Plant disease prescription recommendation based on electronic medical records and sentence embedding retrieval.

BACKGROUND: In the era of Agri 4.0 and the popularity of Plantwise systems, the availability of Plant Electronic Medical Records has provided opportunities to extract valuable disease information and treatment knowledge. However, developing an effective prescription recommendation method based on these records presents unique challenges, such as inadequate labeling data, lack of structural and linguistic specifications, incorporation of new prescriptions, and consideration of multiple factors in practical situations. RESULTS: This study proposes a plant disease prescription recommendation method called PRSER, which is based on sentence embedding retrieval. The semantic matching model is created using a pre-trained language model and a sentence embedding method with contrast learning ideas, and the constructed prescription reference database is retrieved for optimal prescription recommendations. A multi-vegetable disease dataset and a multi-fruit disease dataset are constructed to compare three pre-trained language models, four pooling types, and two loss functions. The PRSER model achieves the best semantic matching performance by combining MacBERT, CoSENT, and CLS pooling, resulting in a Pearson coefficient of 86.34% and a Spearman coefficient of 77.67%. The prescription recommendation capability of the model is also verified. PRSER performs well in closed-set testing with Top-1/Top-3/Top-5 accuracy of 88.20%/96.07%/97.70%; and slightly worse in open-set testing with Top-1/Top-3/Top-5 accuracy of 82.04%/91.50%/94.90%. Finally, a plant disease prescription recommendation system for mobile terminals is constructed and its generalization ability with incomplete inputs is verified. When only symptom information is available without environment and plant information, our model shows slightly lower accuracy with Top-1/Top-3/Top-5 accuracy of 75.24%/88.35%/91.99% in closed-set testing and Top-1/Top-3/Top-5 accuracy of 75.08%/87.54%/89.84% in open-set testing. CONCLUSIONS: The experiments validate the effectiveness and generalization ability of the proposed approach for recommending plant disease prescriptions. This research has significant potential to facilitate the implementation of artificial intelligence in plant disease treatment, addressing the needs of farmers and advancing scientific plant disease management.

A tomato disease identification method based on leaf image automatic labeling algorithm and improved YOLOv5 model.

BACKGROUND: Tomato is one of the most important vegetables in the world. Timely and accurate identification of tomato disease is a critical way to ensure the quality and yield of tomato production. The convolutional neural network is a crucial means of disease identification. However, this method requires manual annotation of a large amount of image data, which wastes the human cost of scientific research. RESULTS: To simplify the process of disease image labeling and improve the accuracy of tomato disease recognition and the balance of various disease recognition effects, a BC-YOLOv5 tomato disease recognition method is proposed to identify healthy growth and nine types of diseased tomato leaves. In the present study, the YOLOv5 model is improved by designing an automatic tomato leaf image labeling algorithm, using the weighted bi-directional feature pyramid network to change the Neck structure, adding the convolution block attention module, and changing the input channel of the detection layer. Experiments show that the BC-YOLOv5 method has an excellent image annotation effect on tomato leaves, with a pass rate exceeding 95%. Furthermore, compared with existing models, the performance indices of BC-YOLOv5 to identify tomato diseases are the best. CONCLUSION: BC-YOLOv5 realizes the automatic labeling of tomato leaf images before the start of training. This method not only identifies nine common tomato diseases, but also improve the accuracy of disease identification and have a more balanced identification effect on various diseases. It provides a reliable method for the identification of tomato disease. © 2023 Society of Chemical Industry.

Tumor-infiltrating CD36+CD8+T cells determine exhausted tumor microenvironment and correlate with inferior response to chemotherapy in non-small cell lung cancer.

BACKGROUND: The scavenger receptor CD36 was reported to be highly expressed on tumor-infiltrating CD8+ T cells, but the clinical role remains obscure. This study aims to explore the infiltration and clinical value of CD36+CD8+ T cells in NSCLC. METHODS: Immunohistochemistry and immunofluorescence were conducted for survival analyses and immunological evaluation in 232 NSCLC patients in Zhongshan Hospital. Flow cytometry analyses were carried out to assess the immune cells from fresh tumor samples, non-tumor tissues and peripheral blood. In vitro tumor infiltrating lymphocytes cultures were conducted to test the effect of CD36 blockage. RESULTS: Accumulation of CD36+CD8+ T cells in tumor tissues was correlated with more advanced stage (p < 0.001), larger tumor size (p < 0.01), and lymph node metastasis (p < 0.0001) in NSCLC. Moreover, high infiltration of CD36+CD8+ T cells indicated poor prognosis in terms of both overall survival (OS) and recurrence-free survival (RFS) and inferior chemotherapy response. CD36+CD8+ T cells showed decreased GZMB (p < 0.0001) and IFN-γ (p < 0.001) with elevated PD-1 (p < 0.0001) and TIGIT (p < 0.0001). Analysis of tumor-infiltrating immune cell landscape revealed a positive correlation between CD36+CD8+ T cells and Tregs (p < 0.01) and M2-polarized macrophages (p < 0.01) but a negative correlation with Th1 (p < 0.05). Notably, inhibition of CD36 partially restored the cytotoxic function of CD8+ T cells by producing more GZMB and IFN-γ. CONCLUSION: CD36+CD8+ T cells exhibit impaired immune function and high infiltration of CD36+CD8+ T cells indicated poor prognosis and inferior chemotherapy response in NSCLC patients. CD36 could be a therapeutic target in combination with chemotherapy in NSCLC patients.

The Gray Mold Spore Detection of Cucumber Based on Microscopic Image and Deep Learning.

Rapid and accurate detection of pathogen spores is an important step to achieve early diagnosis of diseases in precision agriculture. Traditional detection methods are time-consuming, laborious, and subjective, and image processing methods mainly rely on manually designed features that are difficult to cope with pathogen spore detection in complex scenes. Therefore, an MG-YOLO detection algorithm (Multi-head self-attention and Ghost-optimized YOLO) is proposed to detect gray mold spores rapidly. Firstly, Multi-head self-attention is introduced in the backbone to capture the global information of the pathogen spores. Secondly, we combine weighted Bidirectional Feature Pyramid Network (BiFPN) to fuse multiscale features of different layers. Then, a lightweight network is used to construct GhostCSP to optimize the neck part. Cucumber gray mold spores are used as the study object. The experimental results show that the improved MG-YOLO model achieves an accuracy of 0.983 for detecting gray mold spores and takes 0.009 s per image, which is significantly better than the state-of-the-art model. The visualization of the detection results shows that MG-YOLO effectively solves the detection of spores in blurred, small targets, multimorphology, and high-density scenes. Meanwhile, compared with the YOLOv5 model, the detection accuracy of the improved model is improved by 6.8%. It can meet the demand for high-precision detection of spores and provides a novel method to enhance the objectivity of pathogen spore detection.

TTI1 promotes non-small-cell lung cancer progression by regulating the mTOR signaling pathway.

The role of TELO2-interacting protein 1 (TTI1) in the progression of several types of cancer has been reported recently. The aim of this study was to estimate the expression and potential value of TTI1 in non-small-cell lung cancer (NSCLC) patients. The expression of TTI1 and its prognostic value in NSCLC from The Cancer Genome Atlas (TCGA) database and Gene Expression Omnibus (GEO) database were analyzed. To verify the bioinformatics findings, a tissue microarray containing 160 NSCLC and paired peritumoral tissues from NSCLC patients was analyzed by immunohistochemistry for TTI1. Subsequently, the roles of TTI1 in NSCLC cells were investigated in vivo by establishing xenograft models in nude mice and in vitro by transwell, CCK-8, wound healing, and colony formation assays. In addition, quantitative real-time polymerase chain reaction and western blot were applied to explore the underlying mechanism by which TTI1 promotes tumor progression. Finally, the relationship between TTI1 and Ki67 expression level in NSCLC was probed, and Kaplan-Meier and Cox analyses were performed to assess the prognostic merit of TTI1 and Ki67 in NSCLC patients. We found that the expression of TTI1 was significantly upregulated in NSCLC tissues compared to paired peritumoral tissues, which coincides with the bioinformatics findings from the TCGA and GEO databases. TTI1 was highly expressed in NSCLC patients with large tumors, advanced tumor stage, and lymphatic metastasis. In addition, the prognostic analysis identified TTI1 as an independent indication for poor prognosis of NSCLC patients. In vitro, upregulation of TTI1 in NSCLC cells could facilitate cell invasion, metastasis, viability, and proliferation. Mechanistically, our study verified that TTI1 could regulate mTOR activity, which has a pivotal role in human cancer. Consistently, the expressions of TTI1 and Ki67 had a positive relationship in NSCLC cells and tissues. Notably, patients with overexpression of TTI1 or Ki67 had a shorter overall survival rate and a higher disease-free survival rate compared to patients with low expression of TTI1 or Ki67, and the combination of TTI1 and Ki67 was an independent parameter predicting the prognosis and recurrence of NSCLC patients. We conclude that TTI1 promotes NSCLC cell proliferation, metastasis, and invasion by regulating mTOR activity, and the combination of TTI1 and Ki67 is a valuable molecular biomarker for the survival and recurrence of NSCLC patients.

Exosomal circZNF451 restrains anti-PD1 treatment in lung adenocarcinoma via polarizing macrophages by complexing with TRIM56 and FXR1.

BACKGROUND: Although success was achieved in the therapy for a minority of advanced lung adenocarcinoma (LUAD) patients, anti-programmed death 1 (PD1) resistance was found in most LUAD patients. Here, we aimed to uncover a potential role of exosomal circular RNAs (circRNAs) in LUAD refractory to PD1 blockade.  METHODS: circRNA sequencing and qRT-PCR were performed to determine the level of exosomal circRNAs in LUAD patients subsequently treated with anti-PD1. Then, the RNA pulldown, RNA immunoprecipitation, mass spectrometry, chromatin immunoprecipitation, luciferase reporter assays, flow cytometry, RNA sequencing, and in vitro and in vivo models were used to uncover the biological functions and underlying mechanism of circZNF451 in LUAD anti-PD1 treatment resistance. RESULTS: circRNA sequencing and qRT-PCR identified the up-regulation of exosomal circZNF451 from LUAD patients with progressive disease (PD) compared to those with partial remission (PR) after PD1 blockade therapy. Furthermore, elevated circZNF451 was revealed to be associated with poor prognosis of LUAD patients. Additionally, exosomal circZNF451 was demonstrated to induce an anti-inflammatory phenotype in macrophages and exhaustion of cytotoxic CD8+ T cells, and enhanced TRIM56-mediated degradation of FXR1 to activate the ELF4-IRF4 pathway in macrophages. By transgenic mice, knockout of ELF4 in macrophages was found to rescue immunotherapy efficacy in tumors with high level of exosomal circZNF451. CONCLUSION: Exosomal circZNF451 reshapes the tumor immune microenvironment by inducing macrophages polarization via the FXR1- ELF4-IRF4 axis and is a novel biomarker for predicting the sensitivity of PD1 blockade in LUAD.

CDK14 inhibition reduces mammary stem cell activity and suppresses triple negative breast cancer progression.

The Wnt/ß-catenin signaling pathway plays an important role in regulating mammary organogenesis and oncogenesis. However, therapeutic methods targeting the Wnt pathway against breast cancer have been limited. To address this challenge, we investigate the function of cyclin-dependent kinase 14 (CDK14), a member of the Wnt signaling pathway, in mammary development and breast cancer progression. We show that CDK14 is expressed in the mammary basal layer and elevated in triple negative breast cancer (TNBC). CDK14 knockdown reduces the colony-formation ability and regeneration capacity of mammary basal cells and inhibits the progression of murine MMTV-Wnt-1 basal-like mammary tumor. CDK14 knockdown or pharmacological inhibition by FMF-04-159-2 suppresses the progression and metastasis of TNBC. Mechanistically, CDK14 inhibition inhibits mammary regeneration and TNBC progression by attenuating Wnt/ß-catenin signaling. These findings highlight the significance of CDK14 in mammary development and TNBC progression, shedding light on CDK14 as a promising therapeutic target for TNBC.

Attention-optimized DeepLab V3 + for automatic estimation of cucumber disease severity.

BACKGROUND: Automatic and accurate estimation of disease severity is critical for disease management and yield loss prediction. Conventional disease severity estimation is performed using images with simple backgrounds, which is limited in practical applications. Thus, there is an urgent need to develop a method for estimating the disease severity of plants based on leaf images captured in field conditions, which is very challenging since the intensity of sunlight is constantly changing, and the image background is complicated. RESULTS: This study developed a simple and accurate image-based disease severity estimation method using an optimized neural network. A hybrid attention and transfer learning optimized semantic segmentation model was proposed to obtain the disease segmentation map. The severity was calculated by the ratio of lesion pixels to leaf pixels. The proposed method was validated using cucumber downy mildew, and powdery mildew leaves collected under natural conditions. The results showed that hybrid attention with the interaction of spatial attention and channel attention can extract fine lesion and leaf features, and transfer learning can further improve the segmentation accuracy of the model. The proposed method can accurately segment healthy leaves and lesions (MIoU = 81.23%, FWIoU = 91.89%). In addition, the severity of cucumber leaf disease was accurately estimated (R2 = 0.9578, RMSE = 1.1385). Moreover, the proposed model was compared with six different backbones and four semantic segmentation models. The results show that the proposed model outperforms the compared models under complex conditions, and can refine lesion segmentation and accurately estimate the disease severity. CONCLUSIONS: The proposed method was an efficient tool for disease severity estimation in field conditions. This study can facilitate the implementation of artificial intelligence for rapid disease severity estimation and control in agriculture.

Elevated circASCC3 limits antitumor immunity by sponging miR-432-5p to upregulate C5a in non-small cell lung cancer.

Although anti-programmed cell death 1 (PD1) treatment has become a first-line therapy for advanced non-small cell lung cancer (NSCLC), most NSCLC patients are refractory to anti-PD1. Here, we aimed to investigate the mechanism of dysregulated circular RNAs (circRNAs) related to anti-PD1 resistance in NSCLC. The expression of circASCC3 (hsa_circ_0077,495) in NSCLC tissues and cell lines was evaluated by fluorescence in situ hybridization and quantitative reverse transcription-polymerase chain reaction. The functions and mechanisms of circASCC3 in NSCLC progression and anti-PD1 resistance were uncovered in vitro and in vivo. The circASCC3 level was upregulated in NSCLC compared with that in paired normal tissues. Specifically, circASCC3 expression was higher in tissues from NSCLC patients with anti-PD1 refractory than in those from patients who sensitive to anti-PD1. Overexpression of circASCC3 enhanced the malignant phenotype of NSCLC cells and led to an immunosuppressive microenvironment. Mechanistically, circASCC3 sponged miR-432-5p to increase complement C5a levels, which enhanced the progression and dysfunctional immune status of NSCLC. Thus, circASCC3 overexpression reshapes the tumor microenvironment by impacting the complement system in NSCLC and provides a potential strategy to overcome anti-PD1 resistance.

The circular RNA circHMGB2 drives immunosuppression and anti-PD-1 resistance in lung adenocarcinomas and squamous cell carcinomas via the miR-181a-5p/CARM1 axis.

BACKGROUND: Previous studies have confirmed the oncogenic role of HMGB2 in various cancers, but the biological functions of HMGB2-derived circRNAs remain unknown. Thus, we intended to investigate the potential role of HMGB2-derived circRNAs in lung adenocarcinomas (LUAD) and squamous cell carcinomas (LUSC). METHODS: The expression profiles of HMGB2-derived circRNAs in LUAD and LUSC tissues and matched normal tissues were assessed using qRT-PCR. The role of circHMGB2 in the progression of the LUAD and LUSC was determined in vitro by Transwell, CCK-8, flow cytometry and immunohistochemistry assays, as well as in vivo in an immunocompetent mouse model and a humanized mouse model. In addition, in vivo circRNA precipitation assays, luciferase reporter assays and RNA pulldown assays were performed to explore the underlying mechanism by which circHMGB2 promotes anti-PD-1 resistance in the LUAD and LUSC. RESULTS: The expression of circHMGB2 (hsa_circ_0071452) was significantly upregulated in NSCLC tissues, and survival analysis identified circHMGB2 as an independent indicator of poor prognosis in the LUAD and LUSC patients. We found that circHMGB2 exerted a mild effect on the proliferation of the LUAD and LUSC cells, but circHMGB2 substantially reshaped the tumor microenvironment by contributing to the exhaustion of antitumor immunity in an immunocompetent mouse model and a humanized mouse model. Mechanistically, circHMGB2 relieves the inhibition of downstream CARM1 by sponging miR-181a-5p, thus inactivating the type 1 interferon response in the LUAD and LUSC. Moreover, we found that the upregulation of circHMGB2 expression decreased the efficacy of anti-PD-1 therapy, and we revealed that the combination of the CARM1 inhibitor EZM2302 and an anti-PD-1 antibody exerted promising synergistic effects in a preclinical model. CONCLUSION: circHMGB2 overexpression promotes the LUAD and LUSC progression mainly by reshaping the tumor microenvironment and regulating anti-PD-1 resistance in the LUAD and LUSC patients. This study provides a new strategy for the LUAD and LUSC treatment.

CDK16 promotes the progression and metastasis of triple-negative breast cancer by phosphorylating PRC1.

BACKGROUND: Cyclin-dependent kinase 16 (CDK16) is an atypical PCTAIRE kinase, and its activity is dependent on the Cyclin Y (CCNY) family. Ccnys have been reported to regulate mammary stem cell activity and mammary gland development, and CCNY has been recognized as an oncoprotein in various cancers, including breast cancer. However, it remains unclear whether CDK16 has a role in breast cancer and whether it can be used as a therapeutic target for breast cancer. METHODS: Publicly available breast cancer datasets analyses and Kaplan-Meier survival analyses were performed to reveal the expression and clinical relevance of atypical CDKs in breast cancer. CDK16 protein expression was further examined by immunohistochemical and immunoblot analyses of clinical samples. Cell proliferation was measured by colony formation and MTT analyses. Cell cycle and apoptosis were examined by fluorescence-activated cell sorting (FACS) analysis. Wound-healing and trans-well invasion assays were conducted to test cell migration ability. The functions of CDK16 on tumorigenesis and metastasis were evaluated by cell line-derived xenograft, patient-derived organoid/xenograft, lung metastasis and systemic metastasis mouse models. Transcriptomic analysis was performed to reveal the potential molecular mechanisms involved in the function of CDK16. Pharmacological inhibition of CDK16 was achieved by the small molecular inhibitor rebastinib to further assess the anti-tumor utility of targeting CDK16. RESULTS: CDK16 is highly expressed in breast cancer, particularly in triple-negative breast cancer (TNBC). The elevated CDK16 expression is correlated with poor outcomes in breast cancer patients. CDK16 can improve the proliferation and migration ability of TNBC cells in vitro, and promote tumor growth and metastasis of TNBC in vivo. Both genetic knockdown and pharmacological inhibition of CDK16 significantly suppress the tumor progression of TNBC. Mechanistically, CDK16 exerts its function by phosphorylating protein regulator of cytokinesis 1 (PRC1) to regulate spindle formation during mitosis. CONCLUSION: CDK16 plays a critical role in TNBC and is a novel promising therapeutic target for TNBC.

Modeling of Farmers' Vegetable Safety Production Based on Identification of Key Risk Factors From Beijing, China.

Food safety emphasizes risk control in the production process, and has attracted much attention from food regulators and consumers in recent years. The objectives of this study were to conduct early key risk factors identification and risk modeling for vegetable safety production. To achieve these objectives, this article quantitatively identified the key direct and indirect risk factors in vegetable safety production through questionnaire surveys and a multivariate linear model, and modeled the effects of key risk factors affecting vegetable safety production based on the catastrophe progression method. Based on 973 valid farmers' questionnaires from Beijing, China, the results showed that key direct risk factors are production violation, farmland biological control, pesticide and fertilizer use criteria, and agricultural consumable handling; key indirect risk factors included cooperative participation, planting years, prohibited pesticide knowledge, production recording, and product type. Through the empirical analysis, it can be seen that there are regional differences in the production risk of vegetable farmers in Beijing. The production risks of Changping, Huairou, and Shunyi are the most serious; from a city-wide perspective, the risk of farmland biological control is greatest, followed by risk aversion ability. The findings of this research have important implications for safe vegetable production and farmers' production risk control.

CDK inhibitors in cancer therapy, an overview of recent development.

Dysregulated cell division, which leads to aberrant cell proliferation, is one of the key hallmarks of cancer. Therefore, therapeutic targets that block cell division would be effective for cancer treatment. Cell division is mainly controlled by a complex composed of cyclin and cyclin dependent kinases (CDKs). To date, the CDK inhibitors (CDKIs), specifically the ones that block the enzyme activity of CDK4 and CDK6 (CDK4/6), have been approved by FDA for the treatment of metastatic hormone receptor positive breast cancer. However, due to the non-selectivity and significant toxicity, most of the first generation CDK inhibitors (so called pan-CDK inhibitors that target several CDKs), have not been approved for clinical application. Despite this, great efforts and progress have been made to enable pan-CDK inhibitors application in the clinical setting. Notably, the development of combination therapy strategies in recent years has made it possible to reduce the toxicity and side effects of pan-CDK inhibitors. Thus, as a combination therapy approach, pan-CDK inhibitors regain great potential in clinical application. In this review, we introduced the CDK family members and discussed their major functions in cell cycle controlling. Then, we summarized the research progress regarding CDK inhibitors, especially those other than CDK4/6 inhibitors. We reviewed first-generation pan-CDKIs Flavopiridol and Roscovitine, and second-generation CDKIs Dinaciclib, P276-00, AT7519, TG02, Roniciclib, RGB-286638 by focusing on their developing stages, clinical trials and targeting cancers. The specific CDKIs, which targets to increase specificity and decrease the side effects, were also discussed. These CDKIs include CDK4/6, CDK7, CDK9, and CDK12/13 inhibitors. Finally, the efficacy and discrepancy of combination therapy with CDK inhibitors and PD1/PDL1 antibodies were analyzed, which might give insights into the development of promising strategy for cancer treatment.

Chinese tourists in Nordic countries: An analysis of spatio-temporal behavior using geo-located travel blog data.

Geo-located travel blogs, a new data source, enable to achieve more detailed analysis of tourists' spatio-temporal behavior. Taking Chinese tourists in Nordic countries as the research object, this paper focuses on their behavior, seasonal patterns and complex network effects by using geo-located travel blog data collected from Qunar.com. The results show that: (1) Chinese tourists visiting Nordic countries are often experienced in traveling. The local climate during the cold season does not prevent them from pursuing the aurora scenery. (2) The travel behavior of Chinese tourists is spatially heterogeneous. The network analysis reveals that Iceland showcases stronger, compared to the other Nordic countries, community independence and small world effect. (3) During the warm season, Chinese tourists choose a variety of destinations, while in cold season, they tend to choose destinations with higher chances for spotting the northern lights. These results provide helpful information for the tourism management departments of Nordic countries to improve their marketing and development efforts directed for Chinese tourists.

Growth monitoring of greenhouse lettuce based on a convolutional neural network.

Growth-related traits, such as aboveground biomass and leaf area, are critical indicators to characterize the growth of greenhouse lettuce. Currently, nondestructive methods for estimating growth-related traits are subject to limitations in that the methods are susceptible to noise and heavily rely on manually designed features. In this study, a method for monitoring the growth of greenhouse lettuce was proposed by using digital images and a convolutional neural network (CNN). Taking lettuce images as the input, a CNN model was trained to learn the relationship between images and the corresponding growth-related traits, i.e., leaf fresh weight (LFW), leaf dry weight (LDW), and leaf area (LA). To compare the results of the CNN model, widely adopted methods were also used. The results showed that the values estimated by CNN had good agreement with the actual measurements, with R2 values of 0.8938, 0.8910, and 0.9156 and normalized root mean square error (NRMSE) values of 26.00, 22.07, and 19.94%, outperforming the compared methods for all three growth-related traits. The obtained results showed that the CNN demonstrated superior estimation performance for the flat-type cultivars of Flandria and Tiberius compared with the curled-type cultivar of Locarno. Generalization tests were conducted by using images of Tiberius from another growing season. The results showed that the CNN was still capable of achieving accurate estimation of the growth-related traits, with R2 values of 0.9277, 0.9126, and 0.9251 and NRMSE values of 22.96, 37.29, and 27.60%. The results indicated that a CNN with digital images is a robust tool for the monitoring of the growth of greenhouse lettuce.