An Integrated LSTM-Rule-Based Fusion Method for the Localization of Intelligent Vehicles in a Complex Environment.

To improve the accuracy and robustness of autonomous vehicle localization in a complex environment, this paper proposes a multi-source fusion localization method that integrates GPS, laser SLAM, and an odometer model. Firstly, fuzzy rules are constructed to accurately analyze the in-vehicle localization deviation and confidence factor to improve the initial fusion localization accuracy. Then, an odometer model for obtaining the projected localization trajectory is constructed. Considering the high accuracy of the odometer's projected trajectory within a short distance, we used the shape of the projected localization trajectory to inhibit the initial fusion localization noise and used trajectory matching to obtain an accurate localization. Finally, the Dual-LSTM network is constructed to predict the localization and build an electronic fence to guarantee the safety of the vehicle while also guaranteeing the updating of short-distance localization information of the vehicle when the above-mentioned fusion localization is unreliable. Under the limited arithmetic condition of the vehicle platform, accurate and reliable localization is realized in a complex environment. The proposed method was verified by long-time operation on the real vehicle platform, and compared with the EKF fusion localization method, the average root mean square error of localization was reduced by 66%, reaching centimeter-level localization accuracy.

The relationship between innate/adaptive immunity and gastrointestinal cancer : a multi-omics Mendelian randomization study.

BACKGROUND: Innate/adaptive immunity is the key to anti-tumor therapy. However, its causal relationship to Gastrointestinal (GI) cancer remains unclear. METHODS: Immunity genes were extracted from the MSigDB database. The Genome-wide association studies (GWAS) summary data of GI cancer were integrated with expression quantitative trait loci (eQTL) and DNA methylation quantitative trait loci (mQTL) associated with genes. Summary-data-based Mendelian randomization (SMR) and co-localization analysis were used to reveal causal relationships between genes and GI cancer. Two-sample MR analysis was used for sensitivity analysis. Single cell analysis clarified the enrichment of genes. RESULTS: Three-step SMR analysis showed that a putative mechanism, cg17294865 CpG site regulating HLA-DRA expression was negatively associated with gastric cancer risk. HLA-DRA was significantly differentially expressed in monocyte/macrophage and myeloid cells in gastric cancer. CONCLUSION: This study provides evidence that upregulating the expression level of HLA-DRA can reduce the risk of gastric cancer.

Novel targets and therapies of metformin in dementia: old drug, new insights.

Dementia is a devastating disorder characterized by progressive and persistent cognitive decline, imposing a heavy public health burden on the individual and society. Despite numerous efforts by researchers in the field of dementia, pharmacological treatments are limited to relieving symptoms and fail to prevent disease progression. Therefore, studies exploring novel therapeutics or repurposing classical drugs indicated for other diseases are urgently needed. Metformin, a first-line antihyperglycemic drug used to treat type 2 diabetes, has been shown to be beneficial in neurodegenerative diseases including dementia. This review discusses and evaluates the neuroprotective role of metformin in dementia, from the perspective of basic and clinical studies. Mechanistically, metformin has been shown to improve insulin resistance, reduce neuronal apoptosis, and decrease oxidative stress and neuroinflammation in the brain. Collectively, the current data presented here support the future potential of metformin as a potential therapeutic strategy for dementia. This study also inspires a new field for future translational studies and clinical research to discover novel therapeutic targets for dementia.

Epidemiological characteristics and antibody kinetics of elderly population with booster vaccination following both Omicron BA.5 and XBB waves in China.

After the termination of zero-COVID-19 policy, the populace in China has experienced both Omicron BA.5 and XBB waves. Considering the poor antibody responses and severe outcomes observed among the elderly following infection, we conducted a longitudinal investigation to examine the epidemiological characteristics and antibody kinetics among 107 boosted elderly participants following the Omicron BA.5 and XBB waves. We observed that 96 participants (89.7%) were infected with Omicron BA.5, while 59 (55.1%) participants were infected with Omicron XBB. Notably, 52 participants (48.6%) experienced dual infections of both Omicron BA.5 and XBB. The proportion of symptomatic cases appeared to decrease following the XBB wave (18.6%) compared to that after the BA.5 wave (59.3%). Omicron BA.5 breakthrough infection induced lower neutralizing antibody titers against XBB.1.5, BA.2.86, and JN.1, while reinfection with Omicron XBB broadened the antibody responses against all measured Omicron subvariants and may alleviate the wild type-vaccination induced immune imprinting. Boosted vaccination type and comorbidities were the significant factors associated with antibody responses. Updated vaccines based on emerging severe acute respiratory syndrome coronavirus 2 variants are needed to control the Coronavirus Disease 2019 pandemic in the elderly.

Harnessing genetic interactions for prediction of immune checkpoint inhibitors response signature in cancer cells.

Genetic interactions (GIs) refer to two altered genes having a combined effect that is not seen individually. They play a crucial role in influencing drug efficacy. We utilized CGIdb 2.0 (http://www.medsysbio.org/CGIdb2/), an updated database of comprehensively published GIs information, encompassing synthetic lethality (SL), synthetic viability (SV), and chemical-genetic interactions. CGIdb 2.0 elucidates GIs relationships between or within protein complex models by integrating protein-protein physical interactions. Additionally, we introduced GENIUS (GENetic Interactions mediated drUg Signature) to leverage GIs for identifying the response signature of immune checkpoint inhibitors (ICIs). GENIUS identified high MAP4K4 expression as a resistant signature and high HERC4 expression as a sensitive signature for ICIs treatment. Melanoma patients with high expression of MAP4K4 were associated with decreased efficacy and poorer survival following ICIs treatment. Conversely, overexpression of HERC4 in melanoma patients correlated with a positive response to ICIs. Notably, HERC4 enhances sensitivity to immunotherapy by facilitating antigen presentation. Analyses of immune cell infiltration and single-cell data revealed that B cells expressing MAP4K4 may contribute to resistance to ICIs in melanoma. Overall, CGIdb 2.0, provides integrated GIs data, thus serving as a crucial tool for exploring drug effects.

Parallel Driving with Big Models and Foundation Intelligence in Cyber-Physical-Social Spaces.

Recent years have witnessed numerous technical breakthroughs in connected and autonomous vehicles (CAVs). On the one hand, these breakthroughs have significantly advanced the development of intelligent transportation systems (ITSs); on the other hand, these new traffic participants introduce more complex and uncertain elements to ITSs from the social space. Digital twins (DTs) provide real-time, data-driven, precise modeling for constructing the digital mapping of physical-world ITSs. Meanwhile, the metaverse integrates emerging technologies such as virtual reality/mixed reality, artificial intelligence, and DTs to model and explore how to realize improved sustainability, increased efficiency, and enhanced safety. More recently, as a leading effort toward general artificial intelligence, the concept of foundation model was proposed and has achieved significant success, showing great potential to lay the cornerstone for diverse artificial intelligence applications across different domains. In this article, we explore the big models embodied foundation intelligence for parallel driving in cyber-physical-social spaces, which integrate metaverse and DTs to construct a parallel training space for CAVs, and present a comprehensive elucidation of the crucial characteristics and operational mechanisms. Beyond providing the infrastructure and foundation intelligence of big models for parallel driving, this article also discusses future trends and potential research directions, and the "6S" goals of parallel driving.

Determining soil conservation strategies: Ecological risk thresholds of arsenic and the influence of soil properties.

The establishment of ecological risk thresholds for arsenic (As) plays a pivotal role in developing soil conservation strategies. However, despite many studies regarding the toxicological profile of As, such thresholds varying by diverse soil properties have rarely been established. This study aims to address this gap by compiling and critically examining an extensive dataset of As toxicity data sourced from existing literature. Furthermore, to augment the existing information, experimental studies on As toxicity focusing on barley-root elongation were carried out across various soil types. The As concentrations varied from 12.01 to 437.25 mg/kg for the effective concentrations that inhibited 10% of barley-root growth (EC10). The present study applied a machine-learning approach to investigate the complex associations between the toxicity thresholds of As and diverse soil properties. The results revealed that Mn-/Fe-ox and clay content emerged as the most influential factors in predicting the EC10 contribution. Additionally, by using a species sensitivity distribution model and toxicity data from 21 different species, the hazardous concentration for x% of species (HCx) was calculated for four representative soil scenarios. The HC5 values for acidic, neutral, alkaline, and alkaline calcareous soils were 80, 47, 40, and 28 mg/kg, respectively. This study establishes an evidence-based methodology for deriving soil-specific guidance concerning As toxicity thresholds.

Robust Multiobjective Reinforcement Learning Considering Environmental Uncertainties.

Numerous real-world decision or control problems involve multiple conflicting objectives whose relative importance (preference) is required to be weighed in different scenarios. While Pareto optimality is desired, environmental uncertainties (e.g., environmental changes or observational noises) may mislead the agent into performing suboptimal policies. In this article, we present a novel multiobjective optimization paradigm, robust multiobjective reinforcement learning (RMORL) considering environmental uncertainties, to train a single model that can approximate robust Pareto-optimal policies across the entire preference space. To enhance policy robustness against environmental changes, an environmental disturbance is modeled as an adversarial agent across the entire preference space via incorporating a zero-sum game into a multiobjective Markov decision process (MOMDP). Additionally, we devise an adversarial defense technique against observational perturbations, which ensures that policy variations, perturbed by adversarial attacks on state observations, remain within bounds under any specified preferences. The proposed technique is assessed in five multiobjective environments with continuous action spaces, showcasing its effectiveness through comparisons with competitive baselines, which encompass classical and state-of-the-art schemes.

Unraveling Immune Heterogeneity across Pan-Cancer and Deep Insights in Lung Adenocarcinoma based on Alternative Splicing.

Alternative splicing (AS) participates in tumor development and tumor microenvironment formation. However, the landscape of immune infiltrating AS events (IIASE) in pan-cancer and mechanisms of AS in lung adenocarcinoma (LUAD) have not been comprehensively characterized. We systematically profiled the IIASE landscape of pan-cancer using data from The Cancer Genome Atlas (TCGA), analyzing both commonalities and specific characteristics among different cancer types. We found that AS events tend to occur specifically in one cancer type rather than in multiple cancer types. AS events were used to classify 512 LUAD samples into two subtypes by unsupervised clustering: aberrant splicing subtype (ABS) and immune infiltrating subtype (IIS). The two subtypes showed significant differences in clinicopathology, prognosis, transcriptomics, genomics and immune microenvironment. We constructed a classification signature comprising 10 genes involved in 14 AS events using Logistic regression. The robustness of the signature was validated in three independent datasets using survival analysis. To explore AS mechanisms in LUAD, we constructed subtype-specific co-expression networks using Pearson correlation analysis. AS event of AKT3 regulated by splicing factor ENOX1 was associated with poor prognosis in LUAD. Overall, we outline AS events associated with immune infiltration in pan-cancer and this study provides insights into AS mechanisms in LUAD patient classification.

circAMN1-Mediated Ferroptosis Regulates the Expulsion of Placenta in Trophoblast Cells.

After delivery, the death of trophoblast cells can promote the expulsion of the placenta. Ferroptosis, an iron-dependent programmed cell death, is involved in mammalian development. Circular RNAs are associated with placental development; however, it is unclear whether circular RNAs regulate the expulsion of fetal membranes through ferroptosis. The gene expression profiles in the tail vein blood of Holstein cows with normal and retained placentas were investigated using RNA sequencing and a GSE214588 dataset. circAMN1 and SLC39A8 expression was significantly downregulated in the blood of cows with a retained placenta, whereas miR-205_R-1 expression was significantly upregulated. We validated erastin-induced ferroptosis in trophoblast cells. Transfection with si-circAMN1 and miR-205_R-1 mimic reduced intracellular total iron, Fe2+, and glutathione disulfide levels; increased intracellular glutathione levels and glutathione/glutathione disulfide; and enhanced cell viability in these cells. In contrast, transfection with pcDNA3.1 circAMN1 and an miR-205_R-1 inhibitor promoted ferroptosis. As an miR-205_R-1 sponge, circAMN1 regulated the expression of SLC39A8 to control erastin-induced ferroptosis and regulated the proliferation, invasion, and migration of trophoblast cells. Our findings provide a theoretical basis for studying the mechanism by which programmed cell death regulates fetal membrane expulsion and indicate its potential as a therapeutic target for placenta retention.

The safety and efficacy of neoadjuvant immunochemotherapy following laparoscopic gastrectomy for gastric cancer: a multicenter Real-world clinical study.

BACKGROUND: The safety and efficacy of neoadjuvant immunochemotherapy (nICT) for locally advanced gastric cancer (LAGC) remain controversial. METHODS: Patients with LAGC who received either nICT or neoadjuvant chemotherapy (nCT) at 3 tertiary referral teaching hospitals in China between January 2016 and October 2022 were analysed. After propensity-score matching (PSM), comparing the radiological response, pathological response rate, perioperative outcomes, and early recurrence between the two groups. RESULTS: After PSM, 585 patients were included, with 195 and 390 patients comprising the nICT and nCT groups, respectively. The nICT group exhibited a higher objective response rate (79.5% versus [vs.] 59.0%; P<0.001), pathological complete response rate (14.36% vs. 6.41%; P=0.002) and major pathological response rate (39.49% vs. 26.15%; P=0.001) compared with the nCT group. The incidence of surgical complications (17.44% vs. 16.15%, P=0.694) and proportion of perioperative textbook outcomes (80.0% vs. 81.0%; P=0.767) were similar in both groups. The nICT group had a significantly lower proportion of early recurrence than the nCT group (29.7% vs. 40.8%; P=0.047). Furthermore, the multivariable logistic analysis revealed that immunotherapy was an independent protective factor against early recurrence (odds ratio 0.62 [95% CI 0.41-0.92]; P=0.018). No significant difference was found in neoadjuvant therapy drug toxicity between the two groups (51.79% vs. 45.38%; P=0.143). CONCLUSIONS: Compared with nCT, nICT is safe and effective, which significantly enhanced objective and pathological response rates, and reduced the risk for early recurrence among patients with LAGC. TRIAL REGISTRATION: Clinical Trials.gov.

Effects of periodic heat events on the reproduction and longevity of female and male Agasicles hygrophila (Coleoptera: Chrysomelidae).

Alternanthera philoxeroides (Amaranthaceae), commonly known as alligator weed, is a globally invasive and detrimental perennial weed. Agasicles hygrophila serves as an important biocontrol agent for alligator weeds. However, during mid-summer, when temperatures increase, A. hygrophila populations experience a significant decline, leading to ineffective weed control. This study has examined the impact of periodic heat events on the reproduction and survival of A. hygrophila females and males using various mating combinations and durations of temperature treatments. The results demonstrated significant effects on all of the studied parameters across mating combinations when compared with the control. Under the same temperature combination, the fecundity and survival rates of females, as well as the egg-hatching rate, decreased significantly with increasing repeated heat exposure. Furthermore, the egg-hatching rate varied significantly among different temperatures and time-interval combinations. In addition, the females displayed greater sensitivity to heat stress than males in terms of fecundity. These findings enhance our understanding of A. hygrophila population dynamics during summer and provide insights into the release of biocontrol agents in diverse regions with varying climates.

The global distribution and the risk prediction of relapsing fever group Borrelia: a data review with modelling analysis.

BACKGROUND: The recent discovery of emerging relapsing fever group Borrelia (RFGB) species, such as Borrelia miyamotoi, poses a growing threat to public health. However, the global distribution and associated risk burden of these species remain uncertain. We aimed to map the diversity, distribution, and potential infection risk of RFGB. METHODS: We searched PubMed, Web of Science, GenBank, CNKI, and eLibrary from Jan 1, 1874, to Dec 31, 2022, for published articles without language restriction to extract distribution data for RFGB detection in vectors, animals, and humans, and clinical information about human patients. Only articles documenting RFGB infection events were included in this study, and data for RFGB detection in vectors, animals, or humans were composed into a dataset. We used three machine learning algorithms (boosted regression trees, random forest, and least absolute shrinkage and selection operator logistic regression) to assess the environmental, ecoclimatic, biological, and socioeconomic factors associated with the occurrence of four major RFGB species: Borrelia miyamotoi, Borrelia lonestari, Borrelia crocidurae, and Borrelia hermsii; and mapped their worldwide risk level. FINDINGS: We retrieved 13 959 unique studies, among which 697 met the selection criteria and were used for data extraction. 29 RFGB species have been recorded worldwide, of which 27 have been identified from 63 tick species, 12 from 61 wild animals, and ten from domestic animals. 16 RFGB species caused human infection, with a cumulative count of 26 583 cases reported from Jan 1, 1874, to Dec 31, 2022. Borrelia recurrentis (17 084 cases) and Borrelia persica (2045 cases) accounted for the highest proportion of human infection. B miyamotoi showed the widest distribution among all RFGB, with a predicted environmentally suitable area of 6·92 million km2, followed by B lonestari (1·69 million km2), B crocidurae (1·67 million km2), and B hermsii (1·48 million km2). The habitat suitability index of vector ticks and climatic factors, such as the annual mean temperature, have the most significant effect among all predictive models for the geographical distribution of the four major RFGB species. INTERPRETATION: The predicted high-risk regions are considerably larger than in previous reports. Identification, surveillance, and diagnosis of RFGB infections should be prioritised in high-risk areas, especially within low-income regions. FUNDING: National Key Research and Development Program of China.

Evaluation Method of Fatigue Life for Asphalt Pavement on the Steel Bridge Deck Based on the Inhomogeneous Poisson Stochastic Process.

The paving layer on the steel box girder bridge deck is widely used when constructing pavements for steel bridges. Owing to the orthotropic feature of steel decks, a transverse clapboard and rib can lead to a concentration of stress. Consequently, fatigue cracks are often identified in asphalt concrete pavement layers due to re-compaction caused by heavy vehicles. This study aims to derive an evaluation method of fatigue life for asphalt pavement based on the inhomogeneous Poisson stochastic process in view of the highly random and uncertain working conditions of layered composite structures. According to the inhomogeneous Poisson stochastic process, along with Miner's fatigue damage accumulation theory and the linear elastic fracture mechanics theory, the fatigue life formula could be deduced. Meanwhile, fatigue experiments for asphalt concrete are designed to investigate the correlation between the theoretical formula and the actual fatigue damage life of the material. Compared with the test, the accuracy error is within 10%, which is better than other traditional methods. Therefore, the fatigue life prediction model could better reflect the loading order effect and the interaction between loads, providing a new path for the fatigue reliability design of steel bridge deck asphalt pavement.

Examination of the role of the rotating nursing department in the training of nursing staff based on SWOT analysis.

BACKGROUND: The implementation of the rotation system in the Chinese medical industry has achieved significant results. OBJECTIVES: The present study aims to 1) explore the strengths, weaknesses, opportunities and challenges of rotational nursing department implementation and 2) provide references for developing nursing staff's competencies in leadership, performance evaluation, quality of care, communication in relationships and human resources. METHODS: A total of 16 rotational nursing department staff members from a tertiary tuberculosis specialist hospital in Beijing were interviewed, and the interview data were analysed using a strengths, weaknesses, opportunities and threats analysis and class analysis. RESULTS: The advantages of the rotational nursing department included: (1) stimulating the nursing staff's enthusiasm and creativity; (2) strengthening the communication and collaboration between departments; (3) improving the detailed management of nursing quality; and (4) enhancing the nursing staff's comprehensive abilities. The disadvantages included: (1) the design of the rotation programme focusing on practice; (2) a lack of personalisation; and (3) imperfect performance assessment of the rotating staff. Opportunities included: (1) deepening the connotation of nursing job management and (2) developing the construction of nursing discipline and the need for personal career development and value realisation. Threats included the lack of a sound rotation management model to draw on. CONCLUSION: A rotational nursing department is conducive to enhancing the competence of nursing staff in management positions and providing new ideas for hospitals to select and train nursing management talents. By taking full advantage of the benefits of vertical nursing management, designing personalised rotation training programmes, building a diversified learning and training platform and developing a positive performance incentive mechanism is recommended to fully engage the role of rotation in nursing management talent training.

Expanding carbon neutrality strategies: Incorporating out-of-boundary emissions in city-level frameworks.

Cities are increasingly vital in global carbon mitigation efforts, yet few have specifically tailored carbon neutrality pathways. Furthermore, out-of-boundary indirect greenhouse gas (GHG) emissions, aside from those related to electricity and heat imports, are often overlooked in existing pathways, despite their significance in comprehensive carbon mitigation strategies. Addressing this gap, here we introduce an integrated analysis framework focusing on both production and consumption-related GHG emissions. Applied to Wuyishan, a service-oriented city in Southern China, this framework provides a holistic view of a city's carbon neutrality pathway, from a full-scope GHG emission perspective. The findings reveal the equal importance of carbon reduction within and outside the city's boundaries, with out-of-boundary emissions accounting for 42% of Wuyishan's present total GHG emissions. This insight highlights the necessity of including these external factors in GHG accounting and mitigation strategy development. This framework serves as a practical tool for cities, particularly in developing countries, to craft effective carbon neutrality roadmaps that encompass the full spectrum of GHG emissions.

Development and Validation of a Prognostic Model for Postoperative Anastomotic Recurrence in Siewert II or III Adenocarcinomas Without Neoadjuvant Therapy in an East Asian Population.

PURPOSE: Anastomotic recurrence leads to poor prognosis in patients with Siewert II or III adenocarcinoma who undergo radical gastrectomy and do not receive neoadjuvant therapy. We aimed to establish a prognostic model to evaluate the risk of postoperative anastomotic recurrence in patients with Siewert II or III adenocarcinoma who did not receive neoadjuvant therapy. METHODS: We included 366 patients with Siewert II or III adenocarcinoma who were treated with radical gastrectomy without neoadjuvant therapy at Fujian Provincial Hospital (FPH) between 2012 and 2018 as the development cohort. Cox regression was used to verify prognostic factors for anastomotic recurrence, and a nomogram was established. The nomogram was externally validated using a combined cohort of two external centers. Patients were classified into high- or low-risk groups according to the diagnostic threshold and nomogram scores, and recurrence-related survival analysis was analyzed. RESULTS: The average age was 64.6 years, and 285 patients were male. All surgeries were successfully performed (185 open vs 181 laparoscopic). The 3-year anastomotic recurrence rate was significantly lower in the low-risk group (3.5% vs 18.8%, P < 0.001). The predictive performance was verified in the external validation cohort. This model better stratified patient survival than the American Joint Committee on Cancer (AJCC) TNM staging system. CONCLUSIONS: This novel nomogram with surgical margin, postoperative tumor node metastasis (pTNM) stage, and neural invasion as prognostic factors has a significant predictive performance for the risk of anastomotic recurrence after radical gastrectomy in patients with Siewert II or III adenocarcinoma.

LncRNA TYMSOS is a novel prognostic biomarker associated with immune infiltration in prostate cancer.

The long noncoding RNA thymidylate synthetase opposite strand (lncRNA TYMSOS) plays an important role in cancers; however, its impact on prostate cancer (PCa) is still unclear. By analyzing the online data, we found that lncRNA TYMSOS was highly expressed in PCa and associated with T stage, Gleason score, age, and primary therapy outcome. The results of the ROC curve showed that lncRNA TYMSOS has a significant diagnostic ability. Furthermore, Kaplan-Meier analyses suggested that lncRNA TYMSOS plays an important role in progression-free survival (PFS). Increased lncRNA TYMSOS expression was an independent risk factor correlated with PFS in PCa patients. GSEA and GSVA indicated that the lncRNA TYMSOS was involved in the cell cycle, neurodegenerative diseases, oxidative phosphorylation, spliceosomes, and adaptive immune system pathways. Additionally, lncRNA TYMSOS expression was also associated with immune cell infiltrates and tumor mutational burden in PCa. Functional experiments were further conducted, and we verified that lncRNA TYMSOS played an oncogenic role in regulating PCa aggressiveness. Specifically, silencing of lncRNA TYMSOS suppressed cell proliferation, division and epithelial-mesenchymal transition (EMT) but promoted cell apoptosis in PCa cells, and conversely, lncRNA TYMSOS overexpression had the opposite effects. In summary, our study revealed that the lncRNA TYMSOS could be a biomarker and therapeutic target in PCa and participate in tumor-immune cell infiltration.

Smart-seq2 Technology Reveals a Novel Mechanism That Zearalenone Inhibits the In Vitro Maturation of Ovine Oocytes by Influencing TNFAIP6 Expression.

Zearalenone (ZEN), a non-steroidal estrogenic fungal toxin widely present in forage, food, and their ingredients, poses a serious threat to animal and human reproductive health. ZEN also threatens ovine, a major source of human food and breeding stock. However, the mechanisms underlying the impact of ZEN on the in vitro maturation (IVM) of ovine oocytes remain unclear. This study aimed to elucidate these mechanisms using the Smart-seq2 technology. A total of 146 differentially expressed genes were obtained, using Smart-seq2, from sheep oocytes cultured in vitro after ZEN treatment. ZEN treatment inhibited RUNX2 and SPP1 expression in the PI3K signaling pathway, leading to the downregulation of THBS1 and ultimately the downregulation of TNFAIP6; ZEN can also decrease TNFAIP6 by reducing PTPRC and ITGAM. Both inhibit in vitro maturation of ovine oocytes and proliferation of cumulus cells by downregulating TNFAIP6. These findings provide data and a theoretical basis for elucidating ZEN's toxicity mechanisms, screening therapeutic drugs, and reducing ZEN-related losses in the ovine industry.

Sampling Efficient Deep Reinforcement Learning Through Preference-Guided Stochastic Exploration.

Stochastic exploration is the key to the success of the deep Q -network (DQN) algorithm. However, most existing stochastic exploration approaches either explore actions heuristically regardless of their Q values or couple the sampling with Q values, which inevitably introduce bias into the learning process. In this article, we propose a novel preference-guided ϵ -greedy exploration algorithm that can efficiently facilitate exploration for DQN without introducing additional bias. Specifically, we design a dual architecture consisting of two branches, one of which is a copy of DQN, namely, the Q branch. The other branch, which we call the preference branch, learns the action preference that the DQN implicitly follows. We theoretically prove that the policy improvement theorem holds for the preference-guided ϵ -greedy policy and experimentally show that the inferred action preference distribution aligns with the landscape of corresponding Q values. Intuitively, the preference-guided ϵ -greedy exploration motivates the DQN agent to take diverse actions, so that actions with larger Q values can be sampled more frequently, and those with smaller Q values still have a chance to be explored, thus encouraging the exploration. We comprehensively evaluate the proposed method by benchmarking it with well-known DQN variants in nine different environments. Extensive results confirm the superiority of our proposed method in terms of performance and convergence speed.