Deep learning-guided adjuvant chemotherapy selection for elderly patients with breast cancer.

PURPOSE: The efficacy of adjuvant chemotherapy in elderly breast cancer patients is currently controversial. This study aims to provide personalized adjuvant chemotherapy recommendations using deep learning (DL). METHODS: Six models with various causal inference approaches were trained to make individualized chemotherapy recommendations. Patients who received actual treatment recommended by DL models were compared with those who did not. Inverse probability treatment weighting (IPTW) was used to reduce bias. Linear regression, IPTW-adjusted risk difference (RD), and SurvSHAP(t) were used to interpret the best model. RESULTS: A total of 5352 elderly breast cancer patients were included. The median (interquartile range) follow-up time was 52 (30-80) months. Among all models, the balanced individual treatment effect for survival data (BITES) performed best. Treatment according to following BITES recommendations was associated with survival benefit, with a multivariate hazard ratio (HR) of 0.78 (95% confidence interval (CI): 0.64-0.94), IPTW-adjusted HR of 0.74 (95% CI: 0.59-0.93), RD of 12.40% (95% CI: 8.01-16.90%), IPTW-adjusted RD of 11.50% (95% CI: 7.16-15.80%), difference in restricted mean survival time (dRMST) of 12.44 (95% CI: 8.28-16.60) months, IPTW-adjusted dRMST of 7.81 (95% CI: 2.93-11.93) months, and p value of the IPTW-adjusted Log-rank test of 0.033. By interpreting BITES, the debiased impact of patient characteristics on adjuvant chemotherapy was quantified, which mainly included breast cancer subtype, tumor size, number of positive lymph nodes, TNM stages, histological grades, and surgical type. CONCLUSION: Our results emphasize the potential of DL models in guiding adjuvant chemotherapy decisions for elderly breast cancer patients.

Clinical nutrition service capacity of 445 secondary or above hospitals in Sichuan, China: the 2021 annual survey.

BACKGROUND AND OBJECTIVES: To investigate the capacity of clinical nutrition services in secondary and tertiary hospitals in the Sichuan Province, China. METHODS AND STUDY DESIGN: Convenience sampling was used. E-questionnaires were distributed to all eligible medical institutions in Sichuan through the official network of provincial and municipal clinical nutrition quality control centers. The data obtained were sorted in Microsoft Excel and analyzed by SPSS. RESULTS: A total of 519 questionnaires were returned, of which 455 were valid. Only 228 hospitals were accessible to clinical nutrition services, of which 127 hospitals had independently set up clinical nutrition departments (CNDs). The ratio of clinical nutritionists to beds was 1:214. During the last decade, the rate of constructing new CNDs was maintained at approximately 5 units/year. A total of 72.4% of hospitals managed their clinical nutrition units as part of their medical technology departments. The specialist number ratio of senior, associate, intermediate and junior is approximately 1:4:8:10. There were 5 common charges for clinical nutrition. CONCLUSIONS: The sample representation was limited, and the capacity of clinical nutrition services may have been overestimated. Secondary and tertiary hospitals in Sichuan are currently in the second high tide of department establishment, with a positive trend of departmental affiliation standardization and a basic formation of a talent echelon.

First Report of Pantoea stewartii subsp. stewartii Causing Bacterial Leaf Wilt of Sugarcane in China.

The pathogen of Pantoea stewartii subsp. stewartii (Pss) that is the causal agent of Stewart's bacterial wilt of corn also infects numerous experimental hosts of graminaceous plants (Pepper et al., 1967; Wang et al., 2012). However, little is known about this pathogen naturally infecting sugarcane. In 2017, we observed some sugarcane cultivars showing leaf blade bleaching at the disease initiation stage, which further resulted in development of blight and necrotic lesions (Figure 1-A and -B) in Zhanjiang, Guangdong province of China. To diagnose this putative disease, five symptomatic leaf samples were collected from different sugarcane cultivars. The Pss was found to infect these samples using the nested PCR with Pss-specific outer primers PS1/PS4 and inner primers Ps2r/Ps3r that targeted at the 16S rRNA gene of this pathogen (Wang et al., 2009). The expected 262-bp fragments from positive samples were amplified, cloned, and sequenced (GenBank accession no. MW015795-MW015799). BLASTN analysis revealed that these isolates had more than 99.5% nucleotide identify (222 bp out of 262 bp) with each other and with Pss strains (ATCC 8199 and DC283) as well as P. stewartii subsp. indologenes strains (SR2-12 and LMG 2632) after sequences were trimmed at the 5'- and 3'-terminal of inner primer sequences. In addition, these leaf samples were surface-sterilized with 75% alcohol followed by macerated and chopped in sterile water. Upon plating on solid NA medium at 28 °C for 24-36 h, the bacterial colonies exhibited yellow color with circular, convex, smooth and translucent edges (Figure 1-C). Straight rods and non-encapsulated cells were detected under transmission electron microscopy (Figure 1-D). Moreover, an identical colony termed as PSCN1 was isolated from sugarcane cultivar YZ08-1095 and was further confirmed by the PCR with a universal primer pair 63F (5'-CAGGCCTAACACATGCAAGTC-3') and 1387R (5'-GGGCGGWGTGTACAAGGC-3') that targeted at bacterial 16S rRNA gene (Marchesi et al., 1998). A 1362-bp DNA fragment sequence was obtained from PSCN1 strain and deposited on GenBank library (accession no. MW015767). Sequence analysis showed that PSCN1 shared 99.9-100% nucleotide identity (1315 bp out of 1362 bp) with the two reference strains of Pss (ATCC 8199 and DC283) after sequences were trimmed at the 5'- and 3'-terminal of primer sequences. According to Koch's postulates, pathogenicity test was carried out on YZ08-1095 plants with 3-5 fully developed leaf inoculated with the suspended cells (108 cells/ml) of PSCN1 strain by cutting the one-third of leaves before spraying with a suspension. Control plants were mock-inoculated with serial liquid nutrition agar medium. Two independent experiments were performed for pathogenicity assay and more than 28 plants of YZ08-1095 were used in each treatment. Plants were cultured in a growth chamber at 28 °C and 60% humidity under a 16 h light/8 h dark photoperiod. Leaves inoculated by the PSCN1 initially showed bleached, blight and wilting symptoms on leaf edges at seven days post-inoculation (dpi) (Figure 1-E and -F), which were similar to those symptoms observed in the fields. Control plants remained asymptomatic (Figure 1-G). The average incidence of diseased plant was 51.9% at 21 dpi. The bacteria were subsequently re-isolated from diseased leaves, and yielded colonies were completely identical to the PSCN1. Taken together, our data provides the valuable information for diagnosis and controlling this disease in sugarcane.

Enhancing the production of poly-γ-glutamate in Bacillus subtilis ZJS18 by the heat- and osmotic shock and its mechanism.

Poly-γ-glutamate (γ-PGA) is a natural macromolecule peptide, and is widely used in the food, medicine, and pharmaceutical industries. In this study, heat- and osmotic shock were used to improve the production of γ-PGA in Bacillus subtilis ZJS18, and its molecular mechanism was explored. The results indicated that the heat- and osmotic shock significantly promoted the production of γ-PGA owing to the stress response of B. subtilis cells to adverse environment. The highest concentrations of γ-PGA reached 14.53 and 15.98 g/l under heat- and osmotic shock, respectively. The activities of five enzymes related to the metabolism of the endogenous glutamate were determined and analyzed. It was found that the activities of glucose-6-phosphate dehydrogenase, isocitrate dehydrogenase, glutamate dehydrogenase and glutamate synthase were significantly altered during heat- and osmotic shock, while the activity of α-ketoglutarate dehydrogenase only showed a little alteration. This study provides a basis for the industrial production and use of γ-PGA, and for understanding its biosynthetic mechanism in B. subtilis ZJS18.

RNALocate: a resource for RNA subcellular localizations.

Increasing evidence has revealed that RNA subcellular localization is a very important feature for deeply understanding RNA's biological functions after being transported into intra- or extra-cellular regions. RNALocate is a web-accessible database that aims to provide a high-quality RNA subcellular localization resource and facilitate future researches on RNA function or structure. The current version of RNALocate documents more than 37 700 manually curated RNA subcellular localization entries with experimental evidence, involving more than 21 800 RNAs with 42 subcellular localizations in 65 species, mainly including Homo sapiens, Mus musculus and Saccharomyces cerevisiae etc. Besides, RNA homology, sequence and interaction data have also been integrated into RNALocate. Users can access these data through online search, browse, blast and visualization tools. In conclusion, RNALocate will be of help in elucidating the entirety of RNA subcellular localization, and developing new prediction methods. The database is available at http://www.rna-society.org/rnalocate/.

Personalized surgical recommendations and quantitative therapeutic insights for patients with metastatic breast cancer: Insights from deep learning.

Background: The role of surgery in metastatic breast cancer (MBC) is currently controversial. Several novel statistical and deep learning (DL) methods promise to infer the suitability of surgery at the individual level. Objective: The objective of this study was to identify the most applicable DL model for determining patients with MBC who could benefit from surgery and the type of surgery required. Methods: We introduced the deep survival regression with mixture effects (DSME), a semi-parametric DL model integrating three causal inference methods. Six models were trained to make individualized treatment recommendations. Patients who received treatments in line with the DL models' recommendations were compared with those who underwent treatments divergent from the recommendations. Inverse probability weighting (IPW) was used to minimize bias. The effects of various features on surgery selection were visualized and quantified using multivariate linear regression and causal inference. Results: In total, 5269 female patients with MBC were included. DSME was an independent protective factor, outperforming other models in recommending surgery (IPW-adjusted hazard ratio [HR] = 0.39, 95% confidence interval [CI]: 0.19-0.78) and type of surgery (IPW-adjusted HR = 0.66, 95% CI: 0.48-0.93). DSME was superior to other models and traditional guidelines, suggesting a higher proportion of patients benefiting from surgery, especially breast-conserving surgery. The debiased effect of patient characteristics, including age, tumor size, metastatic sites, lymph node status, and breast cancer subtypes, on surgery decision was also quantified. Conclusions: Our findings suggested that DSME could effectively identify patients with MBC likely to benefit from surgery and the specific type of surgery needed. This method can facilitate the development of efficient, reliable treatment recommendation systems and provide quantifiable evidence for decision-making.

Construction of high-quality genomes and gene catalogue for culturable microbes of sugarcane (Saccharum spp.).

Microbes living inside or around sugarcane (Saccharum spp.) are crucial for their resistance to abiotic and biotic stress, growth, and development. Sequences of microbial genomes and genes are helpful to understand the function of these microbes. However, there is currently a lack of such knowledge in sugarcane. Here, we combined Nanopore and Illumina sequencing technologies to successfully construct the first high-quality metagenome-assembled genomes (MAGs) and gene catalogues of sugarcane culturable microbes (GCSCMs), which contained 175 species-level genome bins (SGBs), and 7,771,501 non-redundant genes. The SGBs included 79 novel culturable bacteria genomes, and 3 bacterial genomes with nitrogen-fixing gene clusters. Four single scaffold near-complete circular MAGs (cMAGs) with 0% contamination were obtained from Nanopore sequencing data. In conclusion, we have filled a research gap in the genomes and gene catalogues of culturable microbes of sugarcane, providing a vital data resource for further understanding the genetic basis and functions of these microbes. In addition, our methodology and results can provide guidance and reference for other plant microbial genome and gene catalogue studies.

Determining individual suitability for neoadjuvant systemic therapy in breast cancer patients through deep learning.

BACKGROUND: The survival advantage of neoadjuvant systemic therapy (NST) for breast cancer patients remains controversial, especially when considering the heterogeneous characteristics of individual patients. OBJECTIVE: To discern the variability in responses to breast cancer treatment at the individual level and propose personalized treatment recommendations utilizing deep learning (DL). METHODS: Six models were developed to offer individualized treatment suggestions. Outcomes for patients whose actual treatments aligned with model recommendations were compared to those whose did not. The influence of certain baseline features of patients on NST selection was visualized and quantified by multivariate logistic regression and Poisson regression analyses. RESULTS: Our study included 94,487 female breast cancer patients. The Balanced Individual Treatment Effect for Survival data (BITES) model outperformed other models in performance, showing a statistically significant protective effect with inverse probability treatment weighting (IPTW)-adjusted baseline features [IPTW-adjusted hazard ratio: 0.51, 95% confidence interval (CI), 0.41-0.64; IPTW-adjusted risk difference: 21.46, 95% CI 18.90-24.01; IPTW-adjusted difference in restricted mean survival time: 21.51, 95% CI 19.37-23.80]. Adherence to BITES recommendations is associated with reduced breast cancer mortality and fewer adverse effects. BITES suggests that patients with TNM stage IIB, IIIB, triple-negative subtype, a higher number of positive axillary lymph nodes, and larger tumors are most likely to benefit from NST. CONCLUSIONS: Our results demonstrated the potential of BITES to aid in clinical treatment decisions and offer quantitative treatment insights. In our further research, these models should be validated in clinical settings and additional patient features as well as outcome measures should be studied in depth.

Genetic Predisposition, Sedentary Behavior, and Incident Coronary Artery Disease: A Prospective Chinese Cohort Study.

PURPOSE: Whether the association of sedentary behaviors with coronary artery disease (CAD) can be influenced by genetic susceptibility remains unclear. We aimed to investigate the joint and interplay effects between genetic risk and sedentary time (ST) and to further explore the extent to which the risk for CAD can be counteracted by reducing ST in different genetic groups. METHODS: This prospective cohort study included 39,164 Chinese adults without CAD history. Genetic susceptibility was quantified by a predefined polygenic risk score (PRS) with 540 genetic variants, and daily ST was assessed by questionnaire. We analyzed the modification effect of genetic risk on the association of ST with CAD using the Cox proportional hazards models. RESULTS: During a median follow-up of 11.60 yr, 1156 CAD events were documented. Higher ST and PRS were separately related to elevated CAD risk. Significant additive interaction was also observed (relative excess risk due to interaction: 0.77; 95% confidence interval [CI] = 0.27-1.28). Compared with participants with low genetic risk and low ST (<6 h·d -1 ), those with high genetic risk and high ST (≥10 h·d -1 ) had the highest CAD risk, with the hazard ratio (HR) and 95% CI of 4.22 (2.65-6.71). When stratified by genetic risks, participants with high ST had gradient increment of CAD risks across low, intermediate, and high genetic risk groups, with HR (95% CI) values of 1.21 (0.61-2.40), 1.57 (1.14-2.16), and 2.15 (1.40-3.31), respectively. For the absolute risk reduction, individuals with high genetic risk achieved the greatest benefit from low ST ( Ptrend = 0.024). CONCLUSIONS: Genetic susceptibility may synergistically interact with ST to increase CAD risk. Reducing ST could attenuate the CAD risk, especially among individuals with high genetic risk.

Light-induced Extracellular Vesicle Adsorption.

The role of extracellular vesicles (EVs) in human health and disease has garnered considerable attention over the past two decades. However, while several types of EVs are known to interact dynamically with the extracellular matrix and there is great potential value in producing high-fidelity EV micropatterns, there are currently no label-free, high-resolution, and tunable platform technologies with this capability. We introduce Light-induced Extracellular Vesicle Adsorption (LEVA) as a powerful solution to rapidly advance the study of matrix- and surface-bound EVs and other particles. The versatility of LEVA is demonstrated using commercial GFP-EV standards, EVs from glioblastoma bioreactors, and E. coli outer membrane vesicles (OMVs), with the resulting patterns used for single EV characterization, single cell migration on migrasome-mimetic trails, and OMV-mediated neutrophil swarming. LEVA will enable rapid advancements in the study of matrix- and surface-bound EVs and other particles, and should encourage researchers from many disciplines to create novel diagnostic, biomimetic, immunoengineering, and therapeutic screening assays.

Federated Data Quality Assessment Approach: Robust Learning With Mixed Label Noise.

Federated learning (FL) has been an effective way to train a machine learning model distributedly, holding local data without exchanging them. However, due to the inaccessibility of local data, FL with label noise would be more challenging. Most existing methods assume only open-set or closed-set noise and correspondingly propose filtering or correction solutions, ignoring that label noise can be mixed in real-world scenarios. In this article, we propose a novel FL method to discriminate the type of noise and make the FL mixed noise-robust, named FedMIN. FedMIN employs a composite framework that captures local-global differences in multiparticipant distributions to model generalized noise patterns. By determining adaptive thresholds for identifying mixed label noise in each client and assigning appropriate weights during model aggregation, FedMIN enhances the performance of the global model. Furthermore, FedMIN incorporates a loss alignment mechanism using local and global Gaussian mixture models (GMMs) to mitigate the risk of revealing samplewise loss. Extensive experiments are conducted on several public datasets, which include the simulated FL testbeds, i.e., CIFAR-10, CIFAR-100, and SVHN, and the real-world ones, i.e., Camelyon17 and multiorgan nuclei challenge (MoNuSAC). Compared to FL benchmarks, FedMIN improves model accuracy by up to 9.9% due to its superior noise estimation capabilities.

Iterative Self-Training Based Domain Adaptation for Cross-User sEMG Gesture Recognition.

Surface electromyography (sEMG) based gesture recognition has received broad attention and application in rehabilitation areas for its direct and fine-grained sensing ability. sEMG signals exhibit strong user dependence properties among users with different physiology, causing the inapplicability of the recognition model on new users. Domain adaptation is the most representative method to reduce the user gap with feature decoupling to acquire motion-related features. However, the existing domain adaptation method shows awful decoupling results when handling complex time-series physiological signals. Therefore, this paper proposes an Iterative Self-Training based Domain Adaptation method (STDA) to supervise the feature decoupling process with the pseudo-label generated by self-training and to explore cross-user sEMG gesture recognition. STDA mainly consists of two parts, discrepancy-based domain adaptation (DDA) and pseudo-label iterative update (PIU). DDA aligns existing users' data and new users' unlabeled data with a Gaussian kernel-based distance constraint. PIU Iteratively continuously updates pseudo-labels to generate more accurate labelled data on new users with category balance. Detailed experiments are performed on publicly available benchmark datasets, including the NinaPro dataset (DB-1 and DB-5) and the CapgMyo dataset (DB-a, DB-b, and DB-c). Experimental results show that the proposed method achieves significant performance improvement compared with existing sEMG gesture recognition and domain adaption methods.

Pan-Cancer Bioinformatics Analysis of Gene UBE2C.

Ubiquitin-Conjugating Enzyme E2 C (UBE2C) is a gene that encodes protein. Disorders associated with UBE2C include methotrexate-related lymphatic hyperplasia and complement component 7 deficiency. The encoded protein is necessary for the destruction of mitotic cell cyclins and cell cycle progression, and may be involved in cancer progression. In this paper, on the basis of public databases, we study the expression differential mechanism of gene expression of UBE2C in various tumors and the performance of prognosis, clinical features, immunity, methylation, etc.

Path-ATT-CNN: A Novel Deep Neural Network Method for Key Pathway Identification of Lung Cancer.

Attention convolutional neural networks (ATT-CNNs) have got a huge gain in picture operating and nature language processing. Shortage of interpretability cannot remain the adoption of deep neural networks. It is very conspicuous that is shown in the prediction model of disease aftermath. Biological data are commonly revealed in a nominal grid data structured pattern. ATT-CNN cannot be applied directly. In order to figure out these issues, a novel method which is called the Path-ATT-CNN is proposed by us, making an explicable ATT-CNN model based on united omics data by making use of a recently characterized pathway image. Path-ATT-CNN shows brilliant predictive demonstration difference in primary lung tumor symptom (PLTS) and non-primary lung tumor symptom (non-PLTS) when applied to lung adenocarcinomas (LADCs). The imaginational tool adoption which is linked with statistical analysis enables the status of essential pathways which finally exist in LADCs. In conclusion, Path-ATT-CNN shows that it can be effectively put into use elucidating omics data in an interpretable mode. When people start to figure out key biological correlates of disease, this mode makes promising power in predicting illness.

Causal associations of alcohol consumption with cardiovascular diseases and all-cause mortality among Chinese males.

BACKGROUND: The causal effects of moderate alcohol consumption on cardiovascular diseases (CVDs) are continuously debated, especially on coronary artery disease (CAD). OBJECTIVES: We aimed to explore the causal associations of alcohol consumption with CVDs and all-cause mortality among Chinese males. METHODS: A prospective cohort study was conducted in 40,386 Chinese males, with 17,676 being genotyped for the rs671 variant in the aldehyde dehydrogenase 2 (ALDH2) gene. A Cox proportional hazards model was conducted to estimate the effects of self-reported alcohol consumption. Mendelian randomization (MR) analysis was performed to explore the causality using rs671 as an instrumental variable. RESULTS: During the follow-up of 303,353 person-years, 2406 incident CVDs and 3195 all-cause mortalities were identified. J-shaped associations of self-reported alcohol consumption with incident CVD and all-cause mortality were observed, showing decreased risks for light (≤25 g/d) and moderate drinkers (25-≤60 g/d). However, MR analyses revealed a linear association of genetically predicted alcohol consumption with the incident CVD (P-trend = 0.02), including both CAD (P-trend = 0.03) and stroke (P-trend = 0.02). The HRs (95% CIs) for incident CVD across increasing tertiles of genetically predicted alcohol consumption were 1 (reference), 1.18 (1.01, 1.38), and 1.22 (1.03, 1.46). After excluding heavy drinkers, the risk of incident CVD and all-cause mortality was increased by 27% and 20% per standard drink increment of genetically predicted alcohol consumption, respectively. CONCLUSIONS: Our analyses extend the evidence of the harmful effect of alcohol consumption to total CVD (including CAD) and all-cause mortality, highlighting the potential health benefits of lowering alcohol consumption, even among light-to-moderate male drinkers.

Applications of deep learning in fundus images: A review.

The use of fundus images for the early screening of eye diseases is of great clinical importance. Due to its powerful performance, deep learning is becoming more and more popular in related applications, such as lesion segmentation, biomarkers segmentation, disease diagnosis and image synthesis. Therefore, it is very necessary to summarize the recent developments in deep learning for fundus images with a review paper. In this review, we introduce 143 application papers with a carefully designed hierarchy. Moreover, 33 publicly available datasets are presented. Summaries and analyses are provided for each task. Finally, limitations common to all tasks are revealed and possible solutions are given. We will also release and regularly update the state-of-the-art results and newly-released datasets at https://github.com/nkicsl/Fundus_Review to adapt to the rapid development of this field.

Synthetic Lethality-based Identification of Targets for Anticancer Drugs in the Human Signaling Network.

Chemotherapy agents can cause serious adverse effects by attacking both cancer tissues and normal tissues. Therefore, we proposed a synthetic lethality (SL) concept-based computational method to identify specific anticancer drug targets. First, a 3-step screening strategy (network-based, frequency-based and function-based screening) was proposed to identify the SL gene pairs by mining 697 cancer genes and the human signaling network, which had 6306 proteins and 62937 protein-protein interactions. The network-based screening was composed of a stability score constructed using a network information centrality measure (the average shortest path length) and the distance-based screening between the cancer gene and the non-cancer gene. Then, the non-cancer genes were extracted and annotated using drug-target interaction and drug description information to obtain potential anticancer drug targets. Finally, the human SL data in SynLethDB, the existing drug sensitivity data and text-mining were utilized for target validation. We successfully identified 2555 SL gene pairs and 57 potential anticancer drug targets. Among them, CDK1, CDK2, PLK1 and WEE1 were verified by all three aspects and could be preferentially used in specific targeted therapy in the future.

TLR2/TLR4 activation induces Tregs and suppresses intestinal inflammation caused by Fusobacterium nucleatum in vivo.

Toll-like receptors (TLRs) 2 and 4 play critical roles in intestinal inflammation caused by Fusobacterium nucleatum (F. nucleatum) infection, but the role of TLR2/TLR4 in regulation of proinflammatory cytokines remains unknown. In this study, through microarray analysis and qRT-PCR, we showed that TLR2/TLR4 are involved in the F. nucleatum-induced inflammatory signaling pathway in Caco-2 cells, C57BL/6 mice and human clinical specimens. In TLR2-/- and TLR4-/- mice, F. nucleatum infection resulted in increased colonization of the bacteria and production of the proinflammatory cytokines IL-8, IL-1ß and TNF-α. In addition, the ratio of Foxp3+ CD4+ T cells in the total CD4+ T cells in TLR2-/- and TLR4-/- mice was less than that in wild-type mice, and the ratio in hybrid mice was more than that in knockout mice, which suggested that TLR2/TLR4 mediated the number of Tregs. Furthermore, it was observed that inflammatory cytokine levels were reduced in TLR2-/- mice after Treg transfer. Thus, these data indicate that TLR2/TLR4 regulate F. nucleatum-induced inflammatory cytokines through Tregs in vivo.