PrivFR: Privacy-Enhanced Federated Recommendation With Shared Hash Embedding.

Federated recommender systems (FRSs), with their improved privacy-preserving advantages to jointly train recommendation models from numerous devices while keeping user data distributed, have been widely explored in modern recommender systems (RSs). However, conventional FRSs require transmitting the entire model between the server and clients, which brings a huge carbon footprint for cost-conscious cross-device learning tasks. While several efforts have been dedicated to improving the efficiency of FRSs, it's suboptimal to treat the whole model as the objective of compact design. Besides, current research fails to handle the out-of-vocabulary (OOV) issue in real-world FRSs, where the items only occasionally appear in the testing phase but were not observed during the training process, which is another practical challenge and has not been well studied yet. To this end, we propose a privacy-enhanced federated recommendation framework with shared hash embedding, PrivFR, in cross-device settings, which is an efficient representation mechanism specialized for the embedding parameters without compromising the model capability. Specifically, it represents items in a resource-efficient way by delicately utilizing shared hash embedding and multiple hash functions. As such, it just maintains a small shared pool of hash embedding in local clients, rather than fitting all embedding vectors for each item, which can exactly achieve the dual advantages of conserving resources and handling the OOV issue. What's more, we prove that this mechanism can protect the data privacy of local clients from a theoretical perspective. Extensive experiments show that our method not only effectively reduces storage and communication overheads, but also outperforms state-of-the-art FRSs.

Combating antimicrobial resistance: the silent war.

Once hailed as miraculous solutions, antibiotics no longer hold that status. The excessive use of antibiotics across human healthcare, agriculture, and animal husbandry has given rise to a broad array of multidrug-resistant (MDR) pathogens, posing formidable treatment challenges. Antimicrobial resistance (AMR) has evolved into a pressing global health crisis, linked to elevated mortality rates in the modern medical era. Additionally, the absence of effective antibiotics introduces substantial risks to medical and surgical procedures. The dwindling interest of pharmaceutical industries in developing new antibiotics against MDR pathogens has aggravated the scarcity issue, resulting in an exceedingly limited pipeline of new antibiotics. Given these circumstances, the imperative to devise novel strategies to combat perilous MDR pathogens has become paramount. Contemporary research has unveiled several promising avenues for addressing this challenge. The article provides a comprehensive overview of these innovative therapeutic approaches, highlighting their mechanisms of action, benefits, and drawbacks.

Sources and transport of CO2 in the karst system of Jiguan Cave, Funiu Mountains, China.

Conveyance and modification of carbon-isotope signals within the karst system remain difficult to constrain, due to the complexity of interactions between multiple components, including precipitation, bedrock, soil, atmosphere, and biota. Cave monitoring is thus critical to understanding both their transport in the karst system and dependence on local hydroclimatic conditions. Jiguan Cave, located in Funiu Mountain in central China, is representative of karst tourist caves with relatively thin epikarst zone. We conducted a comprehensive monitoring program of cave climate from 2018 to 2021 and measured δ13C during 2021 in monthly and heavy-rainfall samples of soil CO2, cave CO2, cave water (drip water and underground river), and underground river outlet. Our results demonstrate synchronous variations between CO2 concentration and δ13CCO2 in both soil and cave air on seasonal time scales. Cave pCO2 and carbon-isotope composition further exhibited a high sensitivity to human respiration with fluctuations of ~2000-3000 ppm within 4 days during the cave closure period in July 2021 without tourists. 13C-depleted isotopic signal of cave air in summer is the mixture of human respiration and soil CO2 which varies as a function of regional hydrological conditions of the summer monsoon during the rainy season with high temperatures and humidity. However, respired CO2 from the overlying soil was expected to be the only principal source of the cave CO2 when the anthropogenic CO2 source was removed. The high seasonal amplitude of cave air δ13CCO2 reflects ventilation dynamics, which leads to a prominent contribution from the external atmosphere during winter. Intriguingly, although the δ13C signal reflects complex vertical processes in the vertical karst profile, a heavy summer rainfall event was related to anomalously high δ13C values of cave water that can be utilized to interpret rainfall intensity and regional hydroclimate.

Fully biobased hydrogel based on chitosan and tannic acid coated cotton fabric for underwater superoleophobicity and efficient oil/water separation.

Underwater superoleophobic (UWSO) materials have garnered significant attention in separating oil/water mixtures. But, the majority of these materials are made from non-degradable and non-renewable raw materials, polluting the environment and wasting scarce resources while using them. Against this backdrop, this study aimed to fabricate an environmental-friendly UWSO textile using biobased materials. To achieve this, hydrogel consisting of chitosan (CS) and poly(tannic acid) (PTA) were formed and coated on cotton fabric (CTF) via dip-coating followed by oxidative polymerization. CS&PTA hydrogel endowed the CTF with a rough surface and high surface energy, leading to an UWSO CTF with an underwater oil contact angle as high as 166.84°. The CS&PTA/CTF had excellent separation capability toward various oil/water mixtures, showing separation efficiency above 99.84 % and water flux higher than 23, 999 L m-2 h-1. Moreover, CS&PTA/CTF possessed excellent mechanical and environmental stability with underwater superoleophobicity unchanged after sandpaper friction, ultrasonication, organic solvents, NaCl (m/v, 30 %) solution, and acid/base solution immersion, due to the strong interaction between the hydrogel and cotton fabric generated by the mussel-inspired adhesion owing to the presence of PTA. The fully biobased UWSO CTF exhibits great promising to be an alternative to traditional superwetting materials for separation of oil/water mixtures.

SASAN: Shape-Adaptive Set Abstraction Network for Point-Voxel 3D Object Detection.

Point-voxel 3D object detectors have achieved impressive performance in complex traffic scenes. However, they utilize the 3D sparse convolution (spconv) layers with fixed receptive fields, such as voxel-based detectors, and inherit the fixed sphere radius from point-based methods for generating the features of keypoints, which make them weak in adaptively modeling various geometrical deformations and sizes of real objects. To tackle this issue, we propose a shape-adaptive set abstraction network (SASAN) for point-voxel 3D object detection. First, the proposal and offset generation module is adopted to learn the coordinates and confidences of 3D proposals and shape-adaptive offsets of the certain number of offset points for each voxel. Meanwhile, an extra offset supervision task is employed to guide the learning of shifting values of offset points, aiming at motivating the predicted offsets to preferably adapt to the various shapes of objects. Then, the shape-adaptive set abstraction module is proposed to extract multiscale keypoints features by grouping the neighboring offset points' features, as well as features learned from adjacent raw points and the 2-D bird-view map. Finally, the region of interest (RoI)-grid proposal refinement module is used to aggregate the keypoints features for further proposal refinement and confidence prediction. Extensive experiments on the competitive KITTI 3D detection benchmark demonstrate that the proposed SASAN gains superior performance as compared with state-of-the-art methods.

Fabrication of well-dispersed cellulose nanocrystal reinforced biobased epoxy composites using reversibility of covalent adaptable network.

Cellulose nanocrystal (CNC) shows great potential in reinforced composites but it is difficult to disperse in epoxy thermosets due to its poor dispersity in epoxy monomers. Herein, we reported a novel approach to disperse CNC in epoxidized soybean oil (ESO)-derived epoxy thermosets uniformly by using the reversibility of dynamic imine-containing ESO-derived covalent adaptable network (CAN). The crosslinked CAN was deconstructed by an exchange reaction with ethylenediamine (EDA) in dimethyl formamide (DMF), leading to a solution of deconstructed CAN with plenty of hydroxyl and amino groups, which could form strong hydrogen bonds with hydroxyl groups of CNC and thus facilitated and stabilized dispersion of CNC in the deconstructed CAN solution. Epoxy composite with well-dispersed CNC was finally achieved by a reformation of CAN through the removal of DMF and EDA. In this way, the epoxy composites with CNC content up to 30 wt% were successfully prepared and showed drastically reinforced mechanical properties. The tensile strength and Young's modulus of the CAN were improved by up to ∼70 % and ∼45 times with the incorporation of 20 and 30 wt% CNC, respectively. The composites showed excellent reprocessability without significant loss in mechanical properties after reprocessing.

The Battlefield of Chemotherapy in Pediatric Cancers.

The survival rate for pediatric cancers has remarkably improved in recent years. Conventional chemotherapy plays a crucial role in treating pediatric cancers, especially in low- and middle-income countries where access to advanced treatments may be limited. The Food and Drug Administration (FDA) approved chemotherapy drugs that can be used in children have expanded, but patients still face numerous side effects from the treatment. In addition, multidrug resistance (MDR) continues to pose a major challenge in improving the survival rates for a significant number of patients. This review focuses on the severe side effects of pediatric chemotherapy, including doxorubicin-induced cardiotoxicity (DIC) and vincristine-induced peripheral neuropathy (VIPN). We also delve into the mechanisms of MDR in chemotherapy to the improve survival and reduce the toxicity of treatment. Additionally, the review focuses on various drug transporters found in common types of pediatric tumors, which could offer different therapeutic options.

A multi-functional nano-system combining PI3K-110α/ß inhibitor overcomes P-glycoprotein mediated MDR and improves anti-cancer efficiency.

P-glycoprotein (P-gp/ABCB1)-mediated multidrug resistance (MDR) in cancers severely limit chemotherapeutic efficacy. We recently reported that phosphatidylinositol-3-kinase (PI3K) 110α and 110ß subunits can be novel targets for reversal of P-gp mediated MDR in cancers, and BAY-1082439 as an inhibitor specific for PI3K 110α and 110ß subunits could reverse P-gp-mediated MDR by downregulating P-gp expression in cancer cells. However, BAY-1082439 has very low solubility, short half-life and high in-vivo clearance rate. Till now, nano-system with the functions to target PI3K P110α and P110ß and reverse P-gp mediated MDR in cancers has not been reported. In our study, a tumor targeting drug delivery nano-system PBDF was established, which comprised doxorubicin (DOX) and BAY-1082439 respectively encapsulated by biodegradable PLGA-SH nanoparticles (NPs) that were grafted to gold nanorods (Au NRs) modified with FA-PEG-SH, to enhance the efficacy to reverse P-gp mediated MDR and to target tumor cells, further, to enhance the efficiency to inhibit MDR tumors overexpressing P-gp. In-vitro experiments indicated that PBDF NPs greatly enhanced uptake of DOX, improved the activity to reverse MDR, inhibited the cell proliferation, and induced S-phase arrest and apoptosis in KB-C2 cells, as compared with free DOX combining free BAY-1082439. In-vivo experiments further demonstrated that PBDF NPs improved the anti-tumor ability of DOX and inhibited development of KB-C2 tumors. Notably, the metastasis of KB-C2 cells in livers and lungs of nude mice were inhibited by treatment with PBDF NPs, which showed no obvious in-vitro or in-vivo toxicity.

Fully biobased poly(lactic acid)/lignin composites compatibilized by epoxidized natural rubber.

Biobased poly(lactic acid)/lignin (PLA/lignin) composites are limited by poor mechanical properties resulted from poor compatibility and low interfacial adhesion. Herein, we reported a novel approach to improve compatibility and interfacial adhesion of PLA/lignin composites via reactive compatibilization with epoxidized natural rubber (ENR) as a compatibilizer. Interfacial tension calculation indicated that lignin tended to act as interfacial phase between PLA and ENR, but morphology analysis demonstrated lignin was wrapped with a layer of ENR and dispersed in PLA matrix, which was attributed to the interfacial reaction of ENR with both PLA and lignin. The interfacial reaction was confirmed by Fourier transform infrared spectroscopy. The compatibility and interfacial adhesion between PLA and lignin were improved significantly by incorporation and increase in the content of ENR, as evidenced by the reduced interfacial gaps, blurry phase boundaries, and enhanced elastic response. As such, the mechanical properties of PLA/lignin composites were enhanced significantly. The tensile strength and elongation at break of PLA/lignin (W/W, 80/20) were improved by 15 % and 77 %, respectively, with the incorporation of only 1 wt% ENR. We believe this approach to compatibilize PLA/lignin composites is promising because it would not require costly modification of lignin and would not compromise the sustainability of composites.

Nivolumab and relatlimab for the treatment of melanoma.

Melanoma is a highly lethal type of skin cancer. Although an early diagnosis, in combination with surgery for nonmetastatic melanomas, significantly increases the probability of survival, there are no efficacious treatments for metastatic melanoma. Nivolumab and relatlimab are monoclonal antibodies that selectively interact with and block the proteins programmed cell death protein 1 (PD-1) and lymphocyte activation protein 3 (LAG-3), respectively, and thus, their activation by their cognate ligands. The combination of these immunotherapy drugs was approved in 2022 by the United States Food and Drug Administration (FDA) for the treatment of melanoma. Data from clinical trials indicated that, compared to nivolumab monotherapy, nivolumab and relatlimab produced more than a 2-fold median increase in progression-free survival (PFS) and a higher response rate in melanoma patients. This is an important finding as the response of patients to immunotherapies is limited due to dose-limiting toxicities and secondary drug resistance. This review article will discuss the pathogenesis of melanoma and the pharmacology of nivolumab and relatlimab. In addition, we will provide i) a summary of the anticancer drugs that inhibit LAG-3 and PD-1 in cancer patients and ii) our perspective about the use of nivolumab in combination with relatlimab to treat melanoma.

Prior Knowledge Regularized Self-Representation Model for Partial Multilabel Learning.

Partial multilabel learning (PML) aims to learn from training data, where each instance is associated with a set of candidate labels, among which only a part is correct. The common strategy to deal with such a problem is disambiguation, that is, identifying the ground-truth labels from the given candidate labels. However, the existing PML approaches always focus on leveraging the instance relationship to disambiguate the given noisy label space, while the potentially useful information in label space is not effectively explored. Meanwhile, the existence of noise and outliers in training data also makes the disambiguation operation less reliable, which inevitably decreases the robustness of the learned model. In this article, we propose a prior label knowledge regularized self-representation PML approach, called PAKS, where the self-representation scheme and prior label knowledge are jointly incorporated into a unified framework. Specifically, we introduce a self-representation model with a low-rank constraint, which aims to learn the subspace representations of distinct instances and explore the high-order underlying correlation among different instances. Meanwhile, we incorporate prior label knowledge into the above self-representation model, where the prior label knowledge is regarded as the complement of features to obtain an accurate self-representation matrix. The core of PAKS is to take advantage of the data membership preference, which is derived from the prior label knowledge, to purify the discovered membership of the data and accordingly obtain more representative feature subspace for model induction. Enormous experiments on both synthetic and real-world datasets show that our proposed approach can achieve superior or comparable performance to state-of-the-art approaches.

The intergenerational succession and financialization of Chinese family enterprises: Considering the influence of heirs' growing experience.

As a mixed-methods research in economics and psychology, this study aimed to analyze the influence from the intergenerational succession on the financialization level including asset financialization and revenue financialization, and further test the moderating effect of the heirs' typical growing experience according to The Imprinting Theory, based on the 2009-2020 annual data of listed family enterprises of China. There were two key findings. First, the effect of Chinese family enterprises' intergenerational succession on asset financialization was positively significant while the effect on revenue financialization was not significant, indicating that the financialization behavior has not brought about effective financial profits. Second, among the heirs' typical growing experiences, their parents' entrepreneurial experience during their childhood, oversea study experience, and MBA education experience had the significantly positive moderating effects on the influence from intergeneration succession to asset financialization level of Chinese family enterprises, which was an important internal mechanism for the heirs to promote the financialization process of family enterprises.

The Roles of Exosomal microRNAs in Diffuse Large B-Cell Lymphoma: Diagnosis, Prognosis, Clinical Application, and Biomolecular Mechanisms.

Background: Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous neoplasm and is characterized as the most common subtype of non-Hodgkin lymphoma (NHL). Despite 60-70% of all patients being cured with R-CHOP therapeutic regimen (Cyclophosphamide, doxorubicin, vincristine, and prednisone, combined with rituximab), remaining patients display aggressive disease. Therefore, there is an urgent need to develop novel diagnostic, prognostic, and predictive biomarkers. Recently, exosomal miRNAs have been approved as novel biomarkers in DLBCL due to their potential involvement in lymphomagenesis. Material and Methods: We conducted an investigation on the potential role of exosomal miRNAs as diagnostic, prognostic, and predictive biomarkers in DLBCL in the PubMed, Scopus, and Web of Science search engines. We searched by using a combination of keywords, such as diffuse large B-cell lymphoma, DLBCL, miRNA, microRNA, miR, exosome, exosomes, exosomal, extracellular vesicles, EVs, and secretome. Then, search results were narrowed based on specific inclusion and exclusion criteria. Results: Twelve articles were eligible for our systematic reviews. Among them, nine discussed diagnostic biomarkers, three considered prognostic significance, four evaluated therapeutic efficacy, two studies were conducted in vitro, and three assessed molecular pathways associated with these exosomal miRNAs in DLBCL. Discussion: According to our systematic review, exosomal miRNAs are not only useful for diagnosis and prognosis in DLBCL but are also promising therapeutic tools and predictors of response to therapy. Although promising results so far, more research is required to develop innovative biomarkers.

CDK6-PI3K signaling axis is an efficient target for attenuating ABCB1/P-gp mediated multi-drug resistance (MDR) in cancer cells.

BACKGROUND: Multidrug resistance (MDR) mediated by ATP binding cassette subfamily B member 1 (ABCB1/P-gp) is a major cause of cancer chemotherapy failure, but the regulation mechanisms are largely unknown. METHODS: Based on single gene knockout, we studied the regulation of CDK6-PI3K axis on ABCB1-mediated MDR in human cancer cells. CRISPR/Cas9 technique was performed in KB-C2 cells to knockout cdk6 or cdk4 gene. Western blot, RT-PCR and transcriptome analysis were performed to investigate target gene deletion and expression of critical signaling factors. The effect of cdk4 or cdk6 deficiency on cell apoptosis and the cell cycle was analyzed using flow cytometry. In vivo studies were performed to study the sensitivity of KB-C2 tumors to doxorubicin, tumor growth and metastasis. RESULTS: Deficiency of cdk6 led to remarkable downregulation of ABCB1 expression and reversal of ABCB1-mediated MDR. Transcriptomic analysis revealed that CDK6 knockout regulated a series of signaling factors, among them, PI3K 110α and 110ß, KRAS and MAPK10 were downregulated, and FOS-promoting cell autophagy and CXCL1-regulating multiple factors were upregulated. Notably, PI3K 110α/110ß deficiency in-return downregulated CDK6 and the CDK6-PI3K axis synergizes in regulating ABCB1 expression, which strengthened the regulation of ABCB1 over single regulation by either CDK6 or PI3K 110α/110ß. High frequency of alternative splicing (AS) of premature ABCB1 mRNA induced by CDK6, CDK4 or PI3K 110α/110ß level change was confirmed to alter the ABCB1 level, among them 10 common skipped exon (SE) events were found. In vivo experiments demonstrated that loss of cdk6 remarkably increased the sensitivity of KB-C2 tumors to doxorubicin by increasing drug accumulation of the tumors, resulting in remarkable inhibition of tumor growth and metastasis, as well as KB-C2 survival in the nude mice. CONCLUSIONS: CDK6-PI3K as a new target signaling axis to reverse ABCB1-mediated MDR is reported for the first time in cancers. Pathways leading to inhibition of cancer cell proliferation were revealed to be accompanied by CDK6 deficiency.

Ribociclib Inhibits P-gp-Mediated Multidrug Resistance in Human Epidermoid Carcinoma Cells.

The efficacy of cancer chemotherapy can be attenuated or abrogated by multidrug resistance (MDR) in cancer cells. In this study, we determined the effect of the CDK4/6 inhibitor, ribociclib (or LEE011), on P-glycoprotein (P-gp)-mediated MDR in the human epidermoid carcinoma MDR cell line, KB-C2, which is widely used for studying P-gp-mediated MDR in cancers. The incubation of KB-C2 cells with ribociclib (3-9 µM) increased the efficacy of colchicine, a substrate for P-gp. The cell expression of P-gp was down-regulated at both translation and transcription levels. Furthermore, ribociclib produced a 3.5-fold increase in the basal activity of P-gp ATPase, and the concentration required to increase basal activity by 50% (EC50) was 0.04 µM. Docking studies indicated that ribociclib interacted with the drug-substrate binding site of P-gp. The short-term and long-term intracellular accumulation of doxorubicin greatly increased in the KB-C2 cells co-cultured with ribociclib, indicating ribociclib inhibited the drug efflux activity of P-gp. The results of our study indicate that LEE011 may be a potential agent for combined therapy of the cancers with P-gp mediated MDR.

Mitochondrial oxidative phosphorylation is dispensable for survival of CD34+ chronic myeloid leukemia stem and progenitor cells.

Chronic myeloid leukemia (CML) are initiated and sustained by self-renewing malignant CD34+ stem cells. Extensive efforts have been made to reveal the metabolic signature of the leukemia stem/progenitor cells in genomic, transcriptomic, and metabolomic studies. However, very little proteomic investigation has been conducted and the mechanism regarding at what level the metabolic program was rewired remains poorly understood. Here, using label-free quantitative proteomic profiling, we compared the signature of CD34+ stem/progenitor cells collected from CML individuals with that of healthy donors and observed significant changes in the abundance of enzymes associated with aerobic central carbonate metabolic pathways. Specifically, CML stem/progenitor cells expressed increased tricarboxylic acid cycle (TCA) with decreased glycolytic proteins, accompanying by increased oxidative phosphorylation (OXPHOS) and decreased glycolysis activity. Administration of the well-known OXPHOS inhibitor metformin eradicated CML stem/progenitor cells and re-sensitized CD34+ CML cells to imatinib in vitro and in patient-derived tumor xenograft murine model. However, different from normal CD34+ cells, the abundance and activity of OXPHOS protein were both unexpectedly elevated with endoplasmic reticulum stress induced by metformin in CML CD34+ cells. The four major aberrantly expressed protein sets, in contrast, were downregulated by metformin in CML CD34+ cells. These data challenged the dependency of OXPHOS for CML CD34+ cell survival and underlined the novel mechanism of metformin. More importantly, it suggested a strong rationale for the use of tyrosine kinase inhibitors in combination with metformin in treating CML.

Biobased mussel-inspired underwater superoleophobic chitosan derived complex hydrogel coated cotton fabric for oil/water separation.

Superhydrophilic and underwater superoleophobic materials exhibit excellent oil/water separation performance but are usually fabricated from nonrenewable and nondegradable feedstocks and thus would cause secondary pollution after use. Herein, we report a fully biobased and mussel-inspired underwater superoleophobic hydrogel coated cotton fabric (CF) prepared by surface coating and subsequent oxidation polymerization of chitosan & dopamine mixtures. The obtained chitosan & polydopamine hydrogel coated CF (CS&PDA/CF) showed superhydrophilicity and underwater superoleophobicity, due to the formed rough surface structure with hydrophilic complex hydrogel. The CS&PDA/CF exhibited excellent oil/water separation performance with separation efficiency higher than 99.5% for various oil/water mixtures. Moreover, the CS&PDA/CF showed excellent resistance against various harsh conditions such as boiling water, ultrasonication, and concentrated salt solution, due to the mussel-inspired strong adhesion stabilized structure and morphology. We believe that the fully biobased and mussel-inspired underwater superoleophobic cotton fabric shows great potential as an eco-friendly and high-efficient oil/water separation material.

STAT6 Blockade Abrogates Aspergillus-Induced Eosinophilic Chronic Rhinosinusitis and Asthma, A Model of Unified Airway Disease.

Unified airway disease, including concurrent asthma and chronic rhinosinusitis (CRS), is a common, but poorly understood disorder with no curative treatment options. To establish a murine model of chronic unified eosinophilic airway inflammation, mice were challenged with Aspergillus niger, and sinonasal mucosa and lung tissue were evaluated by immunohistochemistry, flow cytometry, and gene expression. Inhalation of A niger conidia resulted in a Th2-biased lung and sinus inflammation that typifies allergic asthma and CRS. Gene network and pathway analysis correlated with human disease with upregulation of not only the JAK-STAT and helper T-cell pathways, but also less expected pathways governing the spliceosome, osteoclast differentiation, and coagulation pathways. Utilizing a specific inhibitor and gene-deficient mice, we demonstrate that STAT6 is required for mycosis-induced sinus inflammation. These findings confirm the relevance of this new model and portend future studies that further extend our understanding of the immunopathologic basis of airway mycosis and unified airway disease.

Research progress in overcoming ibrutinib drug resistance.

Ibrutinib, an oral small-molecule targeted drug, has been the first Bruton tyrosine kinase (BTK) inhibitor in the world to be approved for the market. It works by regulating cell proliferation, apoptosis and migration, and has been proven to exhibit high efficacy and good safety in the treatment of B-cell lymphomas, including chronic lymphocytic leukemia or small lymphocytic lymphoma and mantle cell lymphoma. However, some patients inevitably have drug resistance and disease recurrence, resulting in a poor prognosis. This article serves as a clinical reference by summarizing the related literature on ibrutinib resistance inhibitors.

Neighborhood Pattern Is Crucial for Graph Convolutional Networks Performing Node Classification.

Graph convolutional networks (GCNs) are widely believed to perform well in the graph node classification task, and homophily assumption plays a core rule in the design of previous GCNs. However, some recent advances on this area have pointed out that homophily may not be a necessity for GCNs. For deeper analysis of the critical factor affecting the performance of GCNs, we first propose a metric, namely, neighborhood class consistency (NCC), to quantitatively characterize the neighborhood patterns of graph datasets. Experiments surprisingly illustrate that our NCC is a better indicator, in comparison to the widely used homophily metrics, to estimate GCN performance for node classification. Furthermore, we propose a topology augmentation graph convolutional network (TA-GCN) framework under the guidance of the NCC metric, which simultaneously learns an augmented graph topology with higher NCC score and a node classifier based on the augmented graph topology. Extensive experiments on six public benchmarks clearly show that the proposed TA-GCN derives ideal topology with higher NCC score given the original graph topology and raw features, and it achieves excellent performance for semi-supervised node classification in comparison to several state-of-the-art (SOTA) baseline algorithms.