Low complexity sub-baud rate sampling reception in a bandwidth-limited IM/DD system.

This Letter presents what is to our knowledge a novel approach to reduce the digital signal processing (DSP) complexity in intensity modulation and direct detection (IM/DD) systems, which is critical for short-reach optical communication systems with severe bandwidth limitations. We propose a sub-baud rate sampling reception method utilizing a polyphase feedforward equalizer-based maximum likelihood sequence estimation (PFFE-MLSE), which could operate effectively under a sampling rate of 0.6 samples per symbol. This new architecture eliminates the need for resampling, allowing the adaptive equalizer to operate with significantly reduced complexity-over 60% compared to traditional FFE-MLSE. An offline experiment, transmitting a 100-Gbaud on-off keying (OOK) signal over a 5-km single-mode fiber (SMF) link, demonstrates the feasibility of our approach with bit error ratio (BER) meeting the KP4-forward error correction (KP4-FEC) threshold in the optical back-to-back (OBTB) scenario and 7% hard-decision FEC (HD-FEC) threshold in the 5-km SMF transmission.

Effectiveness and safety of combined nimotuzumab and S-1 chemotherapy with concurrent radiotherapy for locally advanced esophageal cancer in malnourished and elderly patients: A prospective phase II study.

Objective: Definitive chemoradiotherapy (dCRT) is the standard treatment for unresectable locally advanced esophageal cancer. However, this treatment is associated with substantial toxicity, and most malnourished or elderly patients are unable to complete this therapy. Therefore, there is a need for a more suitable radiotherapy combination regimen for this population. This study was aimed to evaluate the efficacy and safety of a combination regimen comprising chemotherapy with nimotuzumab and S-1 and concurrent radiotherapy for patients with fragile locally advanced esophageal cancer with a high Nutritional Risk Screening 2002 (NRS-2002) score. Methods: Eligible patients with unresectable esophageal carcinoma who had an NRS-2002 score of 2 or higher were enrolled. They were treated with S-1 and nimotuzumab with concurrent radiotherapy, followed by surgery or definitive radiotherapy. The primary endpoint was the locoregional control (LRC) rate. Results: A total of 55 patients who met the study criteria were enrolled. After completion of treatment, surgery was performed in 15 patients and radiotherapy was continued in 40 patients. The median follow-up period was 33.3 [95% confidence interval (95% CI), 31.4-35.1)] months. The LRC rate was 77.2% (95% CI, 66.6%-89.4%) at 1 year in the entire population. The overall survival (OS) rate and event-free survival (EFS) rate were 57.5% and 51.5% at 3 years, respectively. Surgery was associated with better LRC [hazard ratio (HR)=0.16; 95% CI, 0.04-0.70; P=0.015], OS (HR=0.19; 95% CI, 0.04-0.80; P=0.024), and EFS (HR=0.25; 95% CI, 0.08-0.75; P=0.013). Most adverse events were of grade 1 or 2, and no severe adverse events occurred. Conclusions: For malnourished or elderly patients with locally advanced esophageal cancer, radiotherapy combined with nimotuzumab and S-1 is effective and has a good safety profile.

TGC-ARG: Anticipating Antibiotic Resistance via Transformer-Based Modeling and Contrastive Learning.

In various domains, including everyday activities, agricultural practices, and medical treatments, the escalating challenge of antibiotic resistance poses a significant concern. Traditional approaches to studying antibiotic resistance genes (ARGs) often require substantial time and effort and are limited in accuracy. Moreover, the decentralized nature of existing data repositories complicates comprehensive analysis of antibiotic resistance gene sequences. In this study, we introduce a novel computational framework named TGC-ARG designed to predict potential ARGs. This framework takes protein sequences as input, utilizes SCRATCH-1D for protein secondary structure prediction, and employs feature extraction techniques to derive distinctive features from both sequence and structural data. Subsequently, a Siamese network is employed to foster a contrastive learning environment, enhancing the model's ability to effectively represent the data. Finally, a multi-layer perceptron (MLP) integrates and processes sequence embeddings alongside predicted secondary structure embeddings to forecast ARG presence. To evaluate our approach, we curated a pioneering open dataset termed ARSS (Antibiotic Resistance Sequence Statistics). Comprehensive comparative experiments demonstrate that our method surpasses current state-of-the-art methodologies. Additionally, through detailed case studies, we illustrate the efficacy of our approach in predicting potential ARGs.

Signal amplification strategies in photoelectrochemical sensing of carcinoembryonic antigen.

Early detection of cancer markers is critical for cancer diagnosis and cancer therapy since these markers may indicate cancer risk, incidence, and disease prognosis. Carcinoembryonic antigen (CEA) is a type of non-specific and broad-spectrum cancer biomarker commonly utilized for early cancer diagnosis. Moreover, it serves as an essential tool to assess the efficacy of cancer treatment and monitor tumor recurrence as well as metastasis, thus garnering significant attention for precise and sensitive CEA detection. In recent years, photoelectrochemical (PEC) techniques have emerged as prominent methods in CEA detection due to the advantages of PEC, such as simple equipment requirements, cost-effectiveness, high sensitivity, low interference from background signals, and easy of instrument miniaturization. Different signal amplification methods have been reported in PEC sensors for CEA analysis. Based on these, this article reviews PEC sensors based on various signal amplification strategies for detection of CEA during the last five years. The advantages and drawbacks of these sensors were discussed, as well as future challenges.

Preclinical study and phase II trial of adapting low-dose radiotherapy to immunotherapy in small cell lung cancer.

BACKGROUND: Immune checkpoint inhibitors (ICIs) provide modest but unsatisfactory benefits for extensive-stage small cell lung cancer (ES-SCLC). Developing strategies for treating ES-SCLC is critical. METHODS: We preliminarily explored the outcomes of salvage low-dose radiotherapy (LDRT) plus ICI on refractory SCLC patients. Next, we evaluated the combinational efficacy in murine SCLC. The tumor immune microenvironment (TIME) was analyzed for mechanistic study. Subsequently, we conducted a multicenter, prospective phase II trial that administered concurrent thoracic LDRT plus chemoimmunotherapy to treatment-naive ES-SCLC patients (MATCH trial, NCT04622228). The primary endpoint was confirmed objective response rate (ORR), and the key secondary endpoints included progression-free survival (PFS) and safety. FINDINGS: Fifteen refractory SCLC patients treated with LDRT plus ICI were retrospectively reviewed. The ORR was 73.3% (95% confidence interval [CI], 44.9-92.2). We identified a specific dose of LDRT (15 Gy/5 fractions) that exhibited growth retardation and improved survival in murine SCLC when combined with ICIs. This combination recruited a special T cell population, TCF1+ PD-1+ CD8+ stem-like T cells, from tumor-draining lymph nodes into the TIME. The MATCH trial showed a confirmed ORR of 87.5% (95% CI, 75.9-94.8). The median PFS was 6.9 months (95% CI, 5.4-9.3). CONCLUSIONS: These findings verified that LDRT plus chemoimmunotherapy was safe, feasible, and effective for ES-SCLC, warranting further investigation. FUNDING: This research was funded by West China Hospital (no. ZYJC21003), the National Natural Science Foundation of China (no. 82073336), and the MATCH trial was fully funded by Roche (China) Holding Ltd. (RCHL) and Shanghai Roche Pharmaceuticals Ltd. (SRPL).

Ensemble Machine Learning and Predicted Properties Promote Antimicrobial Peptide Identification.

The emergence of antibiotic-resistant microbes raises a pressing demand for novel alternative treatments. One promising alternative is the antimicrobial peptides (AMPs), a class of innate immunity mediators within the therapeutic peptide realm. AMPs offer salient advantages such as high specificity, cost-effective synthesis, and reduced toxicity. Although some computational methodologies have been proposed to identify potential AMPs with the rapid development of artificial intelligence techniques, there is still ample room to improve their performance. This study proposes a predictive framework which ensembles deep learning and statistical learning methods to screen peptides with antimicrobial activity. We integrate multiple LightGBM classifiers and convolution neural networks which leverages various predicted sequential, structural and physicochemical properties from their residue sequences extracted by diverse machine learning paradigms. Comparative experiments exhibit that our method outperforms other state-of-the-art approaches on an independent test dataset, in terms of representative capability measures. Besides, we analyse the discrimination quality under different varieties of attribute information and it reveals that combination of multiple features could improve prediction. In addition, a case study is carried out to illustrate the exemplary favorable identification effect. We establish a web application at http://amp.denglab.org to provide convenient usage of our proposal and make the predictive framework, source code, and datasets publicly accessible at https://github.com/researchprotein/amp .

Preferential segregation of gold at the symmetrical tilt grain boundaries of platinum: an atomic-scale quantitative understanding.

Grain boundary (GB) segregation plays a pivotal role in maintaining and optimizing the remarkable catalytic or mechanical properties of nanocrystalline Pt by reducing the Gibbs free energy and thereby impeding structure degradation. The solute segregation behavior at the Pt GB, however, is not well understood at the atomic level. In this study, we employed first-principles calculations to elucidate the preferential segregation behavior of a single Au atom at the symmetrical tilt GB of Pt. For pure Pt, a linear relationship between the GB energy and excess volume is observed. Therefore, Au exhibits strong segregation tendencies towards GB to release excess energy and volume stored at the strained GB. Although the segregation energy is sensitive to various GB sites, it is interesting to note that the minimum one increases linearly with GB energy. This site-sensitivity of segregation energy can be attributed to mechanical, chemical, and interaction parts, which are quantitatively related to the atomic volume, coordination number, and average bond length, respectively. Finally, the interplay among different structural descriptors is revealed. These insights into the association between GB structures, segregation configuration and energy offers valuable atomic-scale quantitative insights into the segregation behavior of Au in Pt GBs, which holds significant implications for the design of Pt nanomaterials with enhanced thermal stability via GB engineering.

Immune-related adverse events of immune checkpoint inhibitors combined with angiogenesis inhibitors: A real-world pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database (2014-2022).

INTRODUCTION: Although immune checkpoint inhibitors (ICIs) combined with angiogenesis inhibitors (AGIs) has become increasingly used for cancers, the impact of combination therapy on immune-related adverse events (irAEs) in real-world settings has not been well elucidated to date. METHODS: The FDA Adverse Event Reporting System (FAERS) database from 2014 to 2022 was retrospectively queried to extract reports of irAEs referred as standardized MedDRA queries (SMQs), preferred terms (PTs) and system organ classes (SOCs). To perform disproportionality analysis, information component (IC) and reporting odds ratio (ROR) were calculated and lower limit of 95 % confidence interval (CI) for IC (IC025) > 0 or ROR (ROR025) > 1 with at least 3 reports was considered statistically significant. RESULTS: Compared to ICIs alone, ICIs + AGIs demonstrated a lower IC025/ROR025 for irAEs-SMQ (2.343/5.082 vs. 1.826/3.563). Regarding irAEs-PTs, there were fewer irAEs-PTs of significant value in ICIs + AGIs than ICIs alone (57 vs. 150 PTs) and lower signal value for most PTs (88 %) in ICIs + AGIs. Moreover, lower IC025 for most of irAEs-SOCs in ICIs + AGIs (11/13) compared with ICIs alone was observed. As for outcomes of irAEs, ICIs + AGIs showed a lower frequency of "fatal" for irAEs-SMQ than ICIs alone (4.88 % vs. 7.83 %), so as in cardiac disorder (SOC) (15.45 % vs. 26.37 %), and respiratory, thoracic and mediastinal disorders (SOC) (13.74 % vs. 20.06 %). Similarly, there were lower occurrence and fewer fatality of irAEs in ICIs + AGIs + chemotherapy (CT) than ICIs + CT. CONCLUSION: ICIs combined with AGIs may reduce incidence and mortality for most of irAEs compared to ICIs alone whether or not with CT.

Chromosome-level genome assembly of Solanum pimpinellifolium.

Solanum pimpinellifolium, the closest wild relative of the domesticated tomato, has high potential for use in breeding programs aimed at developing multi-pathogen resistance and quality improvement. We generated a chromosome-level genome assembly of S. pimpinellifolium LA1589, with a size of 833 Mb and a contig N50 of 31 Mb. We anchored 98.80% of the contigs into 12 pseudo-chromosomes, and identified 74.47% of the sequences as repetitive sequences. The genome evaluation revealed BUSCO and LAI score of 98.3% and 14.49, respectively, indicating high quality of this assembly. A total of 41,449 protein-coding genes were predicted in the genome, of which 89.17% were functionally annotated. This high-quality genome assembly serves as a valuable resource for accelerating the biological discovery and molecular breeding of this important horticultural crop.

Robust and adaptive star identification algorithm based on linear assignment for multiple large field of view visual imaging systems.

The integration of the visual imaging system and the self-attitude determination system in on-orbit space projects necessitates robust star identification algorithms. However, disturbances in the on-orbit environment pose a challenge to the accuracy and efficiency of star identification algorithms. This paper introduces a novel star identification algorithm, to the best of our knowledge, designed for multiple large field of view (FOV) visual imaging systems, providing stability in the presence of the noise types, including position noise, lost-star noise, and fake-star noise. We employ the dynamic simulated star images generation method to construct simulated star image libraries suitable for various cameras with different FOV angles. Our algorithm comprises two parts: the star edge matching for coarse matching of interstellar angular distances based on linear assignment, and the star point registration for precise matching of star vectors. This innovative combination of local edge generation and global matching approach achieves an impressive 97.83% identification accuracy, maintaining this performance even with a standard deviation of one pixel in image plane errors and up to five missing stars. A comprehensive approach involving both simulated and empirical experiments was employed to validate the algorithm's effectiveness. This integration of the visual imaging system and the self-attitude determination system offers potential benefits such as reduced space equipment weight, simplified satellite launch processes, and decreased maintenance costs.

Efficacy and risk of donor-derived CAR-T treatment of relapsed B-cell acute lymphoblastic leukemia after hematopoietic stem cell transplantation.

The one-year survival rate for patients experiencing a relapse of B-cell acute lymphocytic leukemia (B-ALL) following hematopoietic stem cell transplantation (HSCT) is approximately 30%. Patients experiencing a relapse after allogeneic HSCT frequently encounter difficulties in obtaining autologous CAR-T products. We conducted a study involving 14 patients who received donor-derived CAR-T therapy for relapsed B-ALL following HSCT between August 2019 and May 2023 in our center. The results revealed a CR/CRi rate of 78.6% (11/14), a GVHD rate of 21.4% (3/14), and a 1-year overall survival (OS) rate of 56%. Decreased bone marrow donor cell chimerism in 9 patients recovered after CAR-T therapy. The main causes of death were disease progression and infection. Further analysis showed that GVHD (HR 7.224, 95% CI 1.42-36.82, P = 0.017) and platelet recovery at 30 days (HR 6.807, 95% CI 1.61-28.83, P = 0.009) are significantly associated with OS after CAR-T therapy. Based on the findings, we conclude that donor-derived CAR-T cells are effective in treating relapsed B-ALL patients following HSCT. Additionally, GVHD and poor platelet recovery impact OS, but further verification with a larger sample size is needed.

The impact of vegetation reconstruction on soil erosion in the Loess plateau.

Vegetation restoration not only extensively reshapes spatial land use patterns but also profoundly affects the dynamics of runoff and sediment loss. However, the influence of vegetation restoration on runoff and sediment yield from a regional perspective are scarce. This study therefore focused on 85 sites within the "Grain for Green" Project (GGP) region on the Loess Plateau, to investigate the impacts of the GGP on soil erosion. The results revealed a notable reduction in sediment loss and runoff due to vegetation restoration. Since the inception of the GGP in 1999, approximately 4.1 × 106 ha of degraded lands have been converted into forestlands, shrublands, and grasslands, resulting in an average annual reduction of 1.4 × 109 m3 in runoff and a decrease of 3.6 × 108 t in annual sediment loss on the whole Loess Plateau, with the GGP contributing approximately 26.7% of the sediment reduction in the Yellow River basin. The reduced soil erosion has mainly been regulated by vegetation cover, soil properties (clay, silt, and sand), slope, and precipitation on the Loess Plateau. The insights gained offer valuable contributions to large-scale assessments of changes in soil erosion in response to vegetation reconstruction and enhance our understanding of the spatial configurations associated with soil erosion control measures.

DeepFusionCDR: Employing Multi-Omics Integration and Molecule-Specific Transformers for Enhanced Prediction of Cancer Drug Responses.

Deep learning approaches have demonstrated remarkable potential in predicting cancer drug responses (CDRs), using cell line and drug features. However, existing methods predominantly rely on single-omics data of cell lines, potentially overlooking the complex biological mechanisms governing cell line responses. This paper introduces DeepFusionCDR, a novel approach employing unsupervised contrastive learning to amalgamate multi-omics features, including mutation, transcriptome, methylome, and copy number variation data, from cell lines. Furthermore, we incorporate molecular SMILES-specific transformers to derive drug features from their chemical structures. The unified multi-omics and drug signatures are combined, and a multi-layer perceptron (MLP) is applied to predict IC50 values for cell line-drug pairs. Moreover, this MLP can discern whether a cell line is resistant or sensitive to a particular drug. We assessed DeepFusionCDR's performance on the GDSC dataset and juxtaposed it against cutting-edge methods, demonstrating its superior performance in regression and classification tasks. We also conducted ablation studies and case analyses to exhibit the effectiveness and versatility of our proposed approach. Our results underscore the potential of DeepFusionCDR to enhance CDR predictions by harnessing the power of multi-omics fusion and molecular-specific transformers. The prediction of DeepFusionCDR on TCGA patient data and case study highlight the practical application scenarios of DeepFusionCDR in real-world environments. Source code and datasets can be available on https://github.com/altriavin/DeepFusionCDR.

MvMRL: a multi-view molecular representation learning method for molecular property prediction.

Effective molecular representation learning is very important for Artificial Intelligence-driven Drug Design because it affects the accuracy and efficiency of molecular property prediction and other molecular modeling relevant tasks. However, previous molecular representation learning studies often suffer from limitations, such as over-reliance on a single molecular representation, failure to fully capture both local and global information in molecular structure, and ineffective integration of multiscale features from different molecular representations. These limitations restrict the complete and accurate representation of molecular structure and properties, ultimately impacting the accuracy of predicting molecular properties. To this end, we propose a novel multi-view molecular representation learning method called MvMRL, which can incorporate feature information from multiple molecular representations and capture both local and global information from different views well, thus improving molecular property prediction. Specifically, MvMRL consists of four parts: a multiscale CNN-SE Simplified Molecular Input Line Entry System (SMILES) learning component and a multiscale Graph Neural Network encoder to extract local feature information and global feature information from the SMILES view and the molecular graph view, respectively; a Multi-Layer Perceptron network to capture complex non-linear relationship features from the molecular fingerprint view; and a dual cross-attention component to fuse feature information on the multi-views deeply for predicting molecular properties. We evaluate the performance of MvMRL on 11 benchmark datasets, and experimental results show that MvMRL outperforms state-of-the-art methods, indicating its rationality and effectiveness in molecular property prediction. The source code of MvMRL was released in https://github.com/jedison-github/MvMRL.

Dehydrogenative synthesis of N-functionalized 2-aminophenols from cyclohexanones and amines: Molecular complexities via one-shot assembly.

Polyfunctionalized arenes are privileged structural motifs in both academic and industrial chemistry. Conventional methods for accessing this class of chemicals usually involve stepwise modification of phenyl rings, often necessitating expensive noble metal catalysts and suffering from low reactivity and selectivity when introducing multiple functionalities. We herein report dehydrogenative synthesis of N-functionalized 2-aminophenols from cyclohexanones and amines. The developed reaction system enables incorporating amino and hydroxyl groups into aromatic rings in a one-shot fashion, which simplifies polyfunctionalized 2-aminophenol synthesis by circumventing issues associated with traditional arene modifications. The wide substrate scope and excellent functional group tolerance are exemplified by late-stage modification of complex natural products and pharmaceuticals that are unattainable by existing methods. This dehydrogenative protocol benefits from using 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) as oxidant that offers interesting chemo- and regio-selective oxidation processes. More notably, the essential role of in situ generated water is disclosed, which protects aliphatic amine moieties from overoxidation via hydrogen bond-enabled interaction.

Tab-Cox: An Interpretable Deep Survival Analysis Model for Patients with Nasopharyngeal Carcinoma based on TabNet.

The nutritional status of cancer patients is closely associated with the clinical progression of the disease. A survival analysis model combined with a neural network can predict future disease trends in patients, facilitating early prevention and assisting physicians in making diagnoses. However, the complexity of neural networks and their incompatibility with medical tabular data can reduce the interpretability of the model. To address this issue, thr paper propose a novel survival analysis model called Tab-Cox, which combines TabNet and Cox models. This model is specifically designed to predict the survival outcomes of patients with nasopharyngeal carcinoma. The model utilizes TabNet's sequential attention mechanism to extract more interpretable features, providing an interpretable method for identifying disease risk factors. Consequently, the model ensures accurate survival prediction while also making the results more comprehensible for both patients and doctors. The paper tested the efficacy of the model by conducting experiments on various diverse datasets in comparison with other commonly used survival models. The results showed that the proposed model delivered the highest or second-highest accuracy across all datasets. Furthermore, the paper conducted a comparative interpretability analysis against the classical Cox model. In addition and compare the interpretability of the Tab-Cox model with the classical Cox model and discuss the advantages and disadvantages of its interpretability. This demonstrates that Tab-Cox can assist doctors in identifying risk factors that are challenging to capture using artificial methods.

Enhancing Predictions of Drug Solubility through Multidimensional Structural Characterization Exploitation.

Solubility is not only a significant physical property of molecules but also a vital factor in smallmolecule drug development. Determining drug solubility demands stringent equipment, controlled environments, and substantial human and material resources. The accurate prediction of drug solubility using computational methods has long been a goal for researchers. In this study, we introduce MSCSol, a solubility prediction model that integrates multidimensional molecular structure information. We incorporate a graph neural network with geometric vector perceptrons (GVP-GNN) to encode 3D molecular structures, representing spatial arrangement and orientation of atoms, as well as atomic sequences and interactions. We also employ Selective Kernel Convolution combined with Global and Local attention mechanisms to capture molecular features context at different scales. Additionally, various descriptors are calculated to enrich the molecular representation. For the 2D and 3D structural data of molecules, we design different data augmentation strategies to enhance generalization ability and prevent the model from learning irrelevant information. Extensive experiments on benchmark and independent datasets demonstrate MSCSol's superior performance. Ablation studies further confirm the effectiveness of different modules. Interpretability analysis highlights the importance of various atomic groups and substructures for solubility and verifies that our model effectively captures functional molecular structures and higher-order knowledge. The source code and datasets are freely available at https://github.com/ZiyuFanCSU/MSCSol.

SGCLDGA: unveiling drug-gene associations through simple graph contrastive learning.

Drug repurposing offers a viable strategy for discovering new drugs and therapeutic targets through the analysis of drug-gene interactions. However, traditional experimental methods are plagued by their costliness and inefficiency. Despite graph convolutional network (GCN)-based models' state-of-the-art performance in prediction, their reliance on supervised learning makes them vulnerable to data sparsity, a common challenge in drug discovery, further complicating model development. In this study, we propose SGCLDGA, a novel computational model leveraging graph neural networks and contrastive learning to predict unknown drug-gene associations. SGCLDGA employs GCNs to extract vector representations of drugs and genes from the original bipartite graph. Subsequently, singular value decomposition (SVD) is employed to enhance the graph and generate multiple views. The model performs contrastive learning across these views, optimizing vector representations through a contrastive loss function to better distinguish positive and negative samples. The final step involves utilizing inner product calculations to determine association scores between drugs and genes. Experimental results on the DGIdb4.0 dataset demonstrate SGCLDGA's superior performance compared with six state-of-the-art methods. Ablation studies and case analyses validate the significance of contrastive learning and SVD, highlighting SGCLDGA's potential in discovering new drug-gene associations. The code and dataset for SGCLDGA are freely available at https://github.com/one-melon/SGCLDGA.