Gene regulatory network (GRN) inference is essential for understanding gene interactions that control biological functions and disease progression. Its goal is to uncover regulatory relationships among genes using gene expression data, such as single-cell RNA sequencing (scRNA-seq). With the rise of deep learning, numerous approaches-particularly those using Graph Neural Networks (GNNs)-have shown promising progress in GRN inference. However, scRNA-seq datasets exhibit high sparsity, noise, dropout events, and cellular heterogeneity. These characteristics make it difficult for fixed GNN architectures to generalize across scRNA-seq datasets, often leading to performance degradation. To address this challenge, we propose AutoGRN, an automated graph neural network framework for gene regulatory network inference tasks. AutoGRN incorporates a range of components that potentially influence GRN inference performance into its search space. Through a genetic search algorithm constrained by information entropy, AutoGRN identifies optimal GNN architectures by adapting to the characteristics of individual datasets. Experimental results show that AutoGRN outperforms existing methods in prediction accuracy across multiple public datasets.

Drug repositioning is critical in accelerating drug discovery, which identifies new indications for existing drugs by modeling drug-disease associations. Compared to traditional methods, graph neural networks (GNNs) have recently gained widespread attention due to their ability to effectively aggregate information from neighboring nodes in drug-disease heterogeneous graphs. The GNN-based methods need effective node embeddings for information aggregation. However, they generate the node embeddings by random initialization, rather than incorporating the high-quality expert knowledge involving biological mechanisms in the databases. This limits their capacity to generate interpretable node embeddings aligned with expert knowledge. To bridge this gap, we develop a novel framework dubbed DReKGNN (Drug Repositioning based on expert Knowledge augmented Graph Neural Network). To be specific, DReKGNN will first adopt large language models (LLMs) as a semantic bridge between expert knowledge and GNNs. To ensure the accuracy of expert knowledge, DReKGNN does not rely on prompt templates in LLMs to generate knowledge descriptions for drugs and diseases. Instead, it extracts expert knowledge directly from the DrugBank and OMIM databases. The effective node embeddings with interpretable semantic information will be generated from expert knowledge descriptions involving biological mechanisms by LLMs. Then, we demonstrate that there is a need to mitigate noise when LLM node embeddings serving drug repositioning prediction tasks. Considering this design need, we integrate GNNs with LLM node embeddings by a mean aggregation strategy. The experiment results of performance comparison and case study show the effectiveness of DReKGNN in predicting drug-disease associations. The code is available at https://github.com/csubigdata-Organization/DReKGNN.

Graph neural networks (GNNs) have shown promising performance in the gene regulatory network (G RN) inference task. As mainstream GNNs are developed based on link prediction paradigm, they are susceptible to the impact of noise and imbalanced samples. Recent advances challenge this paradigm by representing gene pairs as nodes in a new graph, thereby recasting GRN inference as graph anomaly detection paradigm and demonstrating superior performance. However, we reveal that diverse GRN datasets require varied model architectures to achieve accurate prediction, and using only a fixed model architecture limits the full potential of graph anomaly detection paradigm, thus leading to suboptimal performance. In this study, we propose AutoGRAD, an automated gene regulatory network inference framework, to systematically explore the benefits of graph anomaly detection paradigm under different model architectures. Through deconstructing graph anomaly detection paradigm for GRN inference, AutoGRAD presents a novel search space consisting of various components, such as similarity computation, graph sparsification, and aggregation function. To this end, AutoGRAD adopts a reinforcement learning search strategy with a proxy training mechanism, which significantly reduces search costs while dynamically identifying the optimal model architecture specific to the dataset. Experimental results on multiple benchmark GRN datasets confirm that AutoGRAD consistently outperforms state-of-the-art baseline models, demonstrating the effectiveness of automatic model architecture adaptation for G RN inference. The code is available at https://github.com/csubigdata-Oraanization/AutoGRAD.

The recent success of Transformer has benefited many real-world applications, with its capability of building long dependency through pairwise dot-products. However, the strong assumption that elements are directly attentive to each other limits the performance of tasks with high-order dependencies such as natural language understanding and Image captioning. To solve such problems, we are the first to define the Jump Self-attention (JAT) to build Transformers. Inspired by the pieces moving of English Draughts, we introduce the spectral convolutional technique to calculate JAT on the dot-product feature map. This technique allows JAT's propagation in each self-attention head and is interchangeable with the canonical self-attention. We further develop the higher-order variants under the multi-hop assumption to increase the generality. Moreover, the proposed architecture is compatible with the pre-trained models. With extensive experiments, we empirically show that our methods significantly increase the performance on ten different tasks.

Certified separate compilation is important for establishing end-to-end guarantees for certified systems consisting of multiple program modules. There has been much work building certified compilers for sequential programs. In this paper, we propose a language-independent framework consisting of the key semantics components and lemmas that bridge the verification gap between the compilers for sequential programs and those for (race-free) concurrent programs, so that the existing verification work for the former can be reused. One of the key contributions of the framework is a novel footprint-preserving compositional simulation as the compilation correctness criterion. The framework also provides a new mechanism to support confined benign races which are usually found in efficient implementations of synchronization primitives.