Recently, significant progress has been made in 2D human pose estimation. While some research has focused on enhancing the accuracy of keypoint detection, others have aimed at reducing model size. However, most models excel in either one aspect or the other, but rarely both simultaneously. In this paper, we address the challenge of balancing accuracy and inference speed. Inspired by large-kernel convolutions and attention mechanisms, we introduce LKConvPose, a hybrid CNN architecture that achieves high keypoint detection accuracy with low computational cost. Specifically, LKConvPose-S attains 76.5 AP on the COCO validation dataset using only 4 GFLOPs, making it the most efficient model at its scale.

Requirements terminology defines and unifies key specialized and/or technical concepts of the software system, which is significant for understanding the application domain in requirements engineering (RE). However, manual terminology extraction from natural language requirements is laborious and expensive, especially with large scale requirements specifications. In this paper, we aim to employ natural language processing (NLP) techniques and machine learning (ML) algorithms to automatically extract and rank the requirements terms to support high-level feature modeling. To this end, we propose an automatic framework composed of noun phrase identification technique for requirements terms extraction and TextRank combined with semantic similarity for terms ranking. The final ranked terms are organized as a hierarchy, which can be used to help name elements when performing feature modeling. In the quantitative evaluation, our extraction method performs better than three baseline methods in recall with comparable precision. Moreover, our adapted TextRank algorithm can rank more relevant terms at the top positions in terms of average precision compared with most baselines. An illustrative example on the smart home domain further shows the usefulness of our framework in aiding elements naming during feature modeling. The research results suggest that proper adoption and adaption of NLP and ML techniques according to the characteristics of specific RE task could provide automation support for problem domain understanding.

With the advances in gene sequencing technologies, millions of somatic mutations have been reported in the past decades, but mining cancer driver genes with oncogenic mutations from these data remains a critical and challenging area of research. In this study, we proposed a network-based classification method for identifying cancer driver genes with merging the multi-biological information. In this method, we construct a cancer specific genetic network from the human protein-protein interactome (PPI) to mine the network structure attributes, and combine biological information such as mutation frequency and differential expression of genes to achieve accurate prediction of cancer driver genes. Across seven different cancer types, the proposed algorithm always achieves high prediction accuracy, which is superior to the existing advanced methods. In the analysis of the predicted results, about 40 percent of the top 10 candidate genes overlap with the Cancer Gene Census database. Interestingly, the feature comparison indicates that the network based features are still more important than the biological features, including the mutation frequency and genetic differential expression. Further analyses also show that the integration of network structure attributes and biological information is valuable for predicting new cancer driver genes.

The lack of studying the complex organization of directed network usually limits the understanding of the underlying relationship between network structures and functions. Structural controllability and structural predictability, two seemingly unrelated subjects, are revealed in this article to be both highly dependent on the critical links previously thought to only be able to influence the number of driver nodes in controllable directed networks. Here, we show that critical links can not only contribute to structural controllability but can also have a significant impact on the structural predictability of networks, suggesting the universal pattern of structural reciprocity in directed networks. In addition, it is shown that the fraction and location of critical links have a strong influence on the performance of prediction algorithms. Moreover, these empirical results are interpreted by introducing the link centrality based on corresponding line graphs. This work bridges the gap between the two independent research fields, and it provides indications of developing advanced control strategies and prediction algorithms from a microscopic perspective.

MOTIVATION: Tumor stratification has a wide range of biomedical and clinical applications, including diagnosis, prognosis and personalized treatment. However, cancer is always driven by the combination of mutated genes, which are highly heterogeneous across patients. Accurately subdividing the tumors into subtypes is challenging. RESULTS: We developed a network-embedding based stratification (NES) methodology to identify clinically relevant patient subtypes from large-scale patients' somatic mutation profiles. The central hypothesis of NES is that two tumors would be classified into the same subtypes if their somatic mutated genes located in the similar network regions of the human interactome. We encoded the genes on the human protein-protein interactome with a network embedding approach and constructed the patients' vectors by integrating the somatic mutation profiles of 7344 tumor exomes across 15 cancer types. We firstly adopted the lightGBM classification algorithm to train the patients' vectors. The AUC value is around 0.89 in the prediction of the patient's cancer type and around 0.78 in the prediction of the tumor stage within a specific cancer type. The high classification accuracy suggests that network embedding-based patients' features are reliable for dividing the patients. We conclude that we can cluster patients with a specific cancer type into several subtypes by using an unsupervised clustering algorithm to learn the patients' vectors. Among the 15 cancer types, the new patient clusters (subtypes) identified by the NES are significantly correlated with patient survival across 12 cancer types. In summary, this study offers a powerful network-based deep learning methodology for personalized cancer medicine. AVAILABILITY AND IMPLEMENTATION: Source code and data can be downloaded from https://github.com/ChengF-Lab/NES. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Tumor-specific genomic alterations allow systematic identification of genetic interactions that promote tumorigenesis and tumor vulnerabilities, offering novel strategies for development of targeted therapies for individual patients. We develop an Individualized Network-based Co-Mutation (INCM) methodology by inspecting over 2.5 million nonsynonymous somatic mutations derived from 6,789 tumor exomes across 14 cancer types from The Cancer Genome Atlas. Our INCM analysis reveals a higher genetic interaction burden on the significantly mutated genes, experimentally validated cancer genes, chromosome regulatory factors, and DNA damage repair genes, as compared to human pan-cancer essential genes identified by CRISPR-Cas9 screenings on 324 cancer cell lines. We find that genes involved in the cancer type-specific genetic subnetworks identified by INCM are significantly enriched in established cancer pathways, and the INCM-inferred putative genetic interactions are correlated with patient survival. By analyzing drug pharmacogenomics profiles from the Genomics of Drug Sensitivity in Cancer database, we show that the network-predicted putative genetic interactions (e.g., BRCA2-TP53) are significantly correlated with sensitivity/resistance of multiple therapeutic agents. We experimentally validated that afatinib has the strongest cytotoxic activity on BT474 (IC50 = 55.5 nM, BRCA2 and TP53 co-mutant) compared to MCF7 (IC50 = 7.7 μM, both BRCA2 and TP53 wild type) and MDA-MB-231 (IC50 = 7.9 μM, BRCA2 wild type but TP53 mutant). Finally, drug-target network analysis reveals several potential druggable genetic interactions by targeting tumor vulnerabilities. This study offers a powerful network-based methodology for identification of candidate therapeutic pathways that target tumor vulnerabilities and prioritization of potential pharmacogenomics biomarkers for development of personalized cancer medicine.

Link prediction is a fundamental problem in social systems, including the prediction of user interaction and the links between users and items, which is also referred to as recommendation. Previous works mainly focus on the prediction task independently, which predict either the links between users or the links between users and items. However, these two prediction tasks are always coupled with each other, where users' preferences will influence the formation of users' interaction in coupled social networks, and vice versa. In this paper, we proposed a multitasking factor graph model (MFG) for predicting the user interaction and links between users and items simultaneously. Firstly, we observed some interesting network transfer structures according to the feature analysis between the user-user network and user-item network. Sequently, we developed a MFG model based on these network transfer structure, which can transfer information mutually from one network to the other to solve the multitasking prediction problem. Extensive experiments conducted on two real-world coupled social networks demonstrate the effectiveness of our methodology, compared with some other baseline algorithms.