuated total reflection (ATR) technique and the fingerprint range (800–1800 cm⁻¹) of 21 maize samples that were contaminated with 300 – 2600 μg/kg deoxynivalenol (DON) was investigated. Separation efficiency after principal component analysis/cluster analysis (PCA/CA) classification was 100%. Cross validation of the PLS model revealed a correlation coefficient of r=0.9926 with a root mean square error of calibration (RMSEC) of 95.01. Validation results gave an r=0.8111 and a root mean square error of cross validation (RMSECV) of 494.5 was calculated. No outliers were reported. Presented at the 25^th Mykotoxin Workshop in Giessen, Germany, May 19–21, 2003     

A Genome-Wide Investigation of MicroRNA Expression Identifies Biologically-Meaningful MicroRNAs That Distinguish between High-Risk and Low-Risk Intraductal Papillary Mucinous Neoplasms of the Pancreas

Background

Intraductal papillary mucinous neoplasms (IPMNs) are pancreatic ductal adenocarcinoma (PDAC) precursors. Differentiating between high-risk IPMNs that warrant surgical resection and low-risk IPMNs that can be monitored is a significant clinical problem, and we sought to discover a panel of mi(cro)RNAs that accurately classify IPMN risk status.

Methodology/Principal Findings

In a discovery phase, genome-wide miRNA expression profiling was performed on 28 surgically-resected, pathologically-confirmed IPMNs (19 high-risk, 9 low-risk) using Taqman MicroRNA Arrays. A validation phase was performed in 21 independent IPMNs (13 high-risk, 8 low-risk). We also explored associations between miRNA expression level and various clinical and pathological factors and examined genes and pathways regulated by the identified miRNAs by integrating data from bioinformatic analyses and microarray analysis of miRNA gene targets. Six miRNAs (miR-100, miR-99b, miR-99a, miR-342-3p, miR-126, miR-130a) were down-regulated in high-risk versus low-risk IPMNs and distinguished between groups (P<10⁻³, area underneath the curve (AUC) = 87%). The same trend was observed in the validation phase (AUC = 74%). Low miR-99b expression was associated with main pancreatic duct involvement (P = 0.021), and serum albumin levels were positively correlated with miR-99a (r = 0.52, P = 0.004) and miR-100 expression (r = 0.49, P = 0.008). Literature, validated miRNA:target gene interactions, and pathway enrichment analysis supported the candidate miRNAs as tumor suppressors and regulators of PDAC development. Microarray analysis revealed that oncogenic targets of miR-130a (ATG2B, MEOX2), miR-342-3p (DNMT1), and miR-126 (IRS-1) were up-regulated in high- versus low-risk IPMNs (P<0.10).

Conclusions

This pilot study highlights miRNAs that may aid in preoperative risk stratification of IPMNs and provides novel insights into miRNA-mediated progression to pancreatic malignancy. The miRNAs identified here and in other recent investigations warrant evaluation in biofluids in a well-powered prospective cohort of individuals newly-diagnosed with IPMNs and other pancreatic cysts and those at increased genetic risk for these lesions.     

Serum DSG2 as a potential biomarker for diagnosis of esophageal squamous cell carcinoma and esophagogastric junction adenocarcinoma

Background: Exploration of serum biomarkers for early detection of upper gastrointestinal cancer is required. Here, we aimed to evaluate the diagnostic potential of serum desmoglein-2 (DSG2) in patients with esophageal squamous cell carcinoma (ESCC) and esophagogastric junction adenocarcinoma (EJA).Methods: Serum DSG2 levels were measured by enzyme-linked immunosorbent assay (ELISA) in 459 participants including 151 patients with ESCC, 96 with EJA, and 212 healthy controls. Receiver operating characteristic (ROC) curves were used to evaluate diagnostic accuracy.Results: Levels of serum DSG2 were significantly higher in patients with ESCC and EJA than those in healthy controls (P<0.001). Detection of serum DSG2 demonstrated an area under the ROC curve (AUC) value of 0.724, sensitivity of 38.1%, and specificity of 84.8% for the diagnosis of ESCC in the training cohort, and AUC 0.736, sensitivity 58.2%, and specificity 84.7% in the validation cohort. For diagnosis of EJA, measurement of DSG2 provided a sensitivity of 29.2%, a specificity of 90.2%, and AUC of 0.698. Similar results were observed for the diagnosis of early-stage ESCC (AUC 0.715 and 0.722, sensitivity 36.3 and 50%, and specificity 84.8 and 84.7%, for training and validation cohorts, respectively) and early-stage EJA (AUC 0.704, sensitivity 44.4%, and specificity 86.9%). Analysis of clinical data indicated that DSG2 levels were significantly associated with patient age and histological grade in ESCC (P<0.05).Conclusion: Serum DSG2 may be a diagnostic biomarker for ESCC and EJA.     

Prediction model of random forest for the risk of hyperuricemia in a Chinese basic health checkup test

Objectives: The present study aimed to develop a random forest (RF) based prediction model for hyperuricemia (HUA) and compare its performance with the conventional logistic regression (LR) model. Methods: This cross-sectional study recruited 91,690 participants (14,032 with HUA, 77,658 without HUA). We constructed a RF-based prediction model in the training sets and evaluated it in the validation sets. Performance of the RF model was compared with the LR model by receiver operating characteristic (ROC) curve analysis. Results: The sensitivity and specificity of the RF models were 0.702 and 0.650 in males, 0.767 and 0.721 in females. The positive predictive value (PPV) and negative predictive value (NPV) were 0.372 and 0.881 in males, 0.159 and 0.978 in females. AUC of the RF models was 0.739 (0.728–0.750) in males and 0.818 (0.799–0.837) in females. AUC of the LR models were 0.730 (0.718–0.741) for males and 0.815 (0.795–0.835) for females. The predictive power of RF was slightly higher than that of LR, but was not statistically significant in females (Delong tests, P=0.0015 for males, P=0.5415 for females). Conclusion: Compared with LR, the good performance in HUA status prediction and the tolerance of features associations or interactions showed great potential of RF in further application. A prospective cohort is necessary for HUA developing prediction. People with high risk factors should be encouraged to actively control to reduce the probability of developing HUA.     

Identification of non-coding RNAs with a new composite feature in the Hybrid Random Forest Ensemble algorithm

To identify non-coding RNA (ncRNA) signals within genomic regions, a classification tool was developed based on a hybrid random forest (RF) with a logistic regression model to efficiently discriminate short ncRNA sequences as well as long complex ncRNA sequences. This RF-based classifier was trained on a well-balanced dataset with a discriminative set of features and achieved an accuracy, sensitivity and specificity of 92.11%, 90.7% and 93.5%, respectively. The selected feature set includes a new proposed feature, SCORE. This feature is generated based on a logistic regression function that combines five significant features—structure, sequence, modularity, structural robustness and coding potential—to enable improved characterization of long ncRNA (lncRNA) elements. The use of SCORE improved the performance of the RF-based classifier in the identification of Rfam lncRNA families. A genome-wide ncRNA classification framework was applied to a wide variety of organisms, with an emphasis on those of economic, social, public health, environmental and agricultural significance, such as various bacteria genomes, the Arthrospira (Spirulina) genome, and rice and human genomic regions. Our framework was able to identify known ncRNAs with sensitivities of greater than 90% and 77.7% for prokaryotic and eukaryotic sequences, respectively. Our classifier is available at http://ncrna-pred.com/HLRF.htm.