A deep learning framework for predicting disease-gene associations with functional modules and graph augmentation.

BACKGROUND: The exploration of gene-disease associations is crucial for understanding the mechanisms underlying disease onset and progression, with significant implications for prevention and treatment strategies. Advances in high-throughput biotechnology have generated a wealth of data linking diseases to specific genes. While graph representation learning has recently introduced groundbreaking approaches for predicting novel associations, existing studies always overlooked the cumulative impact of functional modules such as protein complexes and the incompletion of some important data such as protein interactions, which limits the detection performance. RESULTS: Addressing these limitations, here we introduce a deep learning framework called ModulePred for predicting disease-gene associations. ModulePred performs graph augmentation on the protein interaction network using L3 link prediction algorithms. It builds a heterogeneous module network by integrating disease-gene associations, protein complexes and augmented protein interactions, and develops a novel graph embedding for the heterogeneous module network. Subsequently, a graph neural network is constructed to learn node representations by collectively aggregating information from topological structure, and gene prioritization is carried out by the disease and gene embeddings obtained from the graph neural network. Experimental results underscore the superiority of ModulePred, showcasing the effectiveness of incorporating functional modules and graph augmentation in predicting disease-gene associations. This research introduces innovative ideas and directions, enhancing the understanding and prediction of gene-disease relationships.

PM-CNN: microbiome status recognition and disease detection model based on phylogeny and multi-path neural network.

Motivation: The human microbiome, found throughout various body parts, plays a crucial role in health dynamics and disease development. Recent research has highlighted microbiome disparities between patients with different diseases and healthy individuals, suggesting the microbiome's potential in recognizing health states. Traditionally, microbiome-based status classification relies on pre-trained machine learning (ML) models. However, most ML methods overlook microbial relationships, limiting model performance. Results: To address this gap, we propose PM-CNN (Phylogenetic Multi-path Convolutional Neural Network), a novel phylogeny-based neural network model for multi-status classification and disease detection using microbiome data. PM-CNN organizes microbes based on their phylogenetic relationships and extracts features using a multi-path convolutional neural network. An ensemble learning method then fuses these features to make accurate classification decisions. We applied PM-CNN to human microbiome data for status and disease detection, demonstrating its significant superiority over existing ML models. These results provide a robust foundation for microbiome-based state recognition and disease prediction in future research and applications. Availability and implementation: PM-CNN software is available at https://github.com/qdu-bioinfo/PM_CNN.

Assessment of nonlinear dose-response relationships via nonparametric regression.

We propose a simple approach to assess whether a nonlinear parametric model is appropriate to depict the dose-response relationships and whether two parametric models can be applied to fit a dataset via nonparametric regression. The proposed approach can compensate for the ANOVA, which is sometimes conservative, and is very easy to implement. We illustrate the performance by analyzing experimental examples and a small simulation study.

Gut microbiome predicts selenium supplementation efficiency across different Chinese adult cohorts using hybrid modeling and feature refining.

Selenium (Se) is an essential trace element that plays a vital role in various physiological functions of the human body, despite its small proportion. Due to the inability of the human body to synthesize selenium, there has been increasing concern regarding its nutritional value and adequate intake as a micronutrient. The efficiency of selenium absorption varies depending on individual biochemical characteristics and living environments, underscoring the importance of accurately estimating absorption efficiency to prevent excessive or inadequate intake. As a crucial digestive organ in the human body, gut harbors a complex and diverse microbiome, which has been found to have a significant correlation with the host's overall health status. To investigate the relationship between the gut microbiome and selenium absorption, a two-month intervention experiment was conducted among Chinese adult cohorts. Results indicated that selenium supplementation had minimal impact on the overall diversity of the gut microbiome but was associated with specific subsets of microorganisms. More importantly, these dynamics exhibited variations across regions and sequencing batches, which complicated the interpretation and utilization of gut microbiome data. To address these challenges, we proposed a hybrid predictive modeling method, utilizing refined gut microbiome features and host variable encoding. This approach accurately predicts individual selenium absorption efficiency by revealing hidden microbial patterns while minimizing differences in sequencing data across batches and regions. These efforts provide new insights into the interaction between micronutrients and the gut microbiome, as well as a promising direction for precise nutrition in the future.

Flex Meta-Storms elucidates the microbiome local beta-diversity under specific phenotypes.

MOTIVATION: Beta-diversity quantitatively measures the difference among microbial communities thus enlightening the association between microbiome composition and environment properties or host phenotypes. The beta-diversity analysis mainly relies on distances among microbiomes that are calculated by all microbial features. However, in some cases, only a small fraction of members in a community plays crucial roles. Such a tiny proportion is insufficient to alter the overall distance, which is always missed by end-to-end comparison. On the other hand, beta-diversity pattern can also be interfered due to the data sparsity when only focusing on nonabundant microbes. RESULTS: Here, we develop Flex Meta-Storms (FMS) distance algorithm that implements the "local alignment" of microbiomes for the first time. Using a flexible extraction that considers the weighted phylogenetic and functional relations of microbes, FMS produces a normalized phylogenetic distance among members of interest for microbiome pairs. We demonstrated the advantage of FMS in detecting the subtle variations of microbiomes among different states using artificial and real datasets, which were neglected by regular distance metrics. Therefore, FMS effectively discriminates microbiomes with higher sensitivity and flexibility, thus contributing to in-depth comprehension of microbe-host interactions, as well as promoting the utilization of microbiome data such as disease screening and prediction. AVAILABILITY AND IMPLEMENTATION: FMS is implemented in C++, and the source code is released at https://github.com/qdu-bioinfo/flex-meta-storms.

Communications and High-Precision Positioning (CHP2): Hardware Architecture, Implementation, and Validation.

Spectral congestion and modern consumer applications motivate radio technologies that efficiently cooperate with nearby users and provide several services simultaneously. We designed and implemented a joint positioning-communications system that simultaneously enables network communications, timing synchronization, and localization to a variety of airborne and ground-based platforms. This Communications and High-Precision Positioning (CHP2) system simultaneously performs communications and precise ranging (<10 cm) with a narrow band waveform (10 MHz) at a carrier frequency of 915 MHz (US ISM) or 783 MHz (EU Licensed). The ranging capability may be extended to estimate the relative position and orientation by leveraging the spatial diversity of the multiple-input, multiple-output (MIMO) platforms. CHP2 also digitally synchronizes distributed platforms with sub-nanosecond precision without support from external systems (GNSS, GPS, etc.). This performance is enabled by leveraging precise time-of-arrival (ToA) estimation techniques, a network synchronization algorithm, and the intrinsic cooperation in the joint processing chain that executes these tasks simultaneously. In this manuscript, we describe the CHP2 system architecture, hardware implementation, and in-lab and over-the-air experimental validation.

A novel model-checking approach for dose-response relationships.

We propose a test for assessing nonlinear dose-response models based on a Crámer-von Mises statistic. We establish the asymptotic distribution of the test and demonstrate that the test can detect the local alternative converging to the null at the parametric rate 1/n. We provide a bootstrap resampling technique to calculate the critical values. It is observed that the test has good power performance in small sample sizes. We apply the proposed method to analyze 250 datasets from a pharmacologic study and conduct two small simulation experiments to explore the numerical performance of the proposed test and compare one commonly used test in practice.

Towards multi-label classification: Next step of machine learning for microbiome research.

Machine learning (ML) has been widely used in microbiome research for biomarker selection and disease prediction. By training microbial profiles of samples from patients and healthy controls, ML classifiers constructs data models by community features that highly correlated with the target diseases, so as to determine the status of new samples. To clearly understand the host-microbe interaction of specific diseases, previous studies always focused on well-designed cohorts, in which each sample was exactly labeled by a single status type. However, in fact an individual may be associated with multiple diseases simultaneously, which introduce additional variations on microbial patterns that interferes the status detection. More importantly, comorbidities or complications can be missed by regular ML models, limiting the practical application of microbiome techniques. In this review, we summarize the typical ML approaches of single-label classification for microbiome research, and demonstrate their limitations in multi-label disease detection using a real dataset. Then we prospect a further step of ML towards multi-label classification that potentially solves the aforementioned problem, including a series of promising strategies and key technical issues for applying multi-label classification in microbiome-based studies.

A simple yet powerful test for assessing goodness-of-fit of high-dimensional linear models.

We evaluate the validity of a projection-based test checking linear models when the number of covariates tends to infinity, and analyze two gene expression datasets. We show that the test is still consistent and derive the asymptotic distributions under the null and alternative hypotheses. The asymptotic properties are almost the same as those when the number of covariates is fixed as long as p/n → 0 with additional mild assumptions. The test dramatically gains dimension reduction, and its numerical performance is remarkable.

Method development for cross-study microbiome data mining: Challenges and opportunities.

During the past decade, tremendous amount of microbiome sequencing data has been generated to study on the dynamic associations between microbial profiles and environments. How to precisely and efficiently decipher large-scale of microbiome data and furtherly take advantages from it has become one of the most essential bottlenecks for microbiome research at present. In this mini-review, we focus on the three key steps of analyzing cross-study microbiome datasets, including microbiome profiling, data integrating and data mining. By introducing the current bioinformatics approaches and discussing their limitations, we prospect the opportunities in development of computational methods for the three steps, and propose the promising solutions to multi-omics data analysis for comprehensive understanding and rapid investigation of microbiome from different angles, which could potentially promote the data-driven research by providing a broader view of the "microbiome data space".

Water quality assessment with hierarchical cluster analysis based on Mahalanobis distance.

Water quality assessment is crucial for assessment of marine eutrophication, prediction of harmful algal blooms, and environment protection. Previous studies have developed many numeric modeling methods and data driven approaches for water quality assessment. The cluster analysis, an approach widely used for grouping data, has also been employed. However, there are complex correlations between water quality variables, which play important roles in water quality assessment but have always been overlooked. In this paper, we analyze correlations between water quality variables and propose an alternative method for water quality assessment with hierarchical cluster analysis based on Mahalanobis distance. Further, we cluster water quality data collected form coastal water of Bohai Sea and North Yellow Sea of China, and apply clustering results to evaluate its water quality. To evaluate the validity, we also cluster the water quality data with cluster analysis based on Euclidean distance, which are widely adopted by previous studies. The results show that our method is more suitable for water quality assessment with many correlated water quality variables. To our knowledge, it is the first attempt to apply Mahalanobis distance for coastal water quality assessment.

Network propagation with dual flow for gene prioritization.

Based on the hypothesis that the neighbors of disease genes trend to cause similar diseases, network-based methods for disease prediction have received increasing attention. Taking full advantage of network structure, the performance of global distance measurements is generally superior to local distance measurements. However, some problems exist in the global distance measurements. For example, global distance measurements may mistake non-disease hub proteins that have dense interactions with known disease proteins for potential disease proteins. To find a new method to avoid the aforementioned problem, we analyzed the differences between disease proteins and other proteins by using essential proteins (proteins encoded by essential genes) as references. We find that disease proteins are not well connected with essential proteins in the protein interaction networks. Based on this new finding, we proposed a novel strategy for gene prioritization based on protein interaction networks. We allocated positive flow to disease genes and negative flow to essential genes, and adopted network propagation for gene prioritization. Experimental results on 110 diseases verified the effectiveness and potential of the proposed method.

Vertebral osteomyelitis complicated by iliopsoas muscle abscess in an immunocompetent adolescent: successful conservative treatment.

Vertebral osteomyelitis is rare in children. The lumbar spine is the most commonly involved region. Vertebral osteomyelitis occurs more frequently in the vertebral body, and involvement of posterior element is rare. Vertebral osteomyelitis results from hematogenous seeding, spread from contiguous infections, and direct inoculation from spinal surgery. Initial symptoms include low back pain, difficulty standing, limping gait, and fever. Blood cultures should be obtained for children with vertebral osteomyelitis because it is the definite guide for providing accurate treatment. Computed tomographyi-guided abscess aspiration should be considered for patients with negative blood cultures. Staphylococcus aureus is the most common microorganism in vertebral osteomyelitis, and the incidence of methicillin-resistant S aureus has increased in recent years. Plain radiographs, bone scintigraphy, and magnetic resonance imaging are useful for making the diagnosis. Antimicrobial therapy for 6 weeks is usually successful, and an early transition to oral form does not increase the risk of treatment failure. Debridement with implant removal is required, especially for late-onset infections associated with previous spinal surgery. Vertebral osteomyelitis can cause motor weakness and paralysis. Because of the involvement of spinal development, spinal deformities, including scoliosis and loss of normal lumbar lordosis, should be a concern in pediatric patients. Early diagnosis and adequate treatment for vertebral osteomyelitis are important to prevent severe complications and lifelong disabilities.This article describes the case of a 14-year-old boy with spontaneous lumbar vertebral osteomyelitis who initially presented with low back pain and was successfully treated nonoperatively.