PiLSL: pairwise interaction learning-based graph neural network for synthetic lethality prediction in human cancers

Interpretable high-order knowledge graph neural network for predicting synthetic lethality in human cancers

Synthetic lethality (SL) is a promising gene interaction for cancer therapy. Recent SL prediction methods integrate knowledge graphs (KGs) into graph neural networks (GNNs) and employ attention mechanisms to extract local subgraphs as explanations for target gene pairs. However, attention mechanisms often lack fidelity, typically generate a single explanation per gene pair, and fail to ensure trustworthy high-order structures in their explanations. To overcome these limitations, we propose Diverse Graph Information Bottleneck for Synthetic Lethality (DGIB4SL), a KG-based GNN that generates multiple faithful explanations for the same gene pair and effectively encodes high-order structures. Specifically, we introduce a novel DGIB objective, integrating a determinant point process constraint into the standard information bottleneck objective, and employ 13 motif-based adjacency matrices to capture high-order structures in gene representations. Experimental results show that DGIB4SL outperforms state-of-the-art baselines and provides multiple explanations for SL prediction, revealing diverse biological mechanisms underlying SL inference.

Benchmarking machine learning methods for synthetic lethality prediction in cancer

Synthetic lethality (SL) is a gold mine of anticancer drug targets, exposing cancer-specific dependencies of cellular survival. To complement resource-intensive experimental screening, many machine learning methods for SL prediction have emerged recently. However, a comprehensive benchmarking is lacking. This study systematically benchmarks 12 recent machine learning methods for SL prediction, assessing their performance across diverse data splitting scenarios, negative sample ratios, and negative sampling techniques, on both classification and ranking tasks. We observe that all the methods can perform significantly better by improving data quality, e.g., excluding computationally derived SLs from training and sampling negative labels based on gene expression. Among the methods, SLMGAE performs the best. Furthermore, the methods have limitations in realistic scenarios such as cold-start independent tests and context-specific SLs. These results, together with source code and datasets made freely available, provide guidance for selecting suitable methods and developing more powerful techniques for SL virtual screening.

Synthetic lethal connectivity and graph transformer improve synthetic lethality prediction

Synthetic lethality (SL) has shown great promise for the discovery of novel targets in cancer. CRISPR double-knockout (CDKO) technologies can only screen several hundred genes and their combinations, but not genome-wide. Therefore, good SL prediction models are highly needed for genes and gene pairs selection in CDKO experiments. However, lack of scalable SL properties prevents generalizability of SL interactions to out-of-sample data, thereby hindering modeling efforts. In this paper, we recognize that SL connectivity is a scalable and generalizable SL property. We develop a novel two-step multilayer encoder for individual sample-specific SL prediction model (MLEC-iSL), which predicts SL connectivity first and SL interactions subsequently. MLEC-iSL has three encoders, namely, gene, graph, and transformer encoders. MLEC-iSL achieves high SL prediction performance in K562 (AUPR, 0.73; AUC, 0.72) and Jurkat (AUPR, 0.73; AUC, 0.71) cells, while no existing methods exceed 0.62 AUPR and AUC. The prediction performance of MLEC-iSL is validated in a CDKO experiment in 22Rv1 cells, yielding a 46.8% SL rate among 987 selected gene pairs. The screen also reveals SL dependency between apoptosis and mitosis cell death pathways.

SL-Miner: a web server for mining evidence and prioritization of cancer-specific synthetic lethality

SUMMARY

Synthetic lethality (SL) refers to a type of genetic interaction in which the simultaneous inactivation of two genes leads to cell death, while the inactivation of a single gene does not affect cell viability. It significantly expands the range of potential therapeutic targets for anti-cancer treatments. SL interactions are primarily identified through experimental screening and computational prediction. Although various computational methods have been proposed, they tend to ignore providing evidence to support their predictions of SL. Besides, they are rarely user-friendly for biologists who likely have limited programming skills. Moreover, the genetic context specificity of SL interactions is often not taken into consideration. Here, we introduce a web server called SL-Miner, which is designed to mine the evidence of SL relationships between a primary gene and a few candidate SL partner genes in a specific type of cancer, and to prioritize these candidate genes by integrating various types of evidence. For intuitive data visualization, SL-Miner provides a range of charts (e.g. volcano plot and box plot) to help users get insights from the data.

AVAILABILITY AND IMPLEMENTATION

SL-Miner is available at https://slminer.sist.shanghaitech.edu.cn.

Graph Neural Networks in Cancer and Oncology Research: Emerging and Future Trends

Next-generation cancer and oncology research needs to take full advantage of the multimodal structured, or graph, information, with the graph data types ranging from molecular structures to spatially resolved imaging and digital pathology, biological networks, and knowledge graphs. Graph Neural Networks (GNNs) efficiently combine the graph structure representations with the high predictive performance of deep learning, especially on large multimodal datasets. In this review article, we survey the landscape of recent (2020-present) GNN applications in the context of cancer and oncology research, and delineate six currently predominant research areas. We then identify the most promising directions for future research. We compare GNNs with graphical models and "non-structured" deep learning, and devise guidelines for cancer and oncology researchers or physician-scientists, asking the question of whether they should adopt the GNN methodology in their research pipelines.

AI and machine learning for soil analysis: an assessment of sustainable agricultural practices

Sustainable agricultural practices help to manage and use natural resources efficiently. Due to global climate and geospatial land design, soil texture, soil-water content (SWC), and other parameters vary greatly; thus, real time, robust, and accurate soil analytical measurements are difficult to be developed. Conventional statistical analysis tools take longer to analyze and interpret data, which may have delayed a crucial decision. Therefore, this review paper is presented to develop the researcher's insight toward robust, accurate, and quick soil analysis using artificial intelligence (AI), deep learning (DL), and machine learning (ML) platforms to attain robustness in SWC and soil texture analysis. Machine learning algorithms, such as random forests, support vector machines, and neural networks, can be employed to develop predictive models based on available soil data and auxiliary environmental variables. Geostatistical techniques, including kriging and co-kriging, help interpolate and extrapolate soil property values to unsampled locations, improving the spatial representation of the data set. The false positivity in SWC results and bugs in advanced detection techniques are also evaluated, which may lead to wrong agricultural practices. Moreover, the advantages of AI data processing over general statistical analysis for robust and noise-free results have also been discussed in light of smart irrigation technologies. Conclusively, the conventional statistical tools for SWCs and soil texture analysis are not enough to practice and manage ergonomic land management. The broader geospatial non-numeric data are more suitable for AI processing that may soon help soil scientists develop a global SWC database.

KR4SL: knowledge graph reasoning for explainable prediction of synthetic lethality

MOTIVATION

Synthetic lethality (SL) is a promising strategy for anticancer therapy, as inhibiting SL partners of genes with cancer-specific mutations can selectively kill the cancer cells without harming the normal cells. Wet-lab techniques for SL screening have issues like high cost and off-target effects. Computational methods can help address these issues. Previous machine learning methods leverage known SL pairs, and the use of knowledge graphs (KGs) can significantly enhance the prediction performance. However, the subgraph structures of KG have not been fully explored. Besides, most machine learning methods lack interpretability, which is an obstacle for wide applications of machine learning to SL identification.

RESULTS

We present a model named KR4SL to predict SL partners for a given primary gene. It captures the structural semantics of a KG by efficiently constructing and learning from relational digraphs in the KG. To encode the semantic information of the relational digraphs, we fuse textual semantics of entities into propagated messages and enhance the sequential semantics of paths using a recurrent neural network. Moreover, we design an attentive aggregator to identify critical subgraph structures that contribute the most to the SL prediction as explanations. Extensive experiments under different settings show that KR4SL significantly outperforms all the baselines. The explanatory subgraphs for the predicted gene pairs can unveil prediction process and mechanisms underlying synthetic lethality. The improved predictive power and interpretability indicate that deep learning is practically useful for SL-based cancer drug target discovery.

AVAILABILITY AND IMPLEMENTATION

The source code is freely available at https://github.com/JieZheng-ShanghaiTech/KR4SL.

Towards artificial intelligence to multi-omics characterization of tumor heterogeneity in esophageal cancer

Esophageal cancer is a unique and complex heterogeneous malignancy, with substantial tumor heterogeneity: at the cellular levels, tumors are composed of tumor and stromal cellular components; at the genetic levels, they comprise genetically distinct tumor clones; at the phenotypic levels, cells in distinct microenvironmental niches acquire diverse phenotypic features. This heterogeneity affects almost every process of esophageal cancer progression from onset to metastases and recurrence, etc. Intertumoral and intratumoral heterogeneity are major obstacles in the treatment of esophageal cancer, but also offer the potential to manipulate the heterogeneity themselves as a new therapeutic strategy. The high-dimensional, multi-faceted characterization of genomics, epigenomics, transcriptomics, proteomics, metabonomics, etc. of esophageal cancer has opened novel horizons for dissecting tumor heterogeneity. Artificial intelligence especially machine learning and deep learning algorithms, are able to make decisive interpretations of data from multi-omics layers. To date, artificial intelligence has emerged as a promising computational tool for analyzing and dissecting esophageal patient-specific multi-omics data. This review provides a comprehensive review of tumor heterogeneity from a multi-omics perspective. Especially, we discuss the novel techniques single-cell sequencing and spatial transcriptomics, which have revolutionized our understanding of the cell compositions of esophageal cancer and allowed us to determine novel cell types. We focus on the latest advances in artificial intelligence in integrating multi-omics data of esophageal cancer. Artificial intelligence-based multi-omics data integration computational tools exert a key role in tumor heterogeneity assessment, which will potentially boost the development of precision oncology in esophageal cancer.

Multi-view graph convolutional network for cancer cell-specific synthetic lethality prediction

Synthetic lethal (SL) genetic interactions have been regarded as a promising focus for investigating potential targeted therapeutics to tackle cancer. However, the costly investment of time and labor associated with wet-lab experimental screenings to discover potential SL relationships motivates the development of computational methods. Although graph neural network (GNN) models have performed well in the prediction of SL gene pairs, existing GNN-based models are not designed for predicting cancer cell-specific SL interactions that are more relevant to experimental validation in vitro. Besides, neither have existing methods fully utilized diverse graph representations of biological features to improve prediction performance. In this work, we propose MVGCN-iSL, a novel multi-view graph convolutional network (GCN) model to predict cancer cell-specific SL gene pairs, by incorporating five biological graph features and multi-omics data. Max pooling operation is applied to integrate five graph-specific representations obtained from GCN models. Afterwards, a deep neural network (DNN) model serves as the prediction module to predict the SL interactions in individual cancer cells (iSL). Extensive experiments have validated the model's successful integration of the multiple graph features and state-of-the-art performance in the prediction of potential SL gene pairs as well as generalization ability to novel genes.