Search

Due to the high cost of DNA-binding proteins (DBPs) detection, many machine learning algorithms (ML) have been utilized to large-scale process and detect DBPs. The previous methods took no count of the processing of noise samples. In this study, a fuzzy twin support vector machine (FTWSVM) is employed to detect DBPs. First, multiple types of protein sequence features are formed into kernel matrices; Then, multiple kernel learning (MKL) algorithm is utilized to linear combine multiple kernels; next, self-representation-based membership function is utilized to estimate membership value (weight) of each training sample; finally, we feed the integrated kernel matrix and membership values into the FTWSVM-SR model for training and testing. On comparison with other predictive models, FTWSVM based on SR (FTWSVM-SR) obtains the best performance of Matthew's correlation coefficient (MCC): 0.7410 and 0.5909 on two independent testing sets (PDB186 and PDB2272 datasets), respectively. The results confirm that our method can be an effective DBPs detection tool. Before the biochemical experiment, our model can screen and analyze DBPs on a large scale.

We propose the hierarchical Projective Adaptive Resonance Theory (PART) algorithm for classification of gene expression data. This algorithm is realized by combing transposed quasi-supervised PART and unsupervised PART. We develop the corresponding validation statistics for each process and compare it with other clustering algorithms in a case study of tuberculosis (TB). First, we use sample-based transposed quasi-supervised PART to obtain optimal clustering results of samples distinguished by time post-infection and the representative genes for each cluster including up-regulated, down-regulated and stable genes. The up- and down-regulated genes show more than 90% similarity to the result derived from Linear Models for Microarray Data and are verified by weighted k-nearest neighbor model on TB projection. Second, we use gene-based unsupervised PART algorithm to cluster these representative genes where functional enrichment analysis is conducted in each cluster. We further confirm the main immune response of human macrophage-like THP-1 cells against TB within 2 days is type I interferon-mediated innate immunity. This study demonstrates how hierarchical PART algorithm analyzes microarray data. The sample-based quasi-supervised PART extracts representative genes and narrows down the shortlist of disease-relevant genes and gene-based unsupervised PART classifies representative genes that help to interpret immune response against TB.

Drug–target interaction prediction is important for drug development and drug repurposing. Many computational methods have been proposed for drug–target interaction prediction due to their potential to the time and cost reduction. In this review, we introduce the molecular docking and machine learning-based methods, which have been widely applied to drug–target interaction prediction. Particularly, machine learning-based methods are divided into different types according to the data processing form and task type. For each type of method, we provide a specific description and propose some solutions to improve its capability. The knowledge of heterogeneous network and learning to rank are also summarized in this review. As far as we know, this is the first comprehensive review that summarizes the knowledge of heterogeneous network and learning to rank in the drug–target interaction prediction. Moreover, we propose three aspects that can be explored in depth for future research.

The prediction of drug-target interactions aims to identify potential targets for the treatment of new and rare diseases. The large number of unknown combinations between drugs and targets makes them difficult to verify with experimental methods. There are computational methods that predict drug-target interactions; however, these methods are insufficient in integrating multiple types of data and managing network noise, which affects the accuracy of the prediction. We report a multilayer network representation learning method for drug-target interaction prediction that can integrate useful information from different networks, reduce noise in the multilayer network, and learn the feature vectors of drugs and targets. The feature vectors of the drug and the target are put into the drug-target space to predict the potential drug-target interactions. This work improves the method of multilayer network representation learning and prediction accuracy by increasing the parameter regularization constraints.

BACKGROUND Early-onset obsessive-compulsive disorder (EOCD) is a comparatively severe subtype of obsessive-compulsive disorder (OCD). Olfactory dysfunction is a common symptom of OCD, but all previous studies have focused on late-onset OCD (LOCD). METHODS The current study compared olfactory identification ability in EOCD patients and age-matched and sex-matched LOCD patients and healthy controls. Thirty patients with EOCD, 30 patients with LOCD and 30 healthy controls were included in the study. Olfactory function was measured using the University of Pennsylvania Smell Identification Test. The Logical Memory and Visual Reproduction components of the Revised Wechsler Memory Scale were used to evaluate verbal and visual memory. RESULTS There were significant differences in olfactory identification ability between the three groups. EOCD patients were comparable to LOCD patients, while both patients' group showing worse olfactory identification ability than controls. Olfactory identification ability was not significantly correlated with verbal and visual memory or clinical symptoms in the EOCD group or the LOCD group. CONCLUSIONS The results of the present study suggest that olfactory identification ability may be a relatively stable indicator of OCD, independent of age, duration of illness, verbal and visual memory, and severity of clinical symptoms.

Motivation Identifying whether a pentatricopeptide repeat (PPR) exists in an amino acid is a significant task in the field of bioinformatics. To address this problem, an identification method that combines an optimal feature set selection framework and machine learning algorithms is proposed to recognize the PPR coding genes and proteins in the sequence of amino acid. The original 188-dimensional (D) features are obtained using a feature extraction method, which is successively optimised through a covariance analysis, max-relevant-max-distance processing, and principal component analysis to reduce it to an optimal feature set that has fewer but more expressive features. Four machine learning methods are then used to serve as the classifiers for the identification task. Results The final number of feature data dimensions is reduced from 188 to only 10, and according to the experimental results from support vector machine methods, the loss of the AUC and the F1 values are only 3.26% and 10.1%, respectively. Moreover, after applying the J48, random forest, and naive Bayes methods as classifiers, it was also found that the optimal feature set with 10 dimensions has an almost equivalent performance for a 10-fold validation test.

Machine learning has been widely applied in the fields of biomedicine, computational biology, bioinformatics, image processing, and so on. The performance of machine learning methods mainly relies on feature representation that is the mapping from various types of raw data (i.e., image and genomic data) to a discriminant high-dimensional data space, bridging the raw data with the input of learning/inference algorithms. A good representation is often one that captures the discriminative information from the data and supports effective machine learning. However, over the last few decades, most representation learning approaches are labor-intensive and heavily dependent on the professional knowledge of researchers (dependent on handcrafted feature engineering). To conduct more novel applications in bioinformatics and biomedicine, the weakness of current learning algorithms should be overcome by developing novel feature representation learning algorithms, including supervised representation learning algorithms that are learning features from labeled data, unsupervised feature representation strategies that are learning feature representatives from unlabeled data, and deep feature representation learning algorithms that are learning representative features from data using deep learning architectures.

Graphical abstract First step is the data pre-processing to retrieve the sequence from raw data.The second step is to encode the sequences using on-hot-encoding to make the data readable for the network. The third step is the neural network model construction, and the last step is to classify the sequence as methylated or non-methylated., The most communal post-transcriptional modification, N6-methyladenosine (m6A), is associated with a number of crucial biological processes. The precise detection of m6A sites around the genome is critical for revealing its regulatory function and providing new insights into drug design. Although both experimental and computational models for detecting m6A sites have been introduced, but these conventional methods are laborious and expensive. Furthermore, only a handful of these models are capable of detecting m6A sites in various tissues. Therefore, a more generic and optimized computational method for detecting m6A sites in different tissues is required. In this paper, we proposed a universal model using a deep neural network (DNN) and named it TS-m6A-DL, which can classify m6A sites in several tissues of humans (Homo sapiens), mice (Mus musculus), and rats (Rattus norvegicus). To extract RNA sequence features and to convert the input into numerical format for the network, we utilized one-hot-encoding method. The model was tested using fivefold cross-validation and its stability was measured using independent datasets. The proposed model, TS-m6A-DL, achieved accuracies in the range of 75–85% using the fivefold cross-validation method and 72–84% on the independent datasets. Finally, to authenticate the generalization of the model, we performed cross-species testing and proved the generalization ability by achieving state-of-the-art results.

Deep learning has been increasingly used in bioinformatics, especially in sequence-based protein prediction tasks, as large amounts of biological data are available and deep learning techniques have been developed rapidly in recent years. For sequence-based protein prediction tasks, the selection of a suitable model architecture is essential, whereas sequence data representation is a major factor in controlling model performance. Here, we summarized all the main approaches that are used to represent protein sequence data (amino acid sequence encoding or embedding), which include end-to-end embedding methods, non-contextual embedding methods and embedding methods that use transfer learning and others that are applied for some specific tasks (such as protein sequence embedding based on extracted features for protein structure predictions and graph convolutional network-based embedding for drug discovery tasks). We have also reviewed the architectures of various types of embedding models theoretically and the development of these types of sequence embedding approaches to facilitate researchers and users in selecting the model that best suits their requirements.

With the continuous development of artificial intelligence (AI) technology, big data-supported AI technology with considerable computer and learning capacity has been applied in diagnosing different types of diseases. This study reviews the application of expert systems, neural networks, and deep learning used by AI technology in disease diagnosis. This paper also gives a glimpse of the intelligent diagnosis and treatment of digestive system diseases, respiratory system diseases, and osteoporosis by AI technology.

The advent of transcatheter aortic valve replacement (TAVR) has changed which patients undergo surgical aortic valve replacement (SAVR). We sought to understand the impact of TAVR on the characteristics of SAVR patients in the United States.A cohort of 2959 patients who underwent isolated SAVR at 11 US hospitals that perform both TAVR and SAVR from 2013 through 2017 were grouped by the Society of Thoracic Surgeons (STS) Adult Cardiac Surgery Database version (v)2.73 (2011-2014), v2.81 (2014-2017), and v2.9 (2017) to assess temporal trends in patient characteristics.Over time, SAVR patients were younger with fewer preoperative comorbidities. There was a significant decrease in median STS predicted risk of mortality (PROM) score (2.0 vs. 1.8 vs. 1.3, p 0.001, in v2.73 vs. v2.81 vs. v2.9). Specifically, there were fewer high-risk (STS PROM 8%: 4.3% vs. 4.7% vs. 1.2%, p = 0.03) and intermediate-risk (STS PROM 4% to 8%: 16.3% vs. 11.7% vs. 4.3%, p 0.001) patients. The proportion of patients with bicuspid aortic valve disease increased significantly (11.2% vs. 26.9% vs. 36.6%, p 0.001). There were no differences in operative mortality (1.9% vs. 2.1% vs. 1.4%, p = 0.75).The introduction of TAVR has already impacted the demographics, clinical characteristics and risk profiles of patients undergoing SAVR in the US. Now that TAVR is approved for low-risk patients, SAVR is likely to be reserved for younger patients who are willing to receive a mechanical valve and for patients with aortopathy, coronary artery disease, or concomitant mitral or tricuspid pathology.

Sequence classification is an important data mining task in many real world applications. Over the past few decades, many sequence classification methods have been proposed from different aspects. In particular, the pattern-based method is one of the most important and widely studied sequence classification methods in the literature. In this paper, we present a reference-based sequence classification framework, which can unify existing pattern-based sequence classification methods under the same umbrella. More importantly, this framework can be used as a general platform for developing new sequence classification algorithms. By utilizing this framework as a tool, we propose new sequence classification algorithms that are quite different from existing solutions. Experimental results show that new methods developed under the proposed framework are capable of achieving comparable classification accuracy to those state-of-the-art sequence classification algorithms.

N6-methyladenosine (m6A) is a prevalent methylation modification and plays a vital role in various biological processes, such as metabolism, mRNA processing, synthesis, and transport. Recent studies have suggested that m6A modification is related to common diseases such as cancer, tumours, and obesity. Therefore, accurate prediction of methylation sites in RNA sequences has emerged as a critical issue in the area of bioinformatics. However, traditional high-throughput sequencing and wet bench experimental techniques have the disadvantages of high costs, significant time requirements and inaccurate identification of sites. But through the use of traditional experimental methods, researchers have produced many large databases of m6A sites. With the support of these basic databases and existing deep learning methods, we developed an m6A site predictor named DeepM6ASeq-EL, which integrates an ensemble of five LSTM and CNN classifiers with the combined strategy of hard voting. Compared to the state-of-the-art prediction method WHISTLE (average AUC 0.948 and 0.880), the DeepM6ASeq-EL had a lower accuracy in m6A site prediction (average AUC: 0.861 for the full transcript models and 0.809 for the mature messenger RNA models) when tested on six independent datasets.

The emergence of single cell RNA sequencing has facilitated the studied of genomes, transcriptomes and proteomes. As available single-cell RNA-seq datasets are released continuously, one of the major challenges facing traditional RNA analysis tools is the high-dimensional, high-sparsity, high-noise and large-scale characteristics of single-cell RNA-seq data. Deep learning technologies match the characteristics of single-cell RNA-seq data perfectly and offer unprecedented promise. Here, we give a systematic review for most popular single-cell RNA-seq analysis methods and tools based on deep learning models, involving the procedures of data preprocessing (quality control, normalization, data correction, dimensionality reduction and data visualization) and clustering task for downstream analysis. We further evaluate the deep model-based analysis methods of data correction and clustering quantitatively on 11 gold standard datasets. Moreover, we discuss the data preferences of these methods and their limitations, and give some suggestions and guidance for users to select appropriate methods and tools.

Deep generative models have been an upsurge in the deep learning community since they were proposed. These models are designed for generating new synthetic data including images, videos and texts by fitting the data approximate distributions. In the last few years, deep generative models have shown superior performance in drug discovery especially de novo molecular design. In this study, deep generative models are reviewed to witness the recent advances of de novo molecular design for drug discovery. In addition, we divide those models into two categories based on molecular representations in silico. Then these two classical types of models are reported in detail and discussed about both pros and cons. We also indicate the current challenges in deep generative models for de novo molecular design. De novo molecular design automatically is promising but a long road to be explored.

Background Protein-protein interaction plays a key role in a multitude of biological processes, such as signal transduction, de novo drug design, immune responses, and enzymatic activities. Gaining insights of various binding abilities can deepen our understanding of the interaction. It is of great interest to understand how proteins in a complex interact with each other. Many efficient methods have been developed for identifying protein-protein interface. Results In this paper, we obtain the local information on protein-protein interface, through multi-scale local average block and hexagon structure construction. Given a pair of proteins, we use a trained support vector regression (SVR) model to select best configurations. On Benchmark v4.0, our method achieves average Irmsd value of 3.28Å and overall Fnat value of 63%, which improves upon Irmsd of 3.89Å and Fnat of 49% for ZRANK, and Irmsd of 3.99Å and Fnat of 46% for ClusPro. On CAPRI targets, our method achieves average Irmsd value of 3.45Å and overall Fnat value of 46%, which improves upon Irmsd of 4.18Å and Fnat of 40% for ZRANK, and Irmsd of 5.12Å and Fnat of 32% for ClusPro. The success rates by our method, FRODOCK 2.0, InterEvDock and SnapDock on Benchmark v4.0 are 41.5%, 29.0%, 29.4% and 37.0%, respectively. Conclusion Experiments show that our method performs better than some state-of-the-art methods, based on the prediction quality improved in terms of CAPRI evaluation criteria. All these results demonstrate that our method is a valuable technological tool for identifying protein-protein interface.

Among the large number of known microRNAs (miRNAs), some miRNAs play negligible roles in cell regulation. Therefore, selecting essential miRNAs is an important initial step for a deeper understanding of miRNAs and their functions. In this study, we generated 60 classification models by combining 12 representative feature extraction methods and 5 commonly used classification algorithms. The optimal model for essential miRNA classification that we obtained is based on the Mismatch feature extraction method combined with the random forest algorithm. The F-Measure, area under the curve, and accuracy values of this model were 93.2%, 96.7%, and 93.0%, respectively. We also found that the distribution of the positive and negative examples of the first few features greatly influenced the classification results. The feature extraction methods performed best when the differences between the positive and negative examples were obvious, and this led to better classification of essential miRNAs. Because each classifier’s predictions for the same sample may be different, we employed a novel voting method to improve the accuracy of the classification of essential miRNAs. The performance results showed that the best classification results were obtained when five classification models were used in the voting. The five classification models were constructed based on the Mismatch, pseudo-distance structure status pair composition, Subsequence, Kmer, and Triplet feature extraction methods. The voting result was 95.3%. Our results suggest that the voting method can be an important tool for selecting essential miRNAs. Keywords: miRNA, biological function, classification, feature extraction, voting

Human protein subcellular localization has an important research value in biological processes, also in elucidating protein functions and identifying drug targets. Over the past decade, a number of protein subcellular localization prediction tools have been designed and made freely available online. The purpose of this paper is to summarize the progress of research on the subcellular localization of human proteins in recent years, including commonly used data sets proposed by the predecessors and the performance of all selected prediction tools against the same benchmark data set. We carry out a systematic evaluation of several publicly available subcellular localization prediction methods on various benchmark data sets. Among them, we find that mLASSO-Hum and pLoc-mHum provide a statistically significant improvement in performance, as measured by the value of accuracy, relative to the other methods. Meanwhile, we build a new data set using the latest version of Uniprot database and construct a new GO-based prediction method HumLoc-LBCI in this paper. Then, we test all selected prediction tools on the new data set. Finally, we discuss the possible development directions of human protein subcellular localization. Availability: The codes and data are available from http://www.lbci.cn/syn/.

The identification of therapeutic biomarkers predictive of drug response is crucial in personalized medicine. A number of computational models to predict response of anti-cancer drugs have been developed as the establishment of several pharmacogenomics screening databases. In our study, we proposed a deep cascaded forest model, Deep-Resp-Forest, to classify the anti-cancer drug response as “sensitive” or “resistant”. We made three contributions in this study. Firstly, diverse molecular data could be effectively integrated to provide more information than single type of data for the classification. Combination of two types of data were tested here. Secondly, two structures based on the multi-grained scanning to transform the raw features into high-dimensional feature vectors and integrate the diverse data were proposed in our study. Thirdly, the original deep and time-consuming architecture of cascade forest was improved by a feature optimization operation, which emphasized the most discriminative features across layers. We evaluated the proposed method on the Cancer Cell Line Encyclopedia (CCLE) and Genomics of Drug Sensitivity in Cancer (GDSC) data sets and then compared with the Support Vector Machine. The proposed Deep-Resp-Forest has demonstrated the promising use of deep learning and deep forest approach on the drug response prediction tasks. The R implementation for running our experiments is available at https://github.com/RanSuLab/Deep-Resp-Forest .

Background Since the number of known lncRNA-disease associations verified by biological experiments is quite limited, it has been a challenging task to uncover human disease-related lncRNAs in recent years. Moreover, considering the fact that biological experiments are very expensive and time-consuming, it is important to develop efficient computational models to discover potential lncRNA-disease associations. Results In this manuscript, a novel Collaborative Filtering model called CFNBC for inferring potential lncRNA-disease associations is proposed based on Naïve Bayesian Classifier. In CFNBC, an original lncRNA-miRNA-disease tripartite network is constructed first by integrating known miRNA-lncRNA associations, miRNA-disease associations and lncRNA-disease associations, and then, an updated lncRNA-miRNA-disease tripartite network is further constructed through applying the item-based collaborative filtering algorithm on the original tripartite network. Finally, based on the updated tripartite network, a novel approach based on the Naïve Bayesian Classifier is proposed to predict potential associations between lncRNAs and diseases. The novelty of CFNBC lies in the construction of the updated lncRNA-miRNA-disease tripartite network and the introduction of the item-based collaborative filtering algorithm and Naïve Bayesian Classifier, which guarantee that CFNBC can be applied to predict potential lncRNA-disease associations efficiently without entirely relying on known miRNA-disease associations. Simulation results show that CFNBC can achieve a reliable AUC of 0.8576 in the Leave-One-Out Cross Validation (LOOCV), which is considerably better than previous state-of-the-art results. Moreover, case studies of glioma, colorectal cancer and gastric cancer demonstrate the excellent prediction performance of CFNBC as well. Conclusions According to simulation results, due to the satisfactory prediction performance, CFNBC may be an excellent addition to biomedical researches in the future.

Objectives This study sought to evaluate clinical outcomes and transcatheter heart valve hemodynamics at 1 year after transcatheter aortic valve replacement (TAVR) in low-risk patients. Background Early results from the LRT (Low Risk TAVR) trial demonstrated that TAVR is safe in patients with symptomatic severe aortic stenosis who are at low risk for surgical valve replacement. Methods The LRT trial was an investigator-initiated, prospective, multicenter study and was the first Food and Drug Administration–approved Investigational Device Exemption trial to evaluate feasibility of TAVR in low-risk patients. The primary endpoint was all-cause mortality at 30 days. Secondary endpoints included clinical outcomes and valve hemodynamics at 1 year. Results The LRT trial enrolled 200 low-risk patients with symptomatic severe AS to undergo TAVR at 11 centers. Mean age was 73.6 years and 61.5% were men. At 30 days, there was zero mortality, zero disabling stroke, and low permanent pacemaker implantation rate (5.0%). At 1-year follow-up, mortality was 3.0%, stroke rate was 2.1%, and permanent pacemaker implantation rate was 7.3%. Two (1.0%) subjects underwent surgical reintervention for endocarditis. Of the 14% of TAVR subjects who had evidence of hypoattenuated leaflet thickening at 30 days, there was no impact on valve hemodynamics at 1 year, but the stroke rate was numerically higher (3.8% vs. 1.9%; p = 0.53). Conclusions TAVR in low-risk patients with symptomatic severe aortic stenosis appears to be safe at 1 year. Hypoattenuated leaflet thickening, observed in a minority of TAVR patients at 30 days, did not have an impact on valve hemodynamics in the longer term.

International audience; Hydrogen has long been regarded as an ideal alternative clean energy vector to overcome the drawbacks of fossil technology. However, the direct utilization of hydrogen is challenging, due to low volumetric energy density of hydrogen gas and potential safety issues. Herein, we report an efficient and reversible liquid to liquid organic hydrogen carrier system based on inexpensive, readily available and renewable ethylene glycol. This hydrogen storage system enables the efficient and reversible loading and discharge of hydrogen using a ruthenium pincer complex, with a theoretical hydrogen storage capacity of 6.5 wt%.

N6-methyladenosine (m6A), as one of the most well-studied RNA modifications, has been found to be involved with a wide range of biological processes. Recently, diverse computational methods have been developed for automated identification of m6A sites within RNAs. To identify m6A sites accurately, one of the major challenges is to extract informative features to describe characteristics of m6A sites. However, existing feature representation methods are usually hand-crafted based, and cannot capture discriminative information of m6A sites. In this paper, we develop a m6A site predictor, named DeepM6APred. In this predictor, we propose to use a deep learning based feature descriptor with deep belief network (DBN) to extract high-level latent features. By integrating the deep features with traditional handcrafted features, we train a classification model based on support vector machine and successfully improve the predictive ability of m6A sites. Experimental results on a benchmark dataset show that our proposed method outperforms the state-of-the-art predictors, at least 2% higher in terms of Matthew's correlation coefficient (MCC). Moreover, a webserver that implements the DeepM6APred is established, which is currently available at the website: http://server.malab.cn/DeepM6APred . It is expected to be a useful tool to assist biologists to reveal the functional mechanisms of m6A sites.

Convolutional neural networks (CNNs)-based classifiers improve the accuracy of diagnosis and prediction for Alzheimer's disease (AD). However, exploiting specific brain regions with the AD is essential to understand pathological alteration in the AD and monitor its progression. This paper aims to construct novel AD classification models which have a good performance and interpretation on AD diagnosis. We propose the three classifiers including a simple broaden plain CNNs (SBPCNNs), a major slice-assemble CNNs (SACNNs) and a multi-slice CNNs (MSCNNs), which record the slice positions but have fewer parameters. Specifically, we integrate the ranking and the random forest methods to find the discriminative region that is consistent with domain knowledge about the AD. The results of the visualization explanation of pixel and slice level deliver a clearer understanding of the AD to specialists. The experimental results indicate that the proposed models are meaningful for AD classification.

A definitive pre-mortem diagnosis of prion disease depends on brain biopsy for prion detection currently and no validated alternative preclinical diagnostic tests have been reported to date. To determine the feasibility of using skin for preclinical diagnosis, here we report ultrasensitive serial protein misfolding cyclic amplification (sPMCA) and real-time quaking-induced conversion (RT-QuIC) assays of skin samples from hamsters and humanized transgenic mice (Tg40h) at different time points after intracerebral inoculation with 263K and sCJDMM1 prions, respectively. sPMCA detects skin PrPSc as early as 2 weeks post inoculation (wpi) in hamsters and 4 wpi in Tg40h mice; RT-QuIC assay reveals earliest skin prion-seeding activity at 3 wpi in hamsters and 20 wpi in Tg40h mice. Unlike 263K-inoculated animals, mock-inoculated animals show detectable skin/brain PrPSc only after long cohabitation periods with scrapie-infected animals. Our study provides the proof-of-concept evidence that skin prions could be a biomarker for preclinical diagnosis of prion disease. CJD Foundation National Institutes of Health (NIH) [NS062787, NS087588, NS109532, NS088604] Intramural Research Program of the NIAID, NIH Centers for Disease Control and Prevention [UR8/CCU515004] National Natural Science Foundation of China [81671186] We thank Miriam Warren and Diane Kofskey for skillful histologic and immunohistochemical preparations. We thank Andrew Hughson (RML) for providing recombinant hamster prion protein. Supported by the CJD Foundation and the National Institutes of Health (NIH) NS062787 and NS087588 to W.-Q.Z., NS062787 and NS109532 to W.-Q.Z., and Q.K., NS088604 to Q.K., the Intramural Research Program of the NIAID, NIH to B.C., the Centers for Disease Control and Prevention Contract UR8/CCU515004 to B.S.A. as well as the National Natural Science Foundation of China (No. 81671186) to L.C.

Background: The diagnosis of olfactory dysfunction is challenging given the negligence during routine physical examination, inconvenience of diagnosis in clinical practice, and the inattention to cross-cultural adaptability. The study aimed to develop and validate a simple and effective self-reported olfactory dysfunction questionnaire (SODQ) for the initial screening of clinical olfactory disorders in China. Methods: A total of 121 subjects participated in the study; of these, 96 subjects completed the T&T olfactometer test and 12-item questionnaire, and 25 participants were retested using the SODQ after one week. The T&T olfactometer test examined the olfactory function and the questionnaire measured the ability to perceive common odors in daily life. We evaluated the factor structure, reliability, validity, and discriminative ability of the SODQ. Results: The final version of the SODQ consisted of 10 items with one factor. Test–retest and internal consistency were excellent. Convergent validity of the questionnaire with the T&T olfactory test was high. Furthermore, the discrimination ability was high for the questionnaire with an area under the curve of 0.95 and a cut-off point of 22. Conclusions: The SODQ is a brief, valid, and repeatable tool that has the potential to effectively screen for clinical olfactory disorders from a subjective perspective.

Introduction: A diabetic protein marker is a type of protein that is closely related to diabetes. This kind of protein plays an important role in the prevention and diagnosis of diabetes. Therefore, it is necessary to identify an effective method for predicting diabetic protein markers. In this study, we propose using ensemble methods to predict diabetic protein markers. Methodological issues: The ensemble method consists of two aspects. First, we combine a feature extraction method to obtain mixed features. Next, we classify the protein using ensemble classifiers. We use three feature extraction methods in the ensemble method, including composition and physicochemical features (abbreviated as 188D), adaptive skip gram features (abbreviated as 400D) and g-gap (abbreviated as 670D). There are six traditional classifiers in this study: decision tree, Naive Bayes, logistic regression, part, k-nearest neighbor, and kernel logistic regression. The ensemble classifiers are random forest and vote. First, we used feature extraction methods and traditional classifiers to classify protein sequences. Then, we compared the combined feature extraction methods with single methods. Next, we compared ensemble classifiers to traditional classifiers. Finally, we used ensemble classifiers and combined feature extraction methods to predict samples. Results: The results indicated that ensemble methods outperform single methods with respect to either ensemble classifiers or combined feature extraction methods. When the classifier is a random forest and the feature extraction method is 588D (combined 188D and 400D), the performance is best among all methods. The second best ensemble feature extraction method is 1285D (combining the three methods) with random forest. The best single feature extraction method is 188D, and the worst one is g-gap. Conclusion: According to the results, the ensemble method, either the combined feature extraction method or the ensemble classifier, was better than the single method. We anticipate that ensemble methods will be a useful tool for identifying diabetic protein markers in a cost-effective manner.

Transcription factors (TFs) are essential proteins in regulating the spatiotemporal expression of genes. It is crucial to infer the potential transcription factor binding sites (TFBSs) with high resolution to promote biology and realize precision medicine. Recently, deep learning-based models have shown exemplary performance in the prediction of TFBSs at the base-pair level. However, the previous models fail to integrate nucleotide position information and semantic information without noisy responses. Thus, there is still room for improvement. Moreover, both the inner mechanism and prediction results of these models are challenging to interpret. To this end, the Deep Attentive Encoder-Decoder Neural Network (D-AEDNet) is developed to identify the location of TFs–DNA binding sites in DNA sequences. In particular, our model adopts Skip Architecture to leverage the nucleotide position information in the encoder and removes noisy responses in the information fusion process by Attention Gate. Simultaneously, the Transcription Factor Motif Discovery based on Sliding Window (TF-MoDSW), an approach to discover TFs–DNA binding motifs by utilizing the output of neural networks, is proposed to understand the biological meaning of the predicted result. On ChIP-exo datasets, experimental results show that D-AEDNet has better performance than competing methods. Besides, we authenticate that Attention Gate can improve the interpretability of our model by ways of visualization analysis. Furthermore, we confirm that ability of D-AEDNet to learn TFs–DNA binding motifs outperform the state-of-the-art methods and availability of TF-MoDSW to discover biological sequence motifs in TFs–DNA interaction by conducting experiment on ChIP-seq datasets.

Definitive diagnosis of Parkinson’s disease (PD) and dementia with Lewy bodies (DLB) relies on postmortem finding of disease-associated alpha-synuclein (αSynD) as misfolded protein aggregates in the central nervous system (CNS). The recent development of the real-time quaking induced conversion (RT-QuIC) assay for ultrasensitive detection of αSynD aggregates has revitalized the diagnostic values of clinically accessible biospecimens, including cerebrospinal fluid (CSF) and peripheral tissues. However, the current αSyn RT-QuIC assay platforms vary widely and are thus challenging to implement and standardize the measurements of αSynD across a wide range of biospecimens and in different laboratories. We have streamlined αSyn RT-QuIC assay based on a second generation assay platform that was assembled entirely with commercial reagents. The streamlined RT-QuIC method consisted of a simplified protocol requiring minimal hands-on time, and allowing for a uniform analysis of αSynD in different types of biospecimens from PD and DLB. Ultrasensitive and specific RT-QuIC detection of αSynD aggregates was achieved in million-fold diluted brain homogenates and in nanoliters of CSF from PD and DLB cases but not from controls. Comparative analysis revealed higher seeding activity of αSynD in DLB than PD in both brain homogenates and CSF. Our assay was further validated with CSF samples of 214 neuropathologically confirmed cases from tissue repositories (88 PD, 58 DLB, and 68 controls), yielding a sensitivity of 98% and a specificity of 100%. Finally, a single RT-QuIC assay protocol was employed uniformly to detect seeding activity of αSynD in PD samples across different types of tissues including the brain, skin, salivary gland, and colon. We anticipate that our streamlined protocol will enable interested laboratories to easily and rapidly implement the αSyn RT-QuIC assay for various clinical specimens from PD and DLB. The utilization of commercial products for all assay components will improve the robustness and standardization of the RT-QuIC assay for diagnostic applications across different sites. Due to ultralow sample consumption, the ultrasensitive RT-QuIC assay will facilitate efficient use and sharing of scarce resources of biospecimens. Our streamlined RT-QuIC assay is suitable to track the distribution of αSynD in CNS and peripheral tissues of affected patients. The ongoing evaluation of RT-QuIC assay of αSynD as a potential biomarker for PD and DLB in clinically accessible biospecimens has broad implications for understanding disease pathogenesis, improving early and differential diagnosis, and monitoring therapeutic efficacies in clinical trials.

RNA-binding protein (RBP) is a powerful and wide-ranging regulator that plays an important role in cell development, differentiation, metabolism, health and disease. The prediction of RBPs provides valuable guidance for biologists. Although experimental methods have made great progress in predicting RBP, they are time-consuming and not flexible. Therefore, we developed a network model, rBPDL, by combining a convolutional neural network and long short-term memory for multilabel classification of RBPs. Moreover, to achieve better prediction results, we used a voting algorithm for ensemble learning of the model. We compared rBPDL with state-of-the-art methods and found that rBPDL significantly improved identification performance for the RBP68 dataset, with a macro-Area Under Curve (AUC), micro-AUC, and weighted AUC of 0.936, 0.962, and 0.946, respectively. Furthermore, through AUC statistical analysis of the RBP domain, we analyzed the performance of rBPDL and found that the RBP identification performance in the same domain was similar. In addition, we analyzed the performance preferences and physicochemical properties of the binding protein amino acids and explored the characteristics that affect the binding by using the RBP86 dataset.

Objective: Olfactory deficits have been reported in bipolar disorder, but this finding is controversial. This study investigated whether olfactory deficit can serve as a specific marker for bipolar disorder by comparing olfactory function in different mood episodes of bipolar disorder. We also compared olfactory function in bipolar disorder and other mental disorders – namely, major depressive disorder and schizophrenia. Methods: The study consisted of two experiments. Experiment 1 enrolled 175 bipolar disorder patients (70 depressed subgroup, 70 manic subgroup and 35 euthymic subgroup) and 47 controls. Experiment 2 enrolled the participants from Experiment 1, along with 85 major depressive disorder and 90 schizophrenia patients. The Sniffin’ Sticks test was used to evaluate odour identification ability and odour threshold (as a measure of odour sensitivity). The Hamilton Depression Rating Scale and Young Mania Rating Scale were used to assess depressive symptoms in all subjects and manic symptoms in bipolar disorder patients, respectively. We also used the Positive and Negative Syndrome Scale to assess clinical symptoms in schizophrenia patients. Results: All three bipolar disorder patient subgroups (depressed, manic and euthymic subgroup) showed reduced odour identification ability compared to controls; however, only patients in the acute phase of a mood episode (depressed, and manic subgroup) showed impaired odour sensitivity. Clinical symptoms were negatively correlated with odour sensitivity but not odour identification ability. Bipolar disorder and major depressive disorder patients showed less odour identification and sensitivity impairment than schizophrenia patients. Conclusion: Odour sensitivity is a potential dopaminergic marker for distinguishing between bipolar disorder patients in acute phase vs remission, while odour identification is a trait but a nonspecific marker of bipolar disorder.

How best to utilize the microbial taxonomic abundances in regard to the prediction and explanation of human diseases remains appealing and challenging, and the relative nature of microbiome data necessitates a proper feature selection method to resolve the compositional problem. In this study, we developed an all-in-one platform to address a series of issues in microbiome-based human disease prediction and taxonomic biomarkers discovery. We prioritize the interpretation, runtime and classification accuracy of the distal discriminative balances analysis (DBA-distal) method in selecting a set of distal discriminative balances, and develop DisBalance, a comprehensive platform, to integrate and streamline the workflows of disease model building, disease risk prediction and disease-related biomarker discovery for microbiome-based binary classifications. DisBalance allows the de novo model-building and disease risk prediction in a very fast and convenient way. To facilitate the model-driven and knowledge-driven discoveries, DisBalance dedicates multiple strategies for the mining of microbial biomarkers. The independent validation of the models constructed by the DisBalance pipeline is performed on seven microbiome datasets from the original article of DBA-distal. The implementation of the DisBalance platform is demonstrated by a complete analysis of a shotgun metagenomic dataset of Ulcerative Colitis (UC). As a free and open-source, DisBlance can be accessed at http://lab.malab.cn/soft/DisBalance. The source code and demo data for Disbalance are available at https://github.com/yangfenglong/DisBalance.

Motivation mRNA location corresponds to the location of protein translation and contributes to precise spatial and temporal management of the protein function. However, current assignment of subcellular localization of eukaryotic mRNA reveals important limitations: (1) turning multiple classifications into multiple dichotomies makes the training process tedious; (2) the majority of the models trained by classical algorithm are based on the extraction of single sequence information; (3) the existing state-of-the-art models have not reached an ideal level in terms of prediction and generalization ability. To achieve better assignment of subcellular localization of eukaryotic mRNA, a better and more comprehensive model must be developed. Results In this paper, SubLocEP is proposed as a two-layer integrated prediction model for accurate prediction of the location of sequence samples. Unlike the existing models based on limited features, SubLocEP comprehensively considers additional feature attributes and is combined with LightGBM to generated single feature classifiers. The initial integration model (single-layer model) is generated according to the categories of a feature. Subsequently, two single-layer integration models are weighted (sequence-based: physicochemical properties = 3:2) to produce the final two-layer model. The performance of SubLocEP on independent datasets is sufficient to indicate that SubLocEP is an accurate and stable prediction model with strong generalization ability. Additionally, an online tool has been developed that contains experimental data and can maximize the user convenience for estimation of subcellular localization of eukaryotic mRNA.

Importance Deposition of the pathological α-synuclein (αSynP) in the brain is the hallmark of synucleinopathies, including Parkinson disease (PD), Lewy body dementia (LBD), and multiple system atrophy (MSA). Whether real-time quaking-induced conversion (RT-QuIC) and protein misfolding cyclic amplification (PMCA) assays can sensitively detect skin biomarkers for PD and non-PD synucleinopathies remains unknown. Objective To develop sensitive and specific skin biomarkers for antemortem diagnosis of PD and other synucleinopathies. Design, Setting, and Participants This retrospective and prospective diagnostic study evaluated autopsy and biopsy skin samples from neuropathologically and clinically diagnosed patients with PD and controls without PD. Autopsy skin samples were obtained at 3 medical centers from August 2016 to September 2019, and biopsy samples were collected from 3 institutions from August 2018 to November 2019. Based on neuropathological and clinical diagnoses, 57 cadavers with synucleinopathies and 73 cadavers with nonsynucleinopathies as well as 20 living patients with PD and 21 living controls without PD were included. Specifically, cadavers and participants had PD, LBD, MSA, Alzheimer disease, progressive supranuclear palsy, or corticobasal degeneration or were nonneurodegenerative controls (NNCs). A total of 8 approached biopsy participants either refused to participate in or were excluded from this study due to uncertain clinical diagnosis. Data were analyzed from September 2019 to April 2020. Main Outcomes and Measures Skin αSynPseeding activity was analyzed by RT-QuIC and PMCA assays. Results A total of 160 autopsied skin specimens from 140 cadavers (85 male cadavers [60.7%]; mean [SD] age at death, 76.8 [10.1] years) and 41 antemortem skin biopsies (27 male participants [66%]; mean [SD] age at time of biopsy, 65.3 [9.2] years) were analyzed. RT-QuIC analysis of αSynPseeding activity in autopsy abdominal skin samples from 47 PD cadavers and 43 NNCs revealed 94% sensitivity (95% CI, 85-99) and 98% specificity (95% CI, 89-100). As groups, RT-QuIC also yielded 93% sensitivity (95% CI, 85-97) and 93% specificity (95% CI, 83-97) among 57 cadavers with synucleinopathies (PD, LBD, and MSA) and 73 cadavers without synucleinopathies (Alzheimer disease, progressive supranuclear palsy, corticobasal degeneration, and NNCs). PMCA showed 82% sensitivity (95% CI, 76-88) and 96% specificity (95% CI, 85-100) with autopsy abdominal skin samples from PD cadavers. From posterior cervical and leg skin biopsy tissues from patients with PD and controls without PD, the sensitivity and specificity were 95% (95% CI, 77-100) and 100% (95% CI, 84-100), respectively, for RT-QuIC and 80% (95% CI, 49-96) and 90% (95% CI, 60-100) for PMCA. Conclusions and Relevance This study provides proof-of-concept that skin αSynPseeding activity may serve as a novel biomarker for antemortem diagnoses of PD and other synucleinopathies.

Sharaf-AG Low Uplift is located in the central area of Nugara Depression in Muglad Basin, Sudan. It has similar faulted system as the other oil and gas rich sub-basin but it doesn’t have much discovery in oil and gas. So the study of tectonic characteristics has been carried. After detailed seismic interpretation and analysis, the traps controlled by the northwest normal fault have a ‘beaded’ distribution characteristic from southwest to northeast under an extensional- torsional stress. And maybe it destroyed the storage of oil and gas in this area.

Gene regulatory network (GRN) inference based on gene expression data is still a huge challenge in systems biology. Genomic data, including time-series expression data, steady-state data, knockout data, and other biological data, such as Gene Ontology (GO) annotations, provide information on potential gene regulation. However, most existing methods continue to use only a single dataset for GRN inference. To integrate these types of data and improve the accuracy of inference, we propose a new data-integration strategy based on guided regular random forest (GRRF) for GRN inference, dubbed GRRFNet. Specifically, first, time-series data and steady-state data as main datasets are integrated to generate learning samples; simultaneously, other datasets are processed to design penalty coefficients for guiding feature selection; then, a GRRF model is applied to integrate the prior information with a main dataset to learn the transcription function and evaluate the importance of feature; finally, the score of the feature's importance is used as the possibility of the gene regulatory relationships to construct the GRN. To evaluate the performance of GRRFNet, we compare it with GENIE3, dynGENIE3, and GRIEF at the artificial DREAM4 dataset and the real Escherichia coli dataset. Although GRRFNet does not yield the best performance on every network, its competitiveness is still reflected herein.

With the increasing volume of high-throughput sequencing data from a variety of omics techniques in the field of plant–pathogen interactions, sorting, retrieving, processing and visualizing biological information have become a great challenge. Within the explosion of data, machine learning offers powerful tools to process these complex omics data by various algorithms, such as Bayesian reasoning, support vector machine and random forest. Here, we introduce the basic frameworks of machine learning in dissecting plant–pathogen interactions and discuss the applications and advances of machine learning in plant–pathogen interactions from molecular to network biology, including the prediction of pathogen effectors, plant disease resistance protein monitoring and the discovery of protein–protein networks. The aim of this review is to provide a summary of advances in plant defense and pathogen infection and to indicate the important developments of machine learning in phytopathology.

The peptide therapeutics market is providing new opportunities for the biotechnology and pharmaceutical industries. Therefore, identifying therapeutic peptides and exploring their properties are important. Although several studies have proposed different machine learning methods to predict peptides as being therapeutic peptides, most do not explain the decision factors of model in detail. In this work, an Interpretable Therapeutic Peptide Prediction (ITP-Pred) model based on efficient feature fusion was developed. First, we proposed three kinds of feature descriptors based on sequence and physicochemical property encoded, namely amino acid composition (AAC), group AAC and coding autocorrelation, and concatenated them to obtain the feature representation of therapeutic peptide. Then, we input it into the CNN-Bi-directional Long Short-Term Memory (BiLSTM) model to automatically learn recognition of therapeutic peptides. The cross-validation and independent verification experiments results indicated that ITP-Pred has a higher prediction performance on the benchmark dataset than other comparison methods. Finally, we analyzed the output of the model from two aspects: sequence order and physical and chemical properties, mining important features as guidance for the design of better models that can complement existing methods.

Rice and maize are the principal food crop species worldwide. The mechanism of gene regulation for the yield of rice and maize is still the research focus at present. Seed size, weight and shape are important traits of crop yield in rice and maize. Most members of three gene families, APETALA2/ethylene response factor, auxin response factors and MADS, were identified to be involved in yield traits in rice and maize. Analysis of molecular regulation mechanisms related to yield traits provides theoretical support for the improvement of crop yield. Genetic regulatory network analysis can provide new insights into gene families with the improvement of sequencing technology. Here, we analyzed the evolutionary relationships and the genetic regulatory network for the gene family members to predicted genes that may be involved in yield-related traits in rice and maize. The results may provide some theoretical and application guidelines for future investigations of molecular biology, which may be helpful for developing new rice and maize varieties with high yield traits.

Biocomputing and synthetic biology have been two of the most exciting emerging fields in recent years [...]

Non-classically secreted proteins (NCSPs) are proteins that are located in the extracellular environment, although there is a lack of known signal peptides or secretion motifs. They usually perform different biological functions in intracellular and extracellular environments, and several of their biological functions are linked to bacterial virulence and cell defence. Accurate protein localization is essential for all living organisms, however, the performance of existing methods developed for NCSP identification has been unsatisfactory and in particular suffer from data deficiency and possible overfitting problems. Further improvement is desirable, especially to address the lack of informative features and mining subset-specific features in imbalanced datasets. In the present study, a new computational predictor was developed for NCSP prediction of gram-positive bacteria. First, to address the possible prediction bias caused by the data imbalance problem, ten balanced subdatasets were generated for ensemble model construction. Then, the F-score algorithm combined with sequential forward search was used to strengthen the feature representation ability for each of the training subdatasets. Third, the subset-specific optimal feature combination process was adopted to characterize the original data from different aspects, and all subdataset-based models were integrated into a unified model, NonClasGP-Pred, which achieved an excellent performance with an accuracy of 93.23 %, a sensitivity of 100 %, a specificity of 89.01 %, a Matthew’s correlation coefficient of 87.68 % and an area under the curve value of 0.9975 for ten-fold cross-validation. Based on assessment on the independent test dataset, the proposed model outperformed state-of-the-art available toolkits. For availability and implementation, see: http://lab.malab.cn/~wangchao/softwares/NonClasGP/.

BackgroundThe definite diagnosis of human sporadic Creutzfeldt-Jakob disease (sCJD) largely depends on postmortem neuropathology and PrPSc detection in the brain. The development of prion RT-QuIC of cerebrospinal fluid (CSF) samples makes it possible for premortem diagnosis for sCJD. However the diagnostic potential of RT-QuIC of skin specimen for probable sCJD is not well researched. This study is to evaluate the diagnostic potential of RT-QuIC of skin specimen in human prion diseases.MethodsWe collected the paired skin and CSF samples from 29 recruited alive patients referred to Chinese CJD surveillance center, including 12 probable sCJD, 9 non-CJD, 3 genetic prion disease (gPrD) and 5 cases whose diagnoses still pending. The samples were subjected to RT-QuIC assays using recombinant hamster PrP protein rHaPrP90-231 as the substrate.ResultsAll 12 probable sCJD patients, 4 pending, and 1 T188K genetic CJD (gCJD) cases showed positive prion-seeding activity, while all 9 non-CJD patients were negative. CSF RT-QuIC positive seeding activity was only observed in 5 probable sCJD patients.ConclusionsOur preliminary data indicate high sensitivity and specificity of skin RT-QuIC in prion detection for Chinese probable sCJD and highlight that skin prion-seeding activity is a reliable biomarker for premortem diagnosis of human prion disease.

Olfactory impairment and rapid eye movement sleep behaviour disorder (RBD) are prodromal symptoms of Parkinson’s disease (PD) that may be associated with each other. This review aims to investigate the significance of olfaction in the diagnosis and prognosis of patients with RBD and to assess moderating factors affecting olfactory performance. We searched articles on olfaction in RBD and PD in five electronic databases. We identified 32 studies for the systematic review and used 28 of those, including 2858 participants for meta-analysis. Results revealed significant deficits in odour identification (g=−1.80; 95% CI: −2.17 to −1.43), threshold (g=−1.29; 95% CI: −1.67 to −0.91), discrimination (g=−1.08; 95% CI: −1.28 to −0.87) and overall olfactory function (g=−1.64; 95% CI: −1.94 to −1.35) in patients with RBD. Except for the Unified Parkinson's Disease Rating Scale Part III scores, none of the known moderating variables (including age, sex, disease duration and years of education) accounted for the olfactory function heterogeneity in patients with RBD. We identified similar olfactory impairments in patients with RBD and patients with PD (either with or without underlying RBD). These findings suggest that olfactory impairment may be a sensitive and stable diagnostic biomarker of RBD and appears to be useful for identifying patients with idiopathic RBD at high risk for early conversion to PD.

Objective: The objective of the study was to develop an association between clinicopathologic and sonographic features of patients with papillary thyroid microcarcinoma and the prevalence of lymph node metastasis. Subjects and methods: Clinicopathologic and sonographic features of 415 patients of papillary thyroid microcarcinoma with (n = 102) or without (n = 313) lymph node metastasis were retrospectively reviewed. The thickness of the lymph node ≥ 6 mm with intra-lymph nodal occupying lesions considered lymph node metastasis. Also, it was considered metastasis if lymph node perfusion or blood flow defect was found with any thickness size. Univariate following multivariate analysis was performed for the prediction of sonographic features and clinicopathologic factors for the prevalence of lymph node metastasis. Results: Male gender ( p = 0.041), age < 45 years ( p = 0.042), preoperative calcitonin > 65 pg/ mL ( p = 0.039), nodule size > 5 mm in diameter ( p = 0.038), bilaterality ( p = 0.038), tumor capsular invasion ( p = 0.048), cystic change ( p = 0.047), and hyper vascularity ( p = 0.049) of thyroid nodules were associated with lymph node metastasis. Also, thyroid nodules 5 mm and more in diameter may have high aggressiveness. Conclusion: These data helped the surgeon for individualized treatment in thyroid carcinoma and avoid unnecessary prophylactic surgery of the lymph node.

Anticancer peptides constitute one of the most promising therapeutic agents for combating common human cancers. Using wet experiments to verify whether a peptide displays anticancer characteristics is time-consuming and costly. Hence, in this study, we proposed a computational method named identify anticancer peptides via deep representation learning features (iACP-DRLF) using light gradient boosting machine algorithm and deep representation learning features. Two kinds of sequence embedding technologies were used, namely soft symmetric alignment embedding and unified representation (UniRep) embedding, both of which involved deep neural network models based on long short-term memory networks and their derived networks. The results showed that the use of deep representation learning features greatly improved the capability of the models to discriminate anticancer peptides from other peptides. Also, UMAP (uniform manifold approximation and projection for dimension reduction) and SHAP (shapley additive explanations) analysis proved that UniRep have an advantage over other features for anticancer peptide identification. The python script and pretrained models could be downloaded from https://github.com/zhibinlv/iACP-DRLF or from http://public.aibiochem.net/iACP-DRLF/.

Gene regulatory network (GRN) is the important mechanism of maintaining life process, controlling biochemical reaction and regulating compound level, which plays an important role in various organisms and systems. Reconstructing GRN can help us to understand the molecular mechanism of organisms and to reveal the essential rules of a large number of biological processes and reactions in organisms. Various outstanding network reconstruction algorithms use specific assumptions that affect prediction accuracy, in order to deal with the uncertainty of processing. In order to study why a certain method is more suitable for specific research problem or experimental data, we conduct research from model-based, information-based and machine learning-based method classifications. There are obviously different types of computational tools that can be generated to distinguish GRNs. Furthermore, we discuss several classical, representative and latest methods in each category to analyze core ideas, general steps, characteristics, etc. We compare the performance of state-of-the-art GRN reconstruction technologies on simulated networks and real networks under different scaling conditions. Through standardized performance metrics and common benchmarks, we quantitatively evaluate the stability of various methods and the sensitivity of the same algorithm applying to different scaling networks. The aim of this study is to explore the most appropriate method for a specific GRN, which helps biologists and medical scientists in discovering potential drug targets and identifying cancer biomarkers.

ObjectiveTo determine whether (1) immunofluorescence is a reproducible technique in detecting misfolded α-synuclein in skin nerves and subsequently whether (2) immunofluorescence and real-time quaking-induced conversion (RT-QuIC) (both in skin and CSF) show a comparable in vivo diagnostic accuracy in distinguishing synucleinopathies from non-synucleinopathies in a large cohort of patients.MethodsWe prospectively recruited 90 patients fulfilling clinical and instrumental diagnostic criteria for all synucleinopathies variants and non-synucleinopathies (mainly including Alzheimer disease, tauopathies, and vascular parkinsonism or dementia). Twenty-four patients with mainly peripheral neuropathies were used as controls. Patients underwent skin biopsy for immunofluorescence and RT-QuIC; CSF was examined in patients who underwent lumbar puncture for diagnostic purposes. Immunofluorescence and RT-QuIC analysis were made blinded to the clinical diagnosis.ResultsImmunofluorescence showed reproducible results between 2 pairs of neighboring skin samples. Both immunofluorescence and RT-QuIC showed high sensitivity and specificity in discriminating synucleinopathies from non-synucleinopathies and controls but immunofluorescence presented higher diagnostic accuracy. Immunofluorescence presented a good level of agreement with RT-QuIC in both skin and CSF in synucleinopathies.ConclusionsBoth immunofluorescence and RT-QuIC showed high diagnostic accuracy, although immunofluorescence displayed the better value as well as optimal reproducibility; they presented a good level of agreement in synucleinopathies, supporting the use of less invasive tests such as skin immunofluorescence or RT-QuIC instead of CSF RT-QuIC as a diagnostic tool for synucleinopathies.Classification of EvidenceThis study provides Class III evidence that immunofluorescence or RT-QuIC accurately distinguish synucleinopathies from non-synucleinopathies.

Over the last decade, genome-wide association studies (GWAS) have discovered thousands of genetic variants underlying complex human diseases and agriculturally important traits. These findings have been utilized to dissect the biological basis of diseases, to develop new drugs, to advance precision medicine and to boost breeding. However, the potential of GWAS is still underexploited due to methodological limitations. Many challenges have emerged, including detecting epistasis and single-nucleotide polymorphisms (SNPs) with small effects and distinguishing causal variants from other SNPs associated through linkage disequilibrium. These issues have motivated advancements in GWAS analyses in two contrasting cultures—statistical modelling and machine learning. In this review, we systematically present the basic concepts and the benefits and limitations in both methods. We further discuss recent efforts to mitigate their weaknesses. Additionally, we summarize the state-of-the-art tools for detecting the missed signals, ultrarare mutations and gene–gene interactions and for prioritizing SNPs. Our work can offer both theoretical and practical guidelines for performing GWAS analyses and for developing further new robust methods to fully exploit the potential of GWAS.

Background "Early-onset schizophrenia" (EOS) is defined as disease with onset before the age of 18 years. This subset of schizophrenia exhibits worse cognitive function and carries a worse prognosis than adult-onset schizophrenia (AOS). Olfactory impairment has been found in patients with schizophrenia-spectrum disorders. However, most research has focused on olfactory impairment in patients with AOS: olfactory function in EOS is not known. The aim of this study was to investigate the olfactory identification ability in EOS, and its relationship with negative symptoms. Methods We compared olfactory function between two independent samples: 40 patients with EOS and 40 age- and sex-matched healthy controls (HCs); as well as 40 patients with AOS and 40 age- and sex-matched HCs. The University of Pennsylvania Smell Identification Test was administered. Results The EOS group and AOS group exhibited worse olfactory identification ability than HCs; impairment correlated significantly with negative symptoms. Olfactory identification was worse in patients suffering EOS compared with those suffering AOS. Conclusion Olfactory identification impairment may be a trait marker of schizophrenia.