Abstract Relation prediction is a critical task in knowledge graph completion and associated downstream tasks that rely on knowledge representation. Previous studies indicate that both structural ...features and semantic information are meaningful for predicting missing relations in knowledge graphs. This has led to the development of two types of methods: structure-based methods and semantics-based methods. Since these two approaches represent two distinct learning paradigms, it is difficult to fully utilize both sets of features within a single learning model, especially deep features. As a result, existing studies usually focus on only one type of feature. This leads to an insufficient representation of knowledge in current methods and makes them prone to overlooking certain patterns when predicting missing relations. In this study, we introduce a novel model, RP-ISS, which combines deep semantic and structural features for relation prediction. The RP-ISS model utilizes a two-part architecture, with the first component being a RoBERTa module that is responsible for extracting semantic features from entity nodes. The second part of the system employs an edge-based relational message-passing network designed to capture and interpret structural information within the data. To alleviate the computational burden of the message-passing network on the RoBERTa module during the sampling process, RP-ISS introduces a node embedding memory bank, which updates asynchronously to circumvent excessive computation. The model was assessed on three publicly accessible datasets (WN18RR, WN18, and FB15k-237), and the results revealed that RP-ISS surpasses all baseline methods across all evaluation metrics. Moreover, RP-ISS showcases robust performance in graph inductive learning.
The outbreak of COVID-19 has caused a huge shock for human society. As people experience the attack of the COVID-19 virus, they also are experiencing an information epidemic at the same time. Rumors ...about COVID-19 have caused severe panic and anxiety. Misinformation has even undermined epidemic prevention to some extent and exacerbated the epidemic. Social networks have allowed COVID-19 rumors to spread unchecked. Removing rumors could protect people’s health by reducing people’s anxiety and wrong behavior caused by the misinformation. Therefore, it is necessary to research COVID-19 rumor detection on social networks. Due to the development of deep learning, existing studies have proposed rumor detection methods from different perspectives. However, not all of these approaches could address COVID-19 rumor detection. COVID-19 rumors are more severe and profoundly influenced, and there are stricter time constraints on COVID-19 rumor detection. Therefore, this study proposed and verified the rumor detection method based on the content and user responses in limited time CR-LSTM-BE. The experimental results show that the performance of our approach is significantly improved compared with the existing baseline methods. User response information can effectively enhance COVID-19 rumor detection.
This paper delves into optimizing the rotation of relief supplies within the relief supply chain system, concentrating on reserve quantity decisions for governments and humanitarian organizations ...involved in disaster response. By integrating a trade-in strategy with suppliers, it ensures a precise and timely response to the fluctuating demand for relief supplies post-disaster. Utilizing the newsvendor model, optimization theory, and supply chain coordination principles, we developed a comprehensive model that calculates optimal reserve quantities for pre-positioning demanders. It also outlines the expected profit function for suppliers and a robust supply chain coordination model. The findings highlight that optimal stockpiling decisions for relief supplies are heavily influenced by cost parameters, material characteristics, and the relationship between trade-in pricing and market resale values. Notably, higher trade-in prices generally reduce the government’s optimal reserve quantities, impacting strategic decisions within supply chain coordination. This research adds to disaster management literature by offering strategic insights into how coordination and pricing strategies can improve disaster preparedness and response efficiency and effectiveness.
Clinical named entity recognition is an essential task for humans to analyze large-scale electronic medical records efficiently. Traditional rule-based solutions need considerable human effort to ...build rules and dictionaries; machine learning-based solutions need laborious feature engineering. For the moment, deep learning solutions like Long Short-term Memory with Conditional Random Field (LSTM–CRF) achieved considerable performance in many datasets. In this paper, we developed a multitask attention-based bidirectional LSTM–CRF (Att-biLSTM–CRF) model with pretrained Embeddings from Language Models (ELMo) in order to achieve better performance. In the multitask system, an additional task named entity discovery was designed to enhance the model’s perception of unknown entities. Experiments were conducted on the 2010 Informatics for Integrating Biology & the Bedside/Veterans Affairs (I2B2/VA) dataset. Experimental results show that our model outperforms the state-of-the-art solution both on the single model and ensemble model. Our work proposes an approach to improve the recall in the clinical named entity recognition task based on the multitask mechanism.
This study aimed to propose a new user-friendly, cost effective and robust risk model to facilitate risk stratification for diffuse large B-cell lymphoma (DLBCL) treated with frontline R-CHOP ...regimens.
Data on 998 patients with de novo DLBCL diagnosed between Jan 1st, 2005 and Dec 31st, 2018 at our center, who received frontline R-CHOP or R-CHOP-like regimens, were retrospectively collected. Patients were randomly divided into the training cohort (n = 701) and the validation cohort (n = 297). A new prognostic model for overall survival (OS) was built based on the training cohort. The performance of the new model was compared with International prognostic index (IPI), revised IPI (R-IPI) and National Comprehensive Cancer Network (NCCN)-IPI (NCCN-IPI). The new model was validated in the validation cohort.
The multivariate analysis of the training cohort showed that the IPI, β2-microglobulin, platelet count and red blood cell distribution width were independent factors for OS, which were incorporated into the new prognostic model. Patients were stratified into low risk, low-intermediate risk, high-intermediate risk, high risk and very high risk groups, with distinct survival outcomes. The new model achieved good C-indexes for 5-year OS prediction of 0.750 (95%CI 0.719-0.781) and 0.733 (95%CI 0.682-0.784) in the training and validation cohorts, respectively, and displayed well-fitted calibration curves. The C-index and the time-dependent ROC analysis demonstrated better performance of the new model than the IPI, R-IPI and NCCN-IPI in both training and validation cohorts. The integrated Brier score for predicting 5-year OS of the new model was lower than that of the IPI, R-IPI and NCCN-IPI in both cohorts, and decision curve analysis also showed a higher net benefit, indicating the superiority of the new model over the conventional models.
The new prognostic model might be a useful predictive tool for DLBCL treated with R-CHOP regimens. Further external validation is warranted.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
To explore the clinicobiologic features and outcomes of diffuse large B-cell lymphoma (DLBCL) patients in China according to the primary site.
A total of 1,085 patients diagnosed with DLBCL in ...National Cancer Center/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College during a 6-year period were enrolled. Their clinical characteristics and outcomes were analyzed according to the primary site.
In the 1,085 patients, 679 (62.6%) cases were nodal DLBCL (N-DLBCL) and 406 cases (37.4%) were extranodal DLBCL (EN-DLBCL). The most common sites of N-DLBCL were lymphonodus (64.8%), Waldeyer's ring (19.7%), mediastinum (12.8%) and spleen (2.7%), while in EN-DLBCL, stomach (22.4%), intestine (16.0%), nose and sinuses (8.9%), testis (8.4%), skin (7.9%), thyroid (6.9%), central nervous system (CNS) (6.4%), breast (5.7%), bone (3.4%), and salivary gland (2.7%) were most common. N-DLBCL patients tend to present B symptoms, bulky disease, and elevated LDH more often, while age >60 years, extranodal sites >1, Ann Arbor stage I or II, bone marrow involvement, and Ki-67 index >90% were usually seen in EN-DLBCL. The 5-year overall survival (OS) rate and progression-free survival (PFS) rate for all patients were 62.5% and 54.2%. The 5-year OS rate for patients with N-DLBCL and EN-DLBCL were 65.5% and 56.9% (P=0.008), and the 5-year PFS were 57.0% and 49.0% (P=0.020). Waldeyer's ring originated DLBCL possessed the highest 5-year OS rate (83.6%) and PFS rate (76.9%) in N-DLBCL. The top five EN-DLBCL subtypes with favorable prognosis were stomach, breast, nose and sinuses, lung, salivary gland, with 5-year OS rate: 70.3%, 69.6%, 69.4%, 66.7% and 63.6%, respectively. While CNS, testis, oral cavity and kidney originated EN-DLBCL faced miserable prognosis, with 5-year OS rate of 26.9%, 38.2%, and 42.9%.
In our study, primary sites were associated with clinical characteristics and outcomes. Compared with EN-DLBCL, N-DLBCL had better prognosis.
This study aimed to recognize the hub genes associated with prognosis in follicular lymphoma (FL) treated with first-line rituximab combined with chemotherapy.
RNA sequencing data of dataset GSE65135 ...(n = 24) were included in differentially expressed genes (DEGs) analysis. Weighted gene co-expression network analysis (WGCNA) was applied for exploring the coexpression network and identifying hub genes. Validation of hub genes expression and prognosis were applied in dataset GSE119214 (n = 137) and independent patient cohort from Cancer Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College (n = 32), respectively, by analyzing RNAseq expression data and serum protein concentration quantified by ELISA. The Gene Set Enrichment Analysis (GSEA), gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichments analysis were performed. CIBERSORT was applied for tumor-infiltrating immune cells (TIICs) subset analysis.
A total of 3260 DEGs were obtained, with 1861 genes upregulated and 1399 genes downregulated. Using WGCNA, eight hub genes, PLA2G2D, MMP9, PTGDS, CCL19, NFIB, YAP1, RGL1, and TIMP3 were identified. Kaplan-Meier analysis and multivariate COX regression analysis indicated that CCL19 independently associated with overall survival (OS) for FL patients treated with rituximab and chemotherapy (HR = 0.47, 95% CI 0.25-0.86, p = 0.014). Higher serum CCL19 concentration was associated with longer progression-free survival (PFS, p = 0.014) and OS (p = 0.039). TIICs subset analysis showed that CCL19 expression had a positive correlation with monocytes and macrophages M1, and a negative correlation with naïve B cells and plasma cells.
CCL19 expression was associated with survival outcomes and might be a potential prognostic biomarker for FL treated with first-line chemoimmunotherapy.
Pemetrexed combined with platinum complexes can be used as first-line treatment for advanced non-squamous non-small cell lung cancer (NSCLC), however, the efficacy and safety is varying from ...individuals. There is a need to better understand the genetic variations associated with platinum response.
We performed next-generation sequencing (NGS) based on BGI Oseq-ctDNA panel to analyze 98 longitudinal plasma samples from 32 lung adenocarcinoma patients during platinum-based chemotherapy, and a bioinformatic pipeline was developed to detect point mutations.
We found that mutation burden was decreased after chemotherapy, which reflected chemotherapy sensitivity, especially the frequency of C>G and C>A substitutions. Moreover, neoplastic cells carrying a specific set of somatic mutations, such as EGFR(L858R), KRAS (p.G12C) were obviously correlated with platinum treatment. In addition, the MAPK pathway was found to have a pivotal role in NSCLC and platinum based response. Finally, we found that smokers benefit less from platinum-based chemotherapy.
Collectively, this work described the dynamic changes of ctDNA mutation status during platinum-based treatment, which may contribute to advanced lung adenocarcinoma patients stratification and precision treatment.
Objective: Limited data about the prognostic significance of BCL2 mutations and BCL2 copy number variations in diffuse large B-cell lymphoma (DLBCL) are available. This study aimed to comprehensively ...describe BCL2 genetic alterations in DLBCL patients, and examine correlation of BCL2, TP53 and other genetic alterations with outcomes in patients treated with R-CHOP. Methods: Probe capture-based high-resolution sequencing was performed on 191 patients diagnosed with de novo DLBCL. MYC, BCL2, and BCL6 protein expressions were detected by immunohistochemistry. Results: The presence of BCL2 alterations significantly correlated with poor progression-free survival (PFS) (5-year PFS: 13.7% vs. 40.8%; P = 0.003) and overall survival (OS) (5-year OS: 34.0% vs. 70.9%; P = 0.036). Importantly, patients who harbored BCL2 gain/amplifications (BCL2GA/AMP) also had a remarkably inferior 5-year PFS (11.1% vs. 38.3%; P < 0.001) and OS (22.1% vs. 69.6%; P = 0.009). In contrast, neither BCL2 mutations nor BCL2 translocations were significantly prognostic for survival. Multivariable analyses showed that the presence of BCL2 alterations, especially BCL2GA/AMP, TP53 mutations, and International Prognostic Index (IPI) were significantly associated with inferior PFS and OS. Novel prognostic models for OS were constructed based on 3 risk factors, including BCL2 alterations (Model 1) or BCL2GA/AMP (Model 2), TP53 mutations, and IPI, to stratify patients into 4 risk groups with different survival outcomes. Conclusions: This study showed that DLBCL patients treated with R-CHOP, BCL2 alterations, especially BCL2GA/AMP and TP53 mutations were significantly associated with inferior outcomes, which were independent of the IPI. The novel prognostic models we proposed predicted outcomes for DLBCL patients treated with R-CHOP, but further validation of the prognostic models is still warranted.
Lung adenocarcinoma (LUAD) possesses a poor prognosis with a low 5-year survival rate even for stages I-III resected patients, it is thus critical to understand the determinants that affect the ...survival and discover new potentially prognostic biomarkers. Somatic copy number alterations (CNAs) are major source of genomic variations driving tumor evolution, CNAs screening may identify prognostic biomarkers.
Oncoscan MIP array was used to analyze the patterns of CNAs on formalin fixed paraffin embedded(FFPE) tumor specimens from 163 consecutive stage I-III resected LUAD patients, 145 out of which received platinum-based adjuvant chemotherapy.
Of the 163 patients, 91(55.8%) were recurred within 3 years after surgery. The most common aberrations in our cohort were 1q, 5p, 5q, 7p, 8q, 14p, 16p, 17q, 20q for copy number gains and 8p, 9p, 13p, 16q, 18q for losses. The GISTIC2 analysis produced 45 amplification peaks and 40 deletion peaks, involving some reported genes
, and
, most of which were consistent with TCGA database. The amplifications of 12p12.1 (
) and
were correlated with worse prognosis in our cohort, this result was further validated in 506 LUAD patients from TCGA. In addition, 163 patients could be well-classified into five groups, and the clinical outcomes were significantly different based on threshold copy number at reoccurring alteration peaks. Among the 145 patients who received adjuvant chemotherapy, focal amplification of
and deletion of 4q34.3 were found to be specific in relapsed patients, this result was validated in an independent group of Imielinski et al., demonstrating these two CNAs may contribute to resected LUAD recurrence after adjuvant chemotherapy.
This study suggests that CNAs profiling may be a potential prognostic classifier in resected LAUD patients. Amplifications of 12p12.1 and
might be prognostic biomarkers for LUAD, and amplification of
and deletion of 4q34.3 predicted early relapse after adjuvant chemotherapy. These novel findings may provide implication for better implementation of precision therapy for lung cancer patients.