With the increasing prevalence of autism spectrum disorder (ASD), it is important to identify ASD patients for effective treatment and intervention, especially in early childhood. Neuroimaging ...techniques have been used to characterize the complex biomarkers based on the functional connectivity anomalies in the ASD. However, the diagnosis of ASD still adopts the symptom-based criteria by clinical observation. The existing computational models tend to achieve unreliable diagnostic classification on the large-scale aggregated data sets. In this work, we propose a novel graph-based classification model using the deep belief network (DBN) and the Autism Brain Imaging Data Exchange (ABIDE) database, which is a worldwide multisite functional and structural brain imaging data aggregation. The remarkable connectivity features are selected through a graph extension of <inline-formula> <tex-math notation="LaTeX">{K} </tex-math></inline-formula>-nearest neighbors and then refined by a restricted path-based depth-first search algorithm. Thanks to the feature reduction, lower computational complexity could contribute to the shortening of the training time. The automatic hyperparameter-tuning technique is introduced to optimize the hyperparameters of the DBN by exploring the potential parameter space. The simulation experiments demonstrate the superior performance of our model, which is 6.4% higher than the best result reported on the ABIDE database. We also propose to use the data augmentation and the oversampling technique to identify further the possible subtypes within the ASD. The interpretability of our model enables the identification of the most remarkable autistic neural correlation patterns from the data-driven outcomes.
In the recent few years, an increasing number of studies have shown that microRNAs (miRNAs) play critical roles in many fundamental and important biological processes. As one of pathogenetic factors, ...the molecular mechanisms underlying human complex diseases still have not been completely understood from the perspective of miRNA. Predicting potential miRNA-disease associations makes important contributions to understanding the pathogenesis of diseases, developing new drugs, and formulating individualized diagnosis and treatment for diverse human complex diseases. Instead of only depending on expensive and time-consuming biological experiments, computational prediction models are effective by predicting potential miRNA-disease associations, prioritizing candidate miRNAs for the investigated diseases, and selecting those miRNAs with higher association probabilities for further experimental validation. In this study, Path-Based MiRNA-Disease Association (PBMDA) prediction model was proposed by integrating known human miRNA-disease associations, miRNA functional similarity, disease semantic similarity, and Gaussian interaction profile kernel similarity for miRNAs and diseases. This model constructed a heterogeneous graph consisting of three interlinked sub-graphs and further adopted depth-first search algorithm to infer potential miRNA-disease associations. As a result, PBMDA achieved reliable performance in the frameworks of both local and global LOOCV (AUCs of 0.8341 and 0.9169, respectively) and 5-fold cross validation (average AUC of 0.9172). In the cases studies of three important human diseases, 88% (Esophageal Neoplasms), 88% (Kidney Neoplasms) and 90% (Colon Neoplasms) of top-50 predicted miRNAs have been manually confirmed by previous experimental reports from literatures. Through the comparison performance between PBMDA and other previous models in case studies, the reliable performance also demonstrates that PBMDA could serve as a powerful computational tool to accelerate the identification of disease-miRNA associations.
Objective: Deep learning (DL) techniques have been introduced to assist doctors in the interpretation of medical images by detecting image-derived phenotype abnormality. Yet the privacy-preserving ...policy of medical images disables the effective training of DL model using sufficiently large datasets. As a decentralized computing paradigm to address this issue, federated learning (FL) allows the training process to occur in individual institutions with local datasets, and then aggregates the resultant weights without risk of privacy leakage. Methods: We propose an effective federated multi-task learning (MTL) framework to jointly identify multiple related mental disorders based on functional magnetic resonance imaging data. A federated contrastive learning-based feature extractor is developed to extract high-level features across client models. To ease the optimization conflicts of updating shared parameters in MTL, we present a federated multi-gate mixture of expert classifier for the joint classification. The proposed framework also provides practical modules, including personalized model learning, privacy protection, and federated biomarker interpretation. Results: On real-world datasets, the proposed framework achieves robust diagnosis accuracies of 69.48 <inline-formula><tex-math notation="LaTeX">\pm</tex-math></inline-formula> 1.6%, 71.44 <inline-formula><tex-math notation="LaTeX">\pm</tex-math></inline-formula> 3.2%, and 83.29 <inline-formula><tex-math notation="LaTeX">\pm</tex-math></inline-formula> 3.2% in autism spectrum disorder, attention deficit/hyperactivity disorder, and schizophrenia, respectively. Conclusion: The proposed framework can effectively ease the domain shift between clients via federated MTL. Significance: The current work provides insights into exploiting the advantageous knowledge shared in related mental disorders for improving the generalization capability of computer-aided detection approaches.
Facing the increasing worldwide prevalence of mental disorders, the symptom-based diagnostic criteria struggle to address the urgent public health concern due to the global shortfall in ...well-qualified professionals. Thanks to the recent advances in neuroimaging techniques, functional magnetic resonance imaging (fMRI) has surfaced as a new solution to characterize neuropathological biomarkers for detecting functional connectivity (FC) anomalies in mental disorders. However, the existing computer-aided diagnosis models for fMRI analysis suffer from unstable performance on large datasets. To address this issue, we propose an efficient multitask learning (MTL) framework for joint diagnosis of multiple mental disorders using resting-state fMRI data. A novel multiobjective evolutionary clustering algorithm is presented to group regions of interests (ROIs) into different clusters for FC pattern analysis. On the optimal clustering solution, the multicluster multigate mixture-of-expert model is used for the final classification by capturing the highly consistent feature patterns among related diagnostic tasks. Extensive simulation experiments demonstrate that the performance of the proposed framework is superior to that of the other state-of-the-art methods. Moreover, the potential for practical application of the framework is also validated in terms of limited computational resources, real-time analysis, and insufficient training data. The proposed model can identify the remarkable interpretative biomarkers associated with specific mental disorders for clinical interpretation analysis.
Abstract
From transcriptional noise to dark matter of biology, the rapidly changing view of long non-coding RNA (lncRNA) leads to deep understanding of human complex diseases induced by abnormal ...expression of lncRNAs. There is urgent need to discern potential functional roles of lncRNAs for further study of pathology, diagnosis, therapy, prognosis, prevention of human complex disease and disease biomarker detection at lncRNA level. Computational models are anticipated to be an effective way to combine current related databases for predicting most potential lncRNA functions and calculating lncRNA functional similarity on the large scale. In this review, we firstly illustrated the biological function of lncRNAs from five biological processes and briefly depicted the relationship between mutations or dysfunctions of lncRNAs and human complex diseases involving cancers, nervous system disorders and others. Then, 17 publicly available lncRNA function–related databases containing four types of functional information content were introduced. Based on these databases, dozens of developed computational models are emerging to help characterize the functional roles of lncRNAs. We therefore systematically described and classified both 16 lncRNA function prediction models and 9 lncRNA functional similarity calculation models into 8 types for highlighting their core algorithm and process. Finally, we concluded with discussions about the advantages and limitations of these computational models and future directions of lncRNA function prediction and functional similarity calculation. We believe that constructing systematic functional annotation systems is essential to strengthen the prediction accuracy of computational models, which will accelerate the identification process of novel lncRNA functions in the future.
Accumulating clinical researches have shown that specific microbes with abnormal levels are closely associated with the development of various human diseases. Knowledge of microbe-disease ...associations can provide valuable insights for complex disease mechanism understanding as well as the prevention, diagnosis and treatment of various diseases. However, little effort has been made to predict microbial candidates for human complex diseases on a large scale.
In this work, we developed a new computational model for predicting microbe-disease associations by combining two single recommendation methods. Based on the assumption that functionally similar microbes tend to get involved in the mechanism of similar disease, we adopted neighbor-based collaborative filtering and a graph-based scoring method to compute association possibility of microbe-disease pairs. The promising prediction performance could be attributed to the use of hybrid approach based on two single recommendation methods as well as the introduction of Gaussian kernel-based similarity and symptom-based disease similarity.
To evaluate the performance of the proposed model, we implemented leave-one-out and fivefold cross validations on the HMDAD database, which is recently built as the first database collecting experimentally-confirmed microbe-disease associations. As a result, NGRHMDA achieved reliable results with AUCs of 0.9023 ± 0.0031 and 0.9111 in the validation frameworks of fivefold CV and LOOCV. In addition, 78.2% microbe samples and 66.7% disease samples are found to be consistent with the basic assumption of our work that microbes tend to get involved in the similar disease clusters, and vice versa.
Compared with other methods, the prediction results yielded by NGRHMDA demonstrate its effective prediction performance for microbe-disease associations. It is anticipated that NGRHMDA can be used as a useful tool to search the most potential microbial candidates for various diseases, and therefore boosts the medical knowledge and drug development. The codes and dataset of our work can be downloaded from https://github.com/yahuang1991/NGRHMDA .
An increasing number of evidences indicate microbes are implicated in human physiological mechanisms, including complicated disease pathology. Some microbes have been demonstrated to be associated ...with diverse important human diseases or disorders. Through investigating these disease-related microbes, we can obtain a better understanding of human disease mechanisms for advancing medical scientific progress in terms of disease diagnosis, treatment, prevention, prognosis and drug discovery. Based on the known microbe-disease association network, we developed a semi-supervised computational model of Laplacian Regularized Least Squares for Human Microbe-Disease Association (LRLSHMDA) by introducing Gaussian interaction profile kernel similarity calculation and Laplacian regularized least squares classifier. LRLSHMDA reached the reliable AUCs of 0.8909 and 0.7657 based on the global and local leave-one-out cross validations, respectively. In the framework of 5-fold cross validation, average AUC value of 0.8794 +/-0.0029 further demonstrated its promising prediction ability. In case studies, 9, 9 and 8 of top-10 predicted microbes have been manually certified to be associated with asthma, colorectal carcinoma and chronic obstructive pulmonary disease by published literature evidence. Our proposed model achieves better prediction performance relative to the previous model. We expect that LRLSHMDA could offer insights into identifying more promising human microbe-disease associations in the future.
The offloading of computation-intensive tasks to an edge server near resource-constrained mobile devices can provide improved application performance and user experience. However, with the rapid ...growth of mobile devices connected to the edge server, it is challenging to directly obtain an optimal task offloading scheme due to increasing computational cost and problem scale. In this study, we model the costly task offloading problem (CTOP) in mobile-edge computing networks to achieve efficient joint optimization of energy consumption and processing latency for mobile devices. Inspired by the success of evolutionary multitasking in solving complex optimization problems by leveraging the experience of simple optimization problems, we develop a novel multitasking framework whose effectiveness is demonstrated in solving the CTOP. In this framework, auxiliary tasks are created to optimize the local processing overhead and the edge processing overhead of task offloading. On this basis, we propose an effective multitask evolutionary algorithm that includes segmented knowledge transfer and auxiliary task update. Specifically, source and extended decision variables are considered as different knowledge to be utilized, while the auxiliary tasks are allowed to be updated dynamically. Related knowledge that is learned from cheap and simple auxiliary tasks promotes the evolutionary search for CTOP. Experimental results verify the effectiveness of knowledge transfer. Compared to existing multitasking and single-tasking algorithms, the proposed algorithm shows competitive performance in CTOP instances and achieves better comprehensive performance in terms of energy consumption and processing latency.
The rapid progress of high-throughput DNA sequencing techniques has dramatically reduced the costs of whole genome sequencing, which leads to revolutionary advances in gene industry. The explosively ...increasing volume of raw data outpaces the decreasing disk cost and the storage of huge sequencing data has become a bottleneck of downstream analyses. Data compression is considered as a solution to reduce the dependency on storage. Efficient sequencing data compression methods are highly demanded.
In this article, we present a lossless reference-based compression method namely LW-FQZip 2 targeted at FASTQ files. LW-FQZip 2 is improved from LW-FQZip 1 by introducing more efficient coding scheme and parallelism. Particularly, LW-FQZip 2 is equipped with a light-weight mapping model, bitwise prediction by partial matching model, arithmetic coding, and multi-threading parallelism. LW-FQZip 2 is evaluated on both short-read and long-read data generated from various sequencing platforms. The experimental results show that LW-FQZip 2 is able to obtain promising compression ratios at reasonable time and memory space costs.
The competence enables LW-FQZip 2 to serve as a candidate tool for archival or space-sensitive applications of high-throughput DNA sequencing data. LW-FQZip 2 is freely available at http://csse.szu.edu.cn/staff/zhuzx/LWFQZip2 and https://github.com/Zhuzxlab/LW-FQZip2 .
The high prevalence of mental disorders gradually poses a huge pressure on the public healthcare services. Deep learning-based computer-aided diagnosis (CAD) has emerged to relieve the tension in ...healthcare institutions by detecting abnormal neuroimaging-derived phenotypes. However, training deep learning models relies on sufficient annotated datasets, which can be costly and laborious. Semi-supervised learning (SSL) and transfer learning (TL) can mitigate this challenge by leveraging unlabeled data within the same institution and advantageous information from source domain, respectively. This work is the first attempt to propose an effective semi-supervised transfer learning (SSTL) framework dubbed S3TL for CAD of mental disorders on fMRI data. Within S3TL, a secure cross-domain feature alignment method is developed to generate target-related source model in SSL. Subsequently, we propose an enhanced dual-stage pseudo-labeling approach to assign pseudo-labels for unlabeled samples in target domain. Finally, an advantageous knowledge transfer method is conducted to improve the generalization capability of the target model. Comprehensive experimental results demonstrate that S3TL achieves competitive accuracies of 69.14%, 69.65%, and 72.62% on ABIDE-I, ABIDE-II, and ADHD-200 datasets, respectively. Furthermore, the simulation experiments also demonstrate the application potential of S3TL through model interpretation analysis and federated learning extension.