Predicting gene regulatory links from single-cell RNA-seq data using graph neural networks.

Single-cell RNA-sequencing (scRNA-seq) has emerged as a powerful technique for studying gene expression patterns at the single-cell level. Inferring gene regulatory networks (GRNs) from scRNA-seq data provides insight into cellular phenotypes from the genomic level. However, the high sparsity, noise and dropout events inherent in scRNA-seq data present challenges for GRN inference. In recent years, the dramatic increase in data on experimentally validated transcription factors binding to DNA has made it possible to infer GRNs by supervised methods. In this study, we address the problem of GRN inference by framing it as a graph link prediction task. In this paper, we propose a novel framework called GNNLink, which leverages known GRNs to deduce the potential regulatory interdependencies between genes. First, we preprocess the raw scRNA-seq data. Then, we introduce a graph convolutional network-based interaction graph encoder to effectively refine gene features by capturing interdependencies between nodes in the network. Finally, the inference of GRN is obtained by performing matrix completion operation on node features. The features obtained from model training can be applied to downstream tasks such as measuring similarity and inferring causality between gene pairs. To evaluate the performance of GNNLink, we compare it with six existing GRN reconstruction methods using seven scRNA-seq datasets. These datasets encompass diverse ground truth networks, including functional interaction networks, Loss of Function/Gain of Function data, non-specific ChIP-seq data and cell-type-specific ChIP-seq data. Our experimental results demonstrate that GNNLink achieves comparable or superior performance across these datasets, showcasing its robustness and accuracy. Furthermore, we observe consistent performance across datasets of varying scales. For reproducibility, we provide the data and source code of GNNLink on our GitHub repository: https://github.com/sdesignates/GNNLink.

Application of multivariate time-series model for high performance computing (HPC) fault prediction.

Aiming at the high reliability demand of increasingly large and complex supercomputing systems, this paper proposes a multidimensional fusion CBA-net (CNN-BiLSTAM-Attention) fault prediction model based on HDBSCAN clustering preprocessing classification data, which can effectively extract and learn the spatial and temporal features in the predecessor fault log. The model can effectively extract and learn the spatial and temporal features from the predecessor fault logs, and has the advantages of high sensitivity to time series features and sufficient extraction of local features, etc. The RMSE of the model for fault occurrence time prediction is 0.031, and the prediction accuracy of node location for fault occurrence is 93% on average, as demonstrated by experiments. The model can achieve fast convergence and improve the fine-grained and accurate fault prediction of large supercomputers.

Gene Regulatory Network Inference Using Convolutional Neural Networks from scRNA-seq Data.

In recent years, with the rapid development of single-cell sequencing technology, this brings new opportunities and challenges to reconstruct gene regulatory networks. On the one hand, scRNA-seq data reveal statistical information of gene expression at single-cell resolution, which is beneficial to construct gene expression regulatory networks. On the other hand, the noise and dropout of single-cell data bring great difficulties to the analysis of scRNA-seq data, resulting in lower accuracy of gene regulatory networks reconstructed by traditional methods. In this article, we propose a novel supervised convolutional neural network (CNNSE), which can extract gene expression information from 2D co-expression matrices of gene doublets and identify interactions between genes. Our method can avoid the loss of extreme point interference by constructing a 2D co-expression matrix of gene pairs and significantly improve the regulation precision between gene pairs. And the CNNSE model is able to obtain detailed and high-level semantic information from the 2D co-expression matrix. Our method achieves satisfactory results on simulated data [accuracy (ACC): 0.712, F1: 0.724]. On two real scRNA-seq datasets, our method exhibits higher stability and accuracy in inference tasks compared with other existing gene regulatory network inference algorithms.

Reconstructing gene regulatory networks of biological function using differential equations of multilayer perceptrons.

BACKGROUND: Building biological networks with a certain function is a challenge in systems biology. For the functionality of small (less than ten nodes) biological networks, most methods are implemented by exhausting all possible network topological spaces. This exhaustive approach is difficult to scale to large-scale biological networks. And regulatory relationships are complex and often nonlinear or non-monotonic, which makes inference using linear models challenging. RESULTS: In this paper, we propose a multi-layer perceptron-based differential equation method, which operates by training a fully connected neural network (NN) to simulate the transcription rate of genes in traditional differential equations. We verify whether the regulatory network constructed by the NN method can continue to achieve the expected biological function by verifying the degree of overlap between the regulatory network discovered by NN and the regulatory network constructed by the Hill function. And we validate our approach by adapting to noise signals, regulator knockout, and constructing large-scale gene regulatory networks using link-knockout techniques. We apply a real dataset (the mesoderm inducer Xenopus Brachyury expression) to construct the core topology of the gene regulatory network and find that Xbra is only strongly expressed at moderate levels of activin signaling. CONCLUSION: We have demonstrated from the results that this method has the ability to identify the underlying network topology and functional mechanisms, and can also be applied to larger and more complex gene network topologies.

Dual Properties of Polyvinyl Alcohol-Based Magnetorheological Plastomer with Different Ratio of DMSO/Water.

Polyvinyl alcohol (PVA)-based magnetorheological plastomer (MRP) possesses excellent magnetically dependent mechanical properties such as the magnetorheological effect (MR effect) when exposed to an external magnetic field. PVA-based MRP also shows a shear stiffening (ST) effect, which is very beneficial in fabricating pressure sensor. Thus, it can automatically respond to external stimuli such as shear force without the magnetic field. The dual properties of PVA-based MRP mainly on the ST and MR effect are rarely reported. Therefore, this work empirically investigates the dual properties of this smart material under the influence of different solvent compositions (20:80, 40:60, 60:40, and 80:20) by varying the ratios of binary solvent mixture (dimethyl sulfoxide (DMSO) to water). Upon applying a shear stress with excitation frequencies from 0.01 to 10 Hz, the storage modulus (G') for PVA-based MRP with DMSO to water ratio of 20:40 increases from 6.62 × 10-5 to 0.035 MPa. This result demonstrates an excellent ST effect with the relative shear stiffening effect (RSTE) up to 52,827%. In addition, both the ST and MR effect show a downward trend with increasing DMSO content to water. Notably, the physical state of hydrogel MRP could be changed with different solvent ratios either in the liquid-like or solid-like state. On the other hand, a transient stepwise experiment showed that the solvent's composition had a positive effect on the arrangement of CIPs within the matrix as a function of the external magnetic field. Therefore, the solvent ratio (DMSO/water) can influence both ST and MR effects of hydrogel MRP, which need to be emphasized in the fabrication of hydrogel MRP for appropriate applications primarily with soft sensors and actuators for dynamic motion control.