Search

The identification of cancer driver genes is crucial for early detection, effective therapy, and precision medicine of cancer. Cancer is caused by the dysregulation of several genes at various levels of regulation. However, current techniques only capture a limited amount of regulatory information, which may hinder their efficacy. In this study, we present IMI-driver, a model that integrates multi-omics data into eight biological networks and applies Multi-view Collaborative Network Embedding to embed the gene regulation information from the biological networks into a low-dimensional vector space to identify cancer drivers. We apply IMI-driver to 29 cancer types from The Cancer Genome Atlas (TCGA) and compare its performance with nine other methods on nine benchmark datasets. IMI-driver outperforms the other methods, demonstrating that multi-level network integration enhances prediction accuracy. We also perform a pan-cancer analysis using the genes identified by IMI-driver, which confirms almost all our selected candidate genes as known or potential drivers. Case studies of the new positive genes suggest their roles in cancer development and progression. Author summary: Cancer driver identification is of great value in cancer research. Recent advancements have demonstrated that incorporating the intricate regulatory relationships among genes can facilitate driver genes identification. However, current methods do not fully capture the regulatory relationships at different layers. In this study, we propose a new method for predicting cancer driver genes (IMI-driver), which integrates multi-omics data into eight biological networks (including common networks such as protein-protein interaction networks and various cancer-specific networks) and embeds multi-level network information as features to predict driver gene sets. Experiment demonstrates that our method outperforms mainstream tools, and can discover new drivers. [ABSTRACT FROM AUTHOR]

Cell‐free DNA (cfDNA) fragmentation patterns have immense potential for early cancer detection. However, the definition of fragmentation varies, ranging from the entire genome to specific genomic regions. These patterns have not been systematically compared, impeding broader research and practical implementation. Here, 1382 plasma cfDNA sequencing samples from 8 cancer types are collected. Considering that cfDNA within open chromatin regions is more susceptible to fragmentation, 10 fragmentation patterns within open chromatin regions as features and employed machine learning techniques to evaluate their performance are examined. All fragmentation patterns demonstrated discernible classification capabilities, with the end motif showing the highest diagnostic value for cross‐validation. Combining cross and independent validation results revealed that fragmentation patterns that incorporated both fragment length and coverage information exhibited robust predictive capacities. Despite their diagnostic potential, the predictive power of these fragmentation patterns is unstable. To address this limitation, an ensemble classifier via integrating all fragmentation patterns is developed, which demonstrated notable improvements in cancer detection and tissue‐of‐origin determination. Further functional bioinformatics investigations on significant feature intervals in the model revealed its impressive ability to identify critical regulatory regions involved in cancer pathogenesis. [ABSTRACT FROM AUTHOR]

With the increasing demand for multi-variety and small-batch products, it's necessary to frequently dispatch and replace the progressive press dies on the stamping production lines to ensure the diversity of processed automobile covering parts. This paper formulates a die scheduling problem with twin-crane transportation (DSP-TCT) encountered in the stamping production line, which concentrates on the scheduling of transporting dies between the production line and warehouse by twin cranes with satisfying crane distance constraint, die position constraint, and precedence constraint. To solve DSP-TCT, this paper proposes a node sequence-based ant colony optimisation algorithm (NS-ACO). In this algorithm, each node represents a single die transportation task with action and time information executed by the twin cranes. The combination of adjacent nodes with a high time utilisation rate can be accumulated as heuristic priority knowledge for guiding optimisation. To demonstrate the effectiveness of the NS-ACO algorithm, numerical experiments with three different die stacking strategies are executed. [ABSTRACT FROM AUTHOR]