A combined pre- and intra-operative nomogram in evaluation of degrees of liver cirrhosis predicts post-hepatectomy liver failure: a multicenter prospective study.

Background: Adequate evaluation of degrees of liver cirrhosis is essential in surgical treatment of hepatocellular carcinoma (HCC) patients. The impact of the degrees of cirrhosis on prediction of post-hepatectomy liver failure (PHLF) remains poorly defined. This study aimed to construct and validate a combined pre- and intra-operative nomogram based on the degrees of cirrhosis in predicting PHLF in HCC patients using prospective multi-center's data. Methods: Consecutive HCC patients who underwent hepatectomy between May 18, 2019 and Dec 19, 2020 were enrolled at five tertiary hospitals. Preoperative cirrhotic severity scoring (CSS) and intra-operative direct liver stiffness measurement (DSM) were performed to correlate with the Laennec histopathological grading system. The performances of the pre-operative nomogram and combined pre- and intra-operative nomogram in predicting PHLF were compared with conventional predictive models of PHLF. Results: For 327 patients in this study, histopathological studies showed the rates of HCC patients with no, mild, moderate, and severe cirrhosis were 41.9%, 29.1%, 22.9%, and 6.1%, respectively. Either CSS or DSM was closely correlated with histopathological stages of cirrhosis. Thirty-three (10.1%) patients developed PHLF. The 30- and 90-day mortality rates were 0.9%. Multivariate regression analysis showed four pre-operative variables [HBV-DNA level, ICG-R15, prothrombin time (PT), and CSS], and one intra-operative variable (DSM) to be independent risk factors of PHLF. The pre-operative nomogram was constructed based on these four pre-operative variables together with total bilirubin. The combined pre- and intra-operative nomogram was constructed by adding the intra-operative DSM. The pre-operative nomogram was better than the conventional models in predicting PHLF. The prediction was further improved with the combined pre- and intra-operative nomogram. Conclusions: The combined pre- and intra-operative nomogram further improved prediction of PHLF when compared with the pre-operative nomogram. Trial Registration: Clinicaltrials.gov Identifier: NCT04076631.

Immune checkpoint inhibitor­associated diabetes mellitus in patients with HCC: Report of three cases and literature review.

Treatment with immune checkpoint inhibitors (ICIs) is steadily becoming the standard of care for hepatocellular carcinoma (HCC), with an increasing number of immune-related adverse events (irAEs). However, only a small number of reports on the occurrence of diabetes mellitus (DM) in patients with HCC treated with ICIs have been published. In the present study, the clinical manifestations, laboratory findings, treatment and prognosis of three patients with advanced HCC were reported, who suffered immune-related DM when receiving treatment with ICIs. Furthermore, the relevant literature was reviewed in order to summarize clinical manifestations, possible mechanisms, diagnosis, prognosis of rechallenge and recommended management options, as well as clinical treatment suggestions. ICI-induced diabetes is rare but irAEs are potentially fatal, as diabetic ketoacidosis (DKA) is often the first manifestation. The incidence of immune-related DM is 0.86% and among those cases, the incidence of DKA is 59%. The combination of two ICIs markedly increases the risk. The human leukocyte antigen genotype, islet autoantibodies and autoreactive T cell-mediated ß-cell destruction may be linked to the occurrence of immune-related DM. Patient education and clinicians' awareness of ICI-related DM are good management options. Adequate clinical judgment, close monitoring and early detection are also needed to decide whether to continue immunotherapy or to rechallenge it, so as to achieve the maximum benefit of clinical treatment.

Identification and molecular mapping of a major gene conferring resistance to Phytophthora sansomeana in soybean 'Colfax'.

KEY MESSAGE: The first single dominant resistance gene contributing major resistance to the oomycete pathogen Phytophthora sansomeana was identified and mapped from soybean 'Colfax'. Phytophthora root rot (PRR) is one of the most important diseases in soybean (Glycine max). PRR is well known to be caused by Phytophthora sojae, but recent studies showed that P. sansomeana also causes extensive root rot of soybean. Depending upon the isolate, it might produce aggressive symptoms, especially in seeds and seedlings. Unlike P. sojae which can be effectively managed by Rps genes, no known major resistance genes have yet been reported for P. sansomeana. Our previous study screened 470 soybean germplasm lines for resistance to P. sansomeana and found that soybean 'Colfax' (PI 573008) carries major resistance to the pathogen. In this study, we crossed 'Colfax' with a susceptible parent, 'Senaki', and developed three mapping populations with a total of 234 F2:3 families. Inheritance pattern analysis indicated a 1:2:1 ratio for resistant: segregating: susceptible lines among all the three populations, indicating a single dominant gene conferring the resistance in 'Colfax' (designated as Rpsan1). Linkage analysis using extreme phenotypes anchored Rpsan1 to a 30 Mb region on chromosome 3. By selecting nine polymorphic SNP markers within the region, Rpsan1 was genetically delimited into a 21.3 cM region between Gm03_4487138_A_C and Gm03_5451606_A_C, which corresponds to a 1.06 Mb genomic region containing nine NBS-LRR genes based on Gmax2.0 assembly. The mapping results were then validated using two breeding populations derived from 'E12076T-03' × 'Colfax' and 'E16099' × 'Colfax'. Marker-assisted resistance spectrum analyses with 9 additional isolates of P. sansomeana indicated that Rpsan1 may be effective towards a broader range of P. sansomeana isolates and has strong merit in protecting soybean to this pathogen in the future.

A computational model of circRNA-associated diseases based on a graph neural network: prediction and case studies for follow-up experimental validation.

BACKGROUND: Circular RNAs (circRNAs) have been confirmed to play a vital role in the occurrence and development of diseases. Exploring the relationship between circRNAs and diseases is of far-reaching significance for studying etiopathogenesis and treating diseases. To this end, based on the graph Markov neural network algorithm (GMNN) constructed in our previous work GMNN2CD, we further considered the multisource biological data that affects the association between circRNA and disease and developed an updated web server CircDA and based on the human hepatocellular carcinoma (HCC) tissue data to verify the prediction results of CircDA. RESULTS: CircDA is built on a Tumarkov-based deep learning framework. The algorithm regards biomolecules as nodes and the interactions between molecules as edges, reasonably abstracts multiomics data, and models them as a heterogeneous biomolecular association network, which can reflect the complex relationship between different biomolecules. Case studies using literature data from HCC, cervical, and gastric cancers demonstrate that the CircDA predictor can identify missing associations between known circRNAs and diseases, and using the quantitative real-time PCR (RT-qPCR) experiment of HCC in human tissue samples, it was found that five circRNAs were significantly differentially expressed, which proved that CircDA can predict diseases related to new circRNAs. CONCLUSIONS: This efficient computational prediction and case analysis with sufficient feedback allows us to identify circRNA-associated diseases and disease-associated circRNAs. Our work provides a method to predict circRNA-associated diseases and can provide guidance for the association of diseases with certain circRNAs. For ease of use, an online prediction server ( http://server.malab.cn/CircDA ) is provided, and the code is open-sourced ( https://github.com/nmt315320/CircDA.git ) for the convenience of algorithm improvement.

Targeting LINC01607 sensitizes hepatocellular carcinoma to Lenvatinib via suppressing mitophagy.

Lenvatinib is a standard therapy option for advanced hepatocellular carcinoma (HCC), but resistance limits clinical benefits. In this study, we identified inhibition of ROS levels and reduced redox status in Lenvatinib-resistant HCC. Integrating RNA-seq with unbiased whole-genome CRISPR-Cas9 screen analysis indicated LINC01607 regulated the P62 to enhance drug resistance by affecting mitophagy and antioxidant pathways. Underlying mechanisms were investigated both in vitro and in vivo. We initially confirmed that LINC01607, as a competing endogenous RNA (ceRNA) competing with mirRNA-892b, triggered protective mitophagy by upregulating P62, which reduced ROS levels and promoted drug resistance. Furthermore, LINC01607 was proved to resist oxidative stress by regulating the P62-Nrf2 axis, which transcriptionally regulated the expression of LINC01607 to form a positive feedback loop. Finally, silencing LINC01607 combined with Lenvatinib reversed resistance in animal and patient-derived organoid models. In conclusion, we proposed a novel mechanism of Lenvatinib resistance involving ROS homeostasis. This work contributed to understanding redox homeostasis-related drug resistance and provided new therapeutic targets and strategies for HCC patients.

MYC in liver cancer: mechanisms and targeted therapy opportunities.

MYC, a major oncogenic transcription factor, regulates target genes involved in various pathways such as cell proliferation, metabolism and immune evasion, playing a critical role in the tumor initiation and development in multiple types of cancer. In liver cancer, MYC and its signaling pathways undergo significant changes, exerting a profound impact on liver cancer progression, including tumor proliferation, metastasis, dedifferentiation, metabolism, immune microenvironment, and resistance to comprehensive therapies. This makes MYC an appealing target, despite it being previously considered an undruggable protein. In this review, we discuss the role and mechanisms of MYC in liver physiology, chronic liver diseases, hepatocarcinogenesis, and liver cancer progression, providing a theoretical basis for targeting MYC as an ideal therapeutic target for liver cancer. We also summarize and prospect the strategies for targeting MYC, including direct and indirect approaches to abolish the oncogenic function of MYC in liver cancer.

Evidence for an Interface of Hybrid Cocatalysts Favoring Photocatalytic Hydrogen Evolution Kinetics.

Hybrid cocatalysts have great application potential for improving the photocatalytic hydrogen evolution performance of semiconductors. The interfaces between components of hybrid cocatalysts make a great contribution to the improvement, but the associated mechanisms remain unclear. Herein, we prepared and tested three comparative CdS-based photocatalysts with NiS, NiS/Ni9S8, and Ni9S8 as the cocatalysts separately. The emphasis is placed on investigating the effect of the NiS/Ni9S8 interfaces on the photocatalytic hydrogen evolution performance of CdS. NiS/Ni9S8 exhibits a higher ability than NiS and Ni9S8 in making CdS a more active photocatalyst for water splitting. It shows that NiS, NiS/Ni9S8, and Ni9S8 perform similarly in terms of promoting the charge transfer and separation of CdS based on steady-state and time-resolved photoluminescence studies. At the same time, the linear sweep voltammetry and electrochemical impedance spectroscopy tests combined with the density functional theory calculations reveal that the component interfaces of NiS/Ni9S8 enable us to lower the water splitting activation energy, the charge-transfer resistance from the cocatalyst to sacrificial agent, and hydrogen adsorption Gibbs free energy. It is evidenced from this work that component interfaces of hybrid cocatalysts play a vital role in accelerating the dynamics of hydrogen evolution reactions.

Corrigendum: Machine learning prediction model for post- hepatectomy liver failure in hepatocellular carcinoma: a multicenter study.

[This corrects the article DOI: 10.3389/fonc.2022.986867.].

Identifying the Soybean microRNAs Related to Phytophthora sojae Based on RNA Sequencing and Bioinformatics Analysis.

Phytophthora root rot in soybeans is caused by a pathogen called Phytophthora sojae (P. sojae), which results in a significant decrease in soybean production within affected regions. MicroRNAs (miRNAs) are a class of small non-coding RNA molecules that play a key post-transcriptional regulatory role in eukaryotes. In this paper, the miRNAs that respond to P. sojae were analyzed from the gene level to complement the study of molecular resistance mechanisms in soybean. The study utilized high-throughput sequencing of soybean data to predict miRNAs that respond to P. sojae, analyze their specific functions, and verify regulatory relationships using qRT-PCR. The results showed that the miRNAs in soybean respond to P. sojae infection. MiRNAs can be transcribed independently, suggesting the presence of transcription factor binding sites in the promoter regions. Additionally, we performed an evolutionary analysis on conserved miRNAs that respond to P. sojae. Finally, we investigated the regulatory relationships among miRNAs, genes, and transcription factors, and identified five regulatory patterns. These findings lay the groundwork for future studies on the evolution of miRNAs responsive to P. sojae.

Functions of N6-methyladenosine in cancer metabolism: from mechanism to targeted therapy.

N6-methyladenosine (m6A) is the most abundant modification of eukaryotic mRNA and is involved in almost every stage of RNA metabolism. The m6A modification on RNA has been demonstrated to be a regulator of the occurrence and development of a substantial number of diseases, especially cancers. Increasing evidence has shown that metabolic reprogramming is a hallmark of cancer and is crucial for maintaining the homeostasis of malignant tumors. Cancer cells rely on altered metabolic pathways to support their growth, proliferation, invasion and metastasis in an extreme microenvironment. m6A regulates metabolic pathways mainly by either directly acting on metabolic enzymes and transporters or indirectly influencing metabolism-related molecules. This review discusses the functions of the m6A modification on RNAs, its role in cancer cell metabolic pathways, the possible underlying mechanisms of its effects and the implication of this modification in cancer therapy.

Collagen I-DDR1 signaling promotes hepatocellular carcinoma cell stemness via Hippo signaling repression.

Cancer stem cells (CSCs) are a minority population of cancer cells with stemness and multiple differentiation potentials, leading to cancer progression and therapeutic resistance. However, the concrete mechanism of CSCs in hepatocellular carcinoma (HCC) remains obscure. We found that in advanced HCC tissues, collagen I was upregulated, which is consistent with the expression of its receptor DDR1. Accordingly, high collagen I levels accompanied by high DDR1 expression are associated with poor prognoses in patients with HCC. Collagen I-induced DDR1 activation enhanced HCC cell stemness in vitro and in vivo. Mechanistically, DDR1 interacts with CD44, which acts as a co-receptor that amplifies collagen I-induced DDR1 signaling, and collagen I-DDR1 signaling antagonized Hippo signaling by facilitating the recruitment of PP2AA to MST1, leading to exaggerated YAP activation. The combined inhibition of DDR1 and YAP synergistically abrogated HCC cell stemness in vitro and tumorigenesis in vivo. A radiomic model based on T2 weighted images can noninvasively predict collagen I expression. These findings reveal the molecular basis of collagen I-DDR1 signaling inhibiting Hippo signaling and highlight the role of CD44/DDR1/YAP axis in promoting cancer cell stemness, suggesting that DDR1 and YAP may serve as novel prognostic biomarkers and therapeutic targets in HCC.

Targeting VPS72 inhibits ACTL6A/MYC axis activity in HCC progression.

BACKGROUND AND AIMS: HCC is a highly heterogeneous disease that is caused largely by genomic copy number variations. Herein, the mechanistic and therapeutically targeted role of vacuolar protein sorting 72 homologue (VPS72), a novel copy number variation cis-driven gained gene identified by genome-wide copy number variation and transcriptome analyses in HCC, is not well understood. APPROACH AND RESULTS: First, overexpression of VPS72 enhanced the initiation and progression of HCC in vitro and in vivo . Mechanistically, VPS72 interacted with the oncoproteins MYC and actin-like 6A (ACTL6A) and promoted the formation of the ACTL6A/MYC complex. Furthermore, ACTL6A regulated VPS72 protein stability by weakening the interaction between tripartite motif containing 21 (TRIM21) and VPS72. Thus, the interaction between VPS72 and ACTL6A enhanced the affinity of MYC for its target gene promoters and promoted their transcription, thereby contributing to HCC progression, which was inhibited by adeno-associated virus serotype 8 (AAV8)-mediated short hairpin RNA (shRNA) against VPS72. CONCLUSIONS: This study reveals the molecular mechanism of ACTL6A/VPS72/MYC in HCC, providing a theoretical basis and therapeutic target for this malignancy.

Combination Therapies for Advanced Biliary Tract Cancer.

Biliary tract cancers (BTCs) are a group of malignant neoplasms that have recently increased in incidence and have a poor prognosis. Surgery is the only curative therapy. However, most patients are only indicated for palliative therapy because of advanced-stage disease at diagnosis and rapid progression. The current first-line treatment for advanced BTC is gemcitabine and cisplatin chemotherapy. Nonetheless, many patients develop resistance to this regimen. Over the years, few chemotherapy regimens have managed to improve the overall survival of patients. Accordingly, novel therapies such as targeted therapy have been introduced to treat this patient population. Extensive research on tumorigenesis and the genetic profiling of BTC have revealed the heterogenicity and potential target pathways, such as EGFR, VEGF, MEK/ERK, PI3K and mTOR. Moreover, mutational analysis has documented the presence of IDH1, FGFR2, HER2, PRKACA, PRKACB, BRAF, and KRAS gene aberrations. The emergence of immunotherapy in recent years has expanded the treatment landscape for this group of malignancies. Cancer vaccines, adoptive cell transfer, and immune checkpoint inhibitors have been extensively investigated in trials of BTC. Therefore, patient stratification and a combination of various therapies have become a reasonable and important clinical strategy to improve patient outcomes. This review elaborates the literature on combined treatment strategies for advanced BTC from the past few years and ongoing clinical trials to provide new inspiration for the treatment of advanced BTC.

LINC01608 activated by YY1 facilitate hepatocellular carcinoma progression by modulating the EGFR/ERK axis.

BACKGROUND: Long non-coding RNAs (LncRNAs) have been demonstrated to associate with a variety of cancers. However, the mechanisms of LncRNAs in hepatocellular carcinoma (HCC) progression are still not fully clarified. METHODS: LINC01608 expression level in HCC and adjacent normal tissues was detected by real-time-quantitively PCR (RT-qPCR) in clinical samples and in situ hybridization (ISH) in tissue microarray. Several functional assays were performed to determine the biological effects of LINC01608 in HCC cells in vitro, while subcutaneous xenograft models and lung metastasis models in nude mice and immunohistochemistry (IHC) results showed the role of LINC01608 in HCC progression in vivo. The combination of LINC01608 with miR-875-5p and target genes was elucidated by dual-luciferase report assays, RNA immunoprecipitation (RIP) assays and fluorescence in situ hybridization (FISH) assays. Finally, bioinformatics analysis and chromatin immunoprecipitation (CHIP) were performed to investigate the mechanism of Yin Yang-1 (YY1) regulating LINC01608 transcription. RESULTS: LINC01608 was overexpressed in HCC tissues, and high LINC01608 expression predicted poor overall survival (OS) and disease-free survival (DFS) in HCC patients. LINC01608 could promote HCC cell proliferation, migration, invasion and epithelial-mesenchymal transition (EMT) in vitro and in vivo. Furthermore, we demonstrated that LINC01608 could sponge to miR-875-5p and activate the EGFR/ERK pathway. Moreover, we identified transcriptional factor YY1 could bind to the promoter of LINC01608 and induce its transcription. CONCLUSION: LINC01608 could serve as a promising prognostic biomarker of HCC. YY1-activated LINC01608 could promote HCC progression by associating with miR-875-5p to induce the EGFR/ERK signalling pathway. This discovery might provide therapeutic strategies for HCC.

Machine learning prediction model for post- hepatectomy liver failure in hepatocellular carcinoma: A multicenter study.

Introduction: Post-hepatectomy liver failure (PHLF) is one of the most serious complications and causes of death in patients with hepatocellular carcinoma (HCC) after hepatectomy. This study aimed to develop a novel machine learning (ML) model based on the light gradient boosting machines (LightGBM) algorithm for predicting PHLF. Methods: A total of 875 patients with HCC who underwent hepatectomy were randomized into a training cohort (n=612), a validation cohort (n=88), and a testing cohort (n=175). Shapley additive explanation (SHAP) was performed to determine the importance of individual variables. By combining these independent risk factors, an ML model for predicting PHLF was established. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value, negative predictive value, and decision curve analyses (DCA) were used to evaluate the accuracy of the ML model and compare it to that of other noninvasive models. Results: The AUCs of the ML model for predicting PHLF in the training cohort, validation cohort, and testing cohort were 0.944, 0.870, and 0.822, respectively. The ML model had a higher AUC for predicting PHLF than did other non-invasive models. The ML model for predicting PHLF was found to be more valuable than other noninvasive models. Conclusion: A novel ML model for the prediction of PHLF using common clinical parameters was constructed and validated. The novel ML model performed better than did existing noninvasive models for the prediction of PHLF.

A Novel Nomogram for Prediction of Post-Hepatectomy Liver Failure in Patients with Resectable Hepatocellular Carcinoma: A Multicenter Study.

Objective: To develop a nomogram for predicting post-hepatectomy liver failure (PHLF) in patients with resectable hepatocellular carcinoma (HCC) based on portal hypertension, the extent of resection, ALT, total bilirubin, and platelet count. Methods: Patients with HCC hospitalized from January 2015 to December 2020 were included in a retrospective cohort study. 595 HCC patients were divided into a training cohort (n=416) and a validation cohort (n=179) by random sampling. Univariate and multivariable analyses were performed to identify the independent variables to predict PHLF. The nomogram models for predicting the overall risk of PHLF and the risk of PHLF B+C were constructed based on the independent variables. Comparisons were made by using receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA) with traditional models, such as FIB-4 score, APRI score, CP class (Child-Pugh), MELD score (model of end-stage liver disease), and ALBI score (albumin-bilirubin) to analyze the accuracy and superiority of the nomogram. Results: We discovered that portal hypertension (yes vs no) (OR=1.677,95% CI:1.817-4.083, p=0.002), the extent of liver resection (OR=1.872,95% CI:3.937-47.096, p=0.001), ALT (OR=1.003,95% CI:1.003-1.016, P=0.003), total bilirubin (OR=1.036,95% CI:1.031-1.184, p=0.005), and platelet count (OR= 1.004, 95% CI:0.982-0.998, p=0.020) were independent risk factors for PHLF using multifactorial analysis. The nomogram models were constructed using well-fit calibration curves for each of these five covariates. When compared to the FIB4, ALBI, MELD, and CP score, our nomogram models have a better predictive value for predicting the overall risk of PHLF or the risk of PHLF B+C. The validation cohort's results were consistent. DCA also confirmed the conclusion. Conclusion: Our models, in the form of static nomogram or web application, were developed to predict PHLF overall risk and PHLF B+C risk in patients with HCC, with a high prediction sensitivity and specificity performance than other commonly used scoring systems.

Exploring Soybean Flower and Pod Variation Patterns During Reproductive Period Based on Fusion Deep Learning.

The soybean flower and the pod drop are important factors in soybean yield, and the use of computer vision techniques to obtain the phenotypes of flowers and pods in bulk, as well as in a quick and accurate manner, is a key aspect of the study of the soybean flower and pod drop rate (PDR). This paper compared a variety of deep learning algorithms for identifying and counting soybean flowers and pods, and found that the Faster R-CNN model had the best performance. Furthermore, the Faster R-CNN model was further improved and optimized based on the characteristics of soybean flowers and pods. The accuracy of the final model for identifying flowers and pods was increased to 94.36 and 91%, respectively. Afterward, a fusion model for soybean flower and pod recognition and counting was proposed based on the Faster R-CNN model, where the coefficient of determinationR2 between counts of soybean flowers and pods by the fusion model and manual counts reached 0.965 and 0.98, respectively. The above results show that the fusion model is a robust recognition and counting algorithm that can reduce labor intensity and improve efficiency. Its application will greatly facilitate the study of the variable patterns of soybean flowers and pods during the reproductive period. Finally, based on the fusion model, we explored the variable patterns of soybean flowers and pods during the reproductive period, the spatial distribution patterns of soybean flowers and pods, and soybean flower and pod drop patterns.

Noninvasively Assessed Portal Hypertension Grade Predicts Post-Hepatectomy Liver Failure in Patients With HepatocellCarcinoma: A Multicenter Study.

Purpose: To determine the predictive value of portal hypertension (PH) for the development of post-hepatectomy liver failure (PHLF) in patients with hepatocellular carcinoma (HCC). Patients and methods: This study enrolled a total of 659 patients with HCC that received hepatectomy as a first-line therapy. PH was classified as grade 0, 1, and 2 according to whether the indirect criteria for PH were met: 1) patients had obvious varicose veins and 2) splenomegaly was present and platelet count < 100 × 109/L. The effects of each variable on the occurrence of PHLF were assessed using univariate and multivariate analyses. Results: PH grade 2 (odds ratio [OR] = 2.222, p = 0.011), higher age (OR = 1.031, p = 0.003), hepatitis C infection (OR = 3.711, p = 0.012), open surgery (OR = 2.336, p < 0.001), portal flow blockage (OR = 1.626, p = 0.023), major hepatectomy (OR = 2.919, p = 0.001), hyperbilirubinemia (≥ 17.2 µmol/L, OR = 2.113, p = 0.002), and high levels of alpha-fetoprotein (> 400n g/ml, OR = 1.799, p = 0.008) were significantly associated with PHLF occurrence. We performed a subgroup analysis of liver resection and found that the extent of liver resection and PH grade were good at distinguishing patients at high risk for PHLF, and we developed an easy-to-view roadmap. Conclusion: PH is significantly related to the occurrence of PHLF in patients who underwent hepatectomy. Noninvasively assessing PH grade can predict PHLF risk.

Automatic and Accurate Acquisition of Stem-Related Phenotypes of Mature Soybean Based on Deep Learning and Directed Search Algorithms.

The stem-related phenotype of mature stage soybean is important in soybean material selection. How to improve on traditional manual methods and obtain the stem-related phenotype of soybean more quickly and accurately is a problem faced by producers. With the development of smart agriculture, many scientists have explored soybean phenotypes and proposed new acquisition methods, but soybean mature stem-related phenotype studies are relatively scarce. In this study, we used a deep learning method within the convolutional neural network to detect mature soybean stem nodes and identified soybean structural features through a novel directed search algorithm. We subsequently obtained the pitch number, internodal length, branch number, branching angle, plant type spatial conformation, plant height, main stem length, and new phenotype-stem curvature. After 300 epochs, we compared the recognition results of various detection algorithms to select the best. Among them, YOLOX had a maximum average accuracy (mAP) of 94.36% for soybean stem nodes and scale markers. Through comparison of the phenotypic information extracted by the directed search algorithm with the manual measurement results, we obtained the Pearson correlation coefficients, R, of plant height, pitch number, internodal length, main stem length, stem curvature, and branching angle, which were 0.9904, 0.9853, 0.9861, 0.9925, 0.9084, and 0.9391, respectively. These results show that our algorithm can be used for robust measurements and counting of soybean phenotype information, which can reduce labor intensity, improve efficiency, and accelerate soybean breeding.