Mutational Landscape of Secondary Glioblastoma Guides MET-Targeted Trial in Brain Tumor.

Low-grade gliomas almost invariably progress into secondary glioblastoma (sGBM) with limited therapeutic option and poorly understood mechanism. By studying the mutational landscape of 188 sGBMs, we find significant enrichment of TP53 mutations, somatic hypermutation, MET-exon-14-skipping (METex14), PTPRZ1-MET (ZM) fusions, and MET amplification. Strikingly, METex14 frequently co-occurs with ZM fusion and is present in ∼14% of cases with significantly worse prognosis. Subsequent studies show that METex14 promotes glioma progression by prolonging MET activity. Furthermore, we describe a MET kinase inhibitor, PLB-1001, that demonstrates remarkable potency in selectively inhibiting MET-altered tumor cells in preclinical models. Importantly, this compound also shows blood-brain barrier permeability and is subsequently applied in a phase I clinical trial that enrolls MET-altered chemo-resistant glioma patients. Encouragingly, PLB-1001 achieves partial response in at least two advanced sGBM patients with rarely significant side effects, underscoring the clinical potential for precisely treating gliomas using this therapy.

PITX2 functions as a transcription factor for GPX4 and protects pancreatic cancer cells from ferroptosis.

Ferroptosis inhibits tumor progression in pancreatic cancer cells, while PITX2 is known to function as a pro-oncogenic factor in various tumor types, protecting them from ferroptosis and thereby promoting tumor progression. In this study, we sought to investigate the regulatory role of PITX2 in tumor cell ferroptosis within the context of pancreatic cancer. We conducted PITX2 knockdown experiments using lentiviral infection in two pancreatic cancer cell lines, namely PANC-1 and BxPC-3. We assessed protein expression through immunoblotting and mRNA expression through RT-PCR. To confirm PITX2 as a transcription factor for GPX4, we employed Chromatin Immunoprecipitation (ChIP) and Dual-luciferase assays. Furthermore, we used flow cytometry to measure reactive oxygen species (ROS), lipid peroxidation, and apoptosis and employed confocal microscopy to assess mitochondrial membrane potential. Additionally, electron microscopy was used to observe mitochondrial structural changes and evaluate PITX2's regulation of ferroptosis in pancreatic cancer cells. Our findings demonstrated that PITX2, functioning as a transcription factor for GPX4, promoted GPX4 expression, thereby exerting an inhibitory effect on ferroptosis in pancreatic cancer cells and consequently promoting tumor progression. Moreover, PITX2 enhanced the invasive and migratory capabilities of pancreatic cancer cells by activating the WNT signaling pathway. Knockdown of PITX2 increased ferroptosis and inhibited the proliferation of PANC-1 and BxPC-3 cells. Notably, the inhibitory effect on ferroptosis resulting from PITX2 overexpression in these cells could be countered using RSL3, an inhibitor of GPX4. Overall, our study established PITX2 as a transcriptional regulator of GPX4 that could promote tumor progression in pancreatic cancer by reducing ferroptosis. These findings suggest that PITX2 may serve as a potential therapeutic target for combating ferroptosis in pancreatic cancer.

Pathway-based stratification of gliomas uncovers four subtypes with different TME characteristics and prognosis.

Increasing evidences have found that the interactions between hypoxia, immune response and metabolism status in tumour microenvironment (TME) have clinical importance of predicting clinical outcomes and therapeutic efficacy. This study aimed to develop a reliable molecular stratification based on these key components of TME. The TCGA data set (training cohort) and two independent cohorts from CGGA database (validation cohort) were enrolled in this study. First, the enrichment score of 277 TME-related signalling pathways was calculated by gene set variation analysis (GSVA). Then, consensus clustering identified four stable and reproducible subtypes (AFM, CSS, HIS and GLU) based on TME-related signalling pathways, which were characterized by differences in hypoxia and immune responses, metabolism status, somatic alterations and clinical outcomes. Among the four subtypes, HIS subtype had features of immunosuppression, oxygen deprivation and active energy metabolism, resulting in a worst prognosis. Thus, for better clinical application of this acquired stratification, we constructed a risk signature by using the LASSO regression model to identify patients in HIS subtype accurately. We found that the risk signature could accurately screen out the patients in HIS subtype and had important reference value for individualized treatment of glioma patients. In brief, the definition of the TME-related subtypes was a valuable tool for risk stratification in gliomas. It might serve as a reliable prognostic classifier and provide rational design of individualized treatment, and follow-up scheduling for patients with gliomas.

Decoding selectivity: computational insights into AKR1B1 and AKR1B10 inhibition.

Understanding selectivity mechanisms of inhibitors towards highly homologous proteins is of paramount importance in the design of selective candidates. Human aldo-keto reductases (AKRs) pertain to a superfamily of monomeric oxidoreductases, which serve as NADPH-dependent cytosolic enzymes to catalyze the reduction of carbonyl groups to primary and secondary alcohols using electrons from NADPH. Among AKRs, AKR1B1 is emerging as a promising target for cancer treatment and diabetes, despite its high structural similarity with AKR1B10, which leads to severe adverse events. Therefore, it is crucial to understand the selectivity mechanisms of AKR1B1 and AKR1B10 to discover safe anticancer candidates with optimal therapeutic efficacy. In this study, multiple computational strategies, including sequence alignment, structural comparison, Protein Contacts Atlas analysis, molecular docking, molecular dynamics simulation, MM-GBSA calculation, alanine scanning mutagenesis and pharmacophore modeling analysis were employed to comprehensively understand the selectivity mechanisms of AKR1B1/10 inhibition based on selective inhibitor lidorestat and HAHE. This study would provide substantial evidence in the design of potent and highly selective AKR1B1/10 inhibitors in future.

Analysis of factors influencing pancreatic fistula after minimally invasive pancreaticoduodenectomy and establishment of a new prediction model for clinically relevant pancreatic fistula.

BACKGROUND: Postoperative pancreatic fistula (POPF) is the most prevalent complications following minimally invasive pancreaticoduodenectomy (MIPD). Only one model related to MIPD exists, and previous POPF scoring prediction methods are based on open pancreaticoduodenectomy patients. Our objectives are to determine the variables that may increase the probability of pancreatic fistula following MIPD and to develop and validate a POPF predictive risk model. METHODS: Data from 432 patients who underwent MIPD between July 2015 and May 2022 were retrospectively collected. A nomogram prediction model was created using multivariate logistic regression analysis to evaluate independent factors for POPF in patients undergoing MIPD in the modeling cohort. The area under the curve (AUC) of the receiver operating characteristic curve (ROC) and the calibration curve were used to verify the nomogram prediction model internally and externally within the modeling cohort and the verification cohort. RESULTS: Multivariate logistic regression analysis showed that body mass index (BMI), albumin, triglycerides, pancreatic duct diameter, pathological diagnosis and intraoperative bleeding were independent variables for POPF. On the basis of this information, a model for the prediction of risks associated with POPF was developed. In accordance with the ROC analysis, the modeling cohort's AUC was 0.819 (95% CI 0.747-0.891), the internal validation cohort's AUC was 0.830 (95% CI 0.747-0.912), and the external validation cohort's AUC was 0.793 (95% CI 0.671-0.915). Based on the calibration curve, the estimated values of POPF have a high degree of concordance with the actual values that were measured. CONCLUSIONS: This model for predicting the probability of pancreatic fistula following MIPD has strong predictive capacity and can provide a trustworthy predictive method for the early screening of high-risk patients with pancreatic fistula after MIPD and timely clinical intervention.

Accelerated Redox Reactions Enable Stable Tin-Lead Mixed Perovskite Solar Cells.

The facile oxidation of Sn2+ to Sn4+ poses an inherent challenge that limits the efficiency and stability of tin-lead mixed (Sn-Pb) perovskite solar cells (PSCs) and all-perovskite tandem devices. In this work, we discover the sustainable redox reactions enabling self-healing Sn-Pb perovskites, where their intractable oxidation degradation can be recovered to their original state under light soaking. Quantitative and operando spectroscopies are used to investigate the redox chemistry, revealing that metallic Pb0 from the photolysis of perovskite reacts with Sn4+ to regenerate Pb2+ and Sn2+ spontaneously. Given the sluggish redox reaction kinetics, V3+ /V2+ ionic pair is designed as an effective redox shuttle to accelerate the recovery of Sn-Pb perovskites from oxidation. The target Sn-Pb PSCs enabled by V3+ /V2+ ionic pair deliver an improved power conversion efficiency (PCE) of 21.22 % and excellent device lifespan, retaining nearly 90 % of its initial PCE after maximum power point tracking under light for 1,000 hours.

SETD8 inhibits ferroptosis in pancreatic cancer by inhibiting the expression of RRAD.

BACKGROUND: As an oncogene, SETD8 can promote tumour growth and tumour cell proliferation. This study aims to reveal the relationship between SETD8 and ferroptosis in pancreatic cancer and its role in pancreatic cancer to provide a possible new direction for the comprehensive treatment of pancreatic cancer. METHODS: The downstream targets were screened by RNA sequencing analysis. Western blot, Real-time Quantitative PCR (qPCR) and immunohistochemistry showed the relationship between genes. Cell proliferation analysis and cell metabolite analysis revealed the function of genes. Chromatin immunoprecipitation (CHIP) assays were used to study the molecular mechanism. RESULTS: The potential downstream target of SETD8, RRAD, was screened by RNA sequencing analysis. A negative correlation between SETD8 and RRAD was found by protein imprinting, Real-time Quantitative PCR (qPCR) and immunohistochemistry. Through cell proliferation analysis and cell metabolite analysis, it was found that RRAD can not only inhibit the proliferation of cancer cells but also improve the level of lipid peroxidation of cancer cells. At the same time, chromatin immunoprecipitation analysis (CHIP) was used to explore the molecular mechanism by which SETD8 regulates RRAD expression. SETD8 inhibited RRAD expression. CONCLUSIONS: SETD8 interacts with the promoter region of RRAD, which epigenetically silences the expression of RRAD to reduce the level of lipid peroxidation in pancreatic cancer cells, thereby inhibiting ferroptosis in pancreatic cancer cells and resulting in poor prognosis of pancreatic cancer.

Attenuated African swine fever virus through serial passaging of viruses in cell culture: a brief review on the knowledge gathered during 60 years of research.

African swine fever virus (ASFV) is a highly pathogenic double-stranded DNA virus. It affects various breeds of pigs, causing serious economic losses and health threats because of its rapid spread and high pathogenicity and infectivity. This situation is not helped by the lack of a validated vaccine or effective therapies. Since the 1960s, different strains of ASFV have been subjected to serial passage in a variety of cell lines. The attenuated ASFV strains obtained through serial passage are not only candidates for ASF vaccine research, but also are useful to study the molecular genetic characteristics and pathogenic mechanism of the virus. This review summarizes related studies on the attenuated strains of ASFV acquired through cell passage over the last 60 years, with the aim of providing inspiration for the rational design of vaccines in future.

An MRI radiomics approach to predict survival and tumour-infiltrating macrophages in gliomas.

Preoperative MRI is one of the most important clinical results for the diagnosis and treatment of glioma patients. The objective of this study was to construct a stable and validatable preoperative T2-weighted MRI-based radiomics model for predicting the survival of gliomas. A total of 652 glioma patients across three independent cohorts were covered in this study including their preoperative T2-weighted MRI images, RNA-seq and clinical data. Radiomic features (1731) were extracted from preoperative T2-weighted MRI images of 167 gliomas (discovery cohort) collected from Beijing Tiantan Hospital and then used to develop a radiomics prediction model through a machine learning-based method. The performance of the radiomics prediction model was validated in two independent cohorts including 261 gliomas from the The Cancer Genomae Atlas database (external validation cohort) and 224 gliomas collected in the prospective study from Beijing Tiantan Hospital (prospective validation cohort). RNA-seq data of gliomas from discovery and external validation cohorts were applied to establish the relationship between biological function and the key radiomics features, which were further validated by single-cell sequencing and immunohistochemical staining. The 14 radiomic features-based prediction model was constructed from preoperative T2-weighted MRI images in the discovery cohort, and showed highly robust predictive power for overall survival of gliomas in external and prospective validation cohorts. The radiomic features in the prediction model were associated with immune response, especially tumour macrophage infiltration. The preoperative T2-weighted MRI radiomics prediction model can stably predict the survival of glioma patients and assist in preoperatively assessing the extent of macrophage infiltration in glioma tumours.

OUTER RETINAL TUBULATION IN BIETTI CRYSTALLINE DYSTROPHY ASSOCIATED WITH THE RETINAL PIGMENT EPITHELIUM ATROPHY.

PURPOSE: To determine the prognostic value of outer retinal tubulation (ORT) in the eyes of a Chinese cohort with Bietti crystalline dystrophy (BCD). METHODS: This retrospective, multicenter cohort study enrolled 42 patients with clinically and genetically diagnosed BCD. Eighty eyes with good-quality images of spectral domain optical coherence tomography were included. Demographic details and clinical data were collected. The characteristics of ORT, including prevalence, location, and morphologic characteristics were analyzed. RESULTS: Forty-two patients with BCD harbored potentially CYP4V2 disease-causing mutations. The mutation spectrum comprised 17 unique variants, 9 of which were novel. Fifty-two of these 80 eyes demonstrated evidence of ORT. The incidence of ORT is significantly higher in Stage 2 than other stages ( P < 0.001). ORT was mainly bilateral and located at the margin of the atrophic area of retinal pigment epithelium (RPE), and dynamically changed with the progressive RPE atrophy. The process of RPE atrophy was slower in eyes with ORT ( P = 0.017), with significantly longer intact RPE width in Stage 3 ( P = 0.024). Eyes with ORT had slower vision loss than eyes without ORT ( P = 0.044). CONCLUSION: ORT may be a sign of the onset of RPE atrophy in early-stage BCD and may suggest less risk of rapid progression in late-stage BCD.

Association of normalization of postoperative carbohydrate antigen 125 levels with treatment failure following uterine artery embolization for adenomyosis.

OBJECTIVE: To investigate the association between carbohydrate antigen 125 (CA125) level and adenomyosis treatment failure (TF) after uterine artery embolization (UAE). METHODS: We evaluated 224 patients with symptomatic adenomyosis who underwent UAE between January 2016 and December 2020. Improvements in dysmenorrhea and menorrhagia were assessed on the basis of symptom relief criteria. The factors associated with TF were investigated using a multivariate logistic regression model. Patients were analyzed for preoperative CA125 levels, postoperative CA125 levels, and the normalization of postoperative CA125 levels. Long-term symptom relief and quality of life after UAE were compared between the groups. RESULTS: During the 24-month follow-up, 50 patients (22.3%) experienced TF. Compared to patients in the non-TF group, those in the TF group had significantly higher preoperative and postoperative CA125 levels (p < 0.05). Multivariate analysis revealed that failure to normalize postoperative CA125 levels was independently associated with an increased risk of TF (34.7% vs. 8.5%, p < 0.001; hazard ratio 3.953, 95% confidence interval 1.567-9.973, p = 0.004). After a 3-month follow-up period, patients who normalized their CA125 levels were more likely to achieve complete necrosis on magnetic resonance imaging than those who did not (82.1% vs. 56.8%, p < 0.001). Normalization of postoperative CA125 levels was significantly associated with fewer symptoms and better quality of life 12 months after UAE (p < 0.05). CONCLUSIONS: Following UAE, normalization of postoperative CA125 levels, rather than absolute values, was the strongest predictive marker of TF.

Intelligent Hotel Guidance System via Face Recognition Technology.

In modern large hotels, due to a large number of rooms and complex layouts, it is difficult for customers to find rooms, which increases a lot of workloads for hotel attendants to guide. In this paper, a hotel intelligent guidance system based on face recognition is designed. After entering the customer's facial photos, the room guidance and customer management are carried out through face recognition. With this, hotels can move toward card-free management, green environmental protection, and save on resources. With these improvements, hotel management will be card-free and green. Each monitoring device of the system adopts dual STM32 core architecture, in which STM32H7 is responsible for face recognition, while STM32L4 is the main control chip, which is responsible for data exchange, guest room guidance and other work. The monitoring master not only guides, but also uploads customer check-in information to the cloud platform to facilitate the management of the hotel. The system adopts contactless information collection and guidance, which improves the intelligence and humanization of the hotel, and has a good application prospect.

Multimodal Attention Dynamic Fusion Network for Facial Micro-Expression Recognition.

The emotional changes in facial micro-expressions are combinations of action units. The researchers have revealed that action units can be used as additional auxiliary data to improve facial micro-expression recognition. Most of the researchers attempt to fuse image features and action unit information. However, these works ignore the impact of action units on the facial image feature extraction process. Therefore, this paper proposes a local detail feature enhancement model based on a multimodal dynamic attention fusion network (MADFN) method for micro-expression recognition. This method uses a masked autoencoder based on learnable class tokens to remove local areas with low emotional expression ability in micro-expression images. Then, we utilize the action unit dynamic fusion module to fuse action unit representation to improve the potential representation ability of image features. The state-of-the-art performance of our proposed model is evaluated and verified on SMIC, CASME II, SAMM, and their combined 3DB-Combined datasets. The experimental results demonstrated that the proposed model achieved competitive performance with accuracy rates of 81.71%, 82.11%, and 77.21% on SMIC, CASME II, and SAMM datasets, respectively, that show the MADFN model can help to improve the discrimination of facial image emotional features.

Clinical characterization and immunosuppressive regulation of CD161 (KLRB1) in glioma through 916 samples.

BACKGROUND: Glioblastoma is a paradigm of cancer-associated immunosuppression, limiting the effects of immunotherapeutic strategies. Thus, identifying the molecular mechanisms underlying immune surveillance evasion is critical. Recently, the preferential expression of inhibitory natural killer (NK) cell receptor CD161 on glioma-infiltrating cytotoxic T cells was identified. Focusing on the molecularly annotated, large-scale clinical samples from different ethnic origins, the data presented here provide evidence of this immune modulator's essential roles in brain tumor biology. METHODS: Retrospective RNA-seq data analysis was conducted in a cohort of 313 patients with glioma in the Chinese Glioma Genome Atlas (CGGA) database and 603 patients in The Cancer Genome Atlas (TCGA) database. In addition, single-cell sequencing data from seven surgical specimens of glioblastoma patients and a model in which patient-derived glioma stem cells were cocultured with peripheral lymphocytes, were used to analyze the molecular evolution process during gliomagenesis. RESULTS: CD161 was enriched in high-grade gliomas and isocitrate dehydrogenase (IDH)-wildtype glioma. CD161 acted as a potential biomarker for the mesenchymal subtype of glioma and an independent prognostic factor for the overall survival (OS) of patients with glioma. In addition, CD161 played an essential role in inhibiting the cytotoxicity of T cells in glioma patients. During the process of gliomagenesis, the expression of CD161 on different lymphocytes dynamically evolved. CONCLUSION: The expression of CD161 was closely related to the pathology and molecular pathology of glioma. Meanwhile, CD161 promoted the progression and evolution of gliomas through its unique effect on T cell dysfunction. Thus, CD161 is a promising novel target for immunotherapeutic strategies in glioma treatment.

A systematic review of genotypes and serogroups of African swine fever virus.

African swine fever virus (ASFV) is the causative agent of African swine fever (ASF). The virus causes an acute highly hemorrhagic disease in domestic pigs, with high mortality. Although the overall genome mutation rate of ASFV, a large DNA virus, is relatively low, ASFV exhibits genetic and antigenic diversity. ASFV can be classified into 24 genotypes on the basis of the B646L gene. Cross-protected ASFV strains can be divided into eight serogroups on the basis of antibody-mediated hemadsorption inhibition. Here, we review research progress on ASFV genotyping and serogrouping, and explain how this information assists in the rapid identification of virus origin during ASF outbreaks and will aid in the development of ASF vaccines.

Simultaneous Removal of Antibiotic Resistant Bacteria, Antibiotic Resistance Genes, and Micropollutants by FeS2@GO-Based Heterogeneous Photo-Fenton Process.

The co-occurrence of various chemical and biological contaminants of emerging concerns has hindered the application of water recycling. This study aims to develop a heterogeneous photo-Fenton treatment by fabricating nano pyrite (FeS2) on graphene oxide (FeS2@GO) to simultaneously remove antibiotic resistant bacteria (ARB), antibiotic resistance genes (ARGs), and micropollutants (MPs). A facile and solvothermal process was used to synthesize new pyrite-based composites. The GO coated layer forms a strong chemical bond with nano pyrite, which enables to prevent the oxidation and photocorrosion of pyrite and promote the transfer of charge carriers. Low reagent doses of FeS2@GO catalyst (0.25 mg/L) and H2O2 (1.0 mM) were found to be efficient for removing 6-log of ARB and 7-log of extracellular ARG (e-ARG) after 30 and 7.5 min treatment, respectively, in synthetic wastewater. Bacterial regrowth was not observed even after a two-day incubation. Moreover, four recalcitrant MPs (sulfamethoxazole, carbamazepine, diclofenac, and mecoprop at an environmentally relevant concentration of 10 µg/L each) were completely removed after 10 min of treatment. The stable and recyclable composite generated more reactive species, including hydroxyl radicals (HO•), superoxide radicals (O2• -), singlet oxygen (1O2). These findings highlight that the synthesized FeS2@GO catalyst is a promising heterogeneous photo-Fenton catalyst for the removal of emerging contaminants.

Genetic analysis of genome sequence characteristics of two lumpy skin disease viruses isolated from China.

BACKGROUND: Lumpy skin disease (LSD) is an acute or subacute infectious disease caused by lumpy skin disease virus (LSDV) of genus Capripoxvirus. The outbreaks of LSD were confirmed in the Yili area of the Xinjiang autonomous region in August 2019 and the Fujian province in June 2020. We detected LSDV in our daily monitoring work, then isolated, identified and sequenced the virus, and analyzed the whole genome characteristics of the isolated strain. RESULTS: Whole genome sequencing revealed that the strains isolated were all LSDV and were named as LSDV XJ201901 and LSDV FJ2019. The results showed that the identity based on whole genome sequences between LSDV XJ201901 and LSDV FJ2019 was 100% and the identity based on whole genome sequences between the two isolated strains and the global LSDV strains was 97.28%-99.99%, with the strain LSDV72/PrachuapKhiriKhan/Thailand/2021 (99.99%) having the highest sequence identity. Analysis of potential recombination events revealed that a total of 18 potential recombination events were identified in strains LSDV XJ201901 and LSDV FJ2019. The two strains are a recombination of Neethling vaccine LW 1959 (GeneBank: AF409138.1) with KSGP 0240 (GeneBank: KX683219.1). It was observed that Neethling vaccine LW 1959 (11/18) and KSGP 0240 (10/18) are involved in most of the potential recombination events. CONCLUSIONS: The virus isolate in this study was LSDV and was identified as a vaccine recombinant strain. The most likely potential parent strains of the two strains in this study are Neethling vaccine LW 1959 and KSGP 0240. The strains in this study are very similar to those isolated in East and Southeast Asia since 2019.

Choroidal and retinal thickness in patients with vitamin C deficiency using swept-source optical coherence tomography.

BACKGROUND: To investigate the effects of vitamin C on central retinal thickness and choroidal thickness. METHODS: A total of 69 patients diagnosed with vitamin C deficiency and 1:1 age- and gender-matched 69 healthy individuals with normal serum vitamin C were included in this study. Demographic characteristics of the individuals were collected. All patients underwent a comprehensive ophthalmic examination. Subfoveal choroidal thickness and retinal thickness were measured using a swept-source optical coherence tomography (SS-OCT). RESULTS: The average retinal thickness was 269.07 ± 13.51 µm in the vitamin C deficiency group and 276.92 ± 13.51 µm in the control group. The average choroidal thickness was 195.62 ± 66.40 µm in the in the vitamin C deficiency group and 238.86 ± 55.08 µm in the control group. There was a significant decrease in both average choroidal thickness and retinal thickness in vitamin C deficiency group compared with normal individuals (p < 0.001, and = 0.001 respectively). CONCLUSION: The central retinal and choroidal thickness were thinner in vitamin C deficiency group compared with normal individuals. These findings suggested that vitamin C deficiency might play an important role in retinal and choroidal diseases.

Stable Gaze Tracking with Filtering Based on Internet of Things.

Gaze tracking is basic research in the era of the Internet of Things. This study attempts to improve the performance of gaze tracking in an active infrared source gaze-tracking system. Owing to unavoidable noise interference, the estimated points of regard (PORs) tend to fluctuate within a certain range. To reduce the fluctuation range and obtain more stable results, we introduced a Kalman filter (KF) to filter the gaze parameters. Considering that the effect of filtering is relevant to the motion state of the gaze, we design the measurement noise that varies with the speed of the gaze. In addition, we used a correlation filter-based tracking method to quickly locate the pupil, instead of the detection method. Experiments indicated that the variance of the estimation error decreased by 73.83%, the size of the extracted pupil image decreased by 93.75%, and the extraction speed increased by 1.84 times. We also comprehensively discussed the advantages and disadvantages of the proposed method, which provides a reference for related research. It must be pointed out that the proposed algorithm can also be adopted in any eye camera-based gaze tracker.

Complex Embedding with Type Constraints for Link Prediction.

Large-scale knowledge graphs not only store entities and relations but also provide ontology-based information about them. Type constraints that exist in this information are of great importance for link prediction. In this paper, we proposed a novel complex embedding method, CHolE, in which complex circular correlation was introduced to extend the classic real-valued compositional representation HolE to complex domains, and type constraints were integrated into complex representational embeddings for improving link prediction. The proposed model consisted of two functional components, the type constraint model and the relation learning model, to form type constraints such as modulus constraints and acquire the relatedness between entities accurately by capturing rich interactions in the modulus and phase angles of complex embeddings. Experimental results on benchmark datasets showed that CHolE outperformed previous state-of-the-art methods, and the impartment of type constraints improved its performance on link prediction effectively.