Study on heat transfer law of moving temperature variable gas in thermoacoustic plate stack.

Taking gas and the heat transfer process between gas and plate as the research object, the mathematical model of heat transfer in one working cycle by moving variable temperature air mass under the action of sound field is established, which provides a new idea for understanding thermoacoustic effect. The influence factors in the model are analyzed and it is found that the amplitude of the air mass has a significant influence on the heat transfer, and the heat transfer of the air mass in one working cycle is proportional to the square of the amplitude. In a certain working environment, the thermoacoustic refrigerator has a critical operating frequency, and only when the operating frequency is greater than the critical frequency can refrigeration be realized. The critical operating frequency is independent of the amplitude and increases with the increase of the stack temperature gradient. With the pressure belly point as the reference position, the greater the distance from the reference position, the greater the critical operating frequency. On this basis, the idea of short plate overlapping is put forward and the formation mechanism of temperature difference between two ends of plate overlapping is explained.

Risk assessment and molecular mechanism of Sclerotium rolfsii resistance to boscalid.

Boscalid, a widely used SDHI fungicide, has been employed in plant disease control for over two decades. However, there is currently no available information regarding its antifungal activity against Sclerotium rolfsii and the potential risk of resistance development in this pathogen. In this study, we evaluated the sensitivity of 100 S. rolfsii strains collected from five different regions in China during 2018-2019 to boscalid using mycelial growth inhibition method and assessed the risk of resistance development. The EC50 values for boscalid ranged from 0.2994 µg/mL to 1.0766 µg/mL against the tested strains, with an average EC50 value of 0.7052 ± 0.1473 µg/mL. Notably, a single peak sensitivity baseline was curved, indicating the absence of any detected resistant strains. Furtherly, 10 randomly selected strains of S. rolfsii were subjected to chemical taming to evaluate its resistance risk to boscalid, resulting in the successful generation of six stable and inheritable resistant mutants. These mutants exhibited significantly reduced mycelial growth, sclerotia production, and virulence compared to their respective parental strains. Cross-resistance tests revealed a correlation between boscalid and flutolanil, benzovindiflupyr, pydiflumetofen, fluindapyr, and thifluzamide; however, no cross-resistance was observed between boscalid and azoxystrobin. Thus, we conclude that the development risk of resistance in S. rolfsii to boscalid is low. Boscalid can be used as an alternative fungicide for controlling peanut sclerotium blight when combined with other fungicides that have different mechanisms of action. Finally, the target genes SDHB, SDHC, and SDHD in S. rolfsii were initially identified, cloned and sequenced to elucidate the mechanism of S. rolfsii resistance to boscalid. Two mutation genotypes were found in the mutants: SDHD-D111H and SDHD-H121Y. The mutants carrying SDHD-H121Y exhibited moderate resistance, while the mutants with SDHD-D111H showed low resistance. These findings contribute to our comprehensive understanding of molecular mechanisms underlying plant pathogens resistance to SDHI fungicides.

Modulating CO2 Electrocatalytic Conversion to the Organics Pathway by the Catalytic Site Dimension.

Electrochemical reduction of carbon dioxide to organic chemicals provides a value-added route for mitigating greenhouse gas emissions. We report a family of carbon-supported Sn electrocatalysts with the tin size varying from single atom, ultrasmall clusters to nanocrystallites. High single-product Faradaic efficiency (FE) and low onset potential of CO2 conversion to acetate (FE = 90% @ -0.6 V), ethanol (FE = 92% @ -0.4 V), and formate (FE = 91% @ -0.6 V) were achieved over the catalysts of different active site dimensions. The CO2 conversion mechanism behind these highly selective, size-modulated p-block element catalysts was elucidated by structural characterization and computational modeling, together with kinetic isotope effect investigation.

[Methods for traditional Chinese medicine syndrome-based efficacy evaluation: a review].

Traditional Chinese medicine(TCM) syndrome-based efficacy is an evaluation index which is unique to TCM and can reflect the advantages of TCM. The development of the methods and measurement tools for evaluating TCM syndrome-based efficacy can provide objective and quantitative evidence for the clinical efficacy evaluation of TCM and the development of new Chinese medicine preparations, being the exploration direction of innovative methods and technologies for evaluating TCM efficacy. The conventional evaluation methods are subjective and limited to the mitigation of symptoms and the improvement of physical signs, which make it difficult to form a unified evaluation standard. In addition, the evaluation methods lack unity, objectivity, and quantitative research. The scientific connotation, evaluation ideas and methods, and key technologies of the evaluation for the therapeutic effect on syndromes remain unclear, which leads to diverse evaluation modes, methods, and indexes. The syndrome-based efficacy scale provides a new idea for the objective quantification and standardization of TCM syndromes. This review systematically summarizes the methods and problems, introduces the research progress in the evaluation scales, and puts forward some thoughts on the characteristics of TCM syndrome-based efficacy evaluation, aiming to provide insights for the research in this field.

High-sensitivity dual U-shaped PCF-SPR refractive index sensor for the detection of gas and liquid analytes.

A dual U-shaped photonic crystal fiber (PCF) biochemical sensor based on surface plasmon resonance (SPR) is designed for the simultaneous detection of gas and liquid analytes, and the properties are analyzed by the full vector finite element method (FEM). SPR is excited by placing gold nanowires on the inner surface of the U-shaped device. In this technique, the traditional metal deposition process can be replaced, subsequently reducing the difficulty and complexity of actual production and improving the phase matching between the basic mode and plasmonic modes. To improve the detection properties, the structural parameters of the sensor including the air hole diameter, spacing, gold nanowire diameter, and polishing depth are optimized, and to better evaluate and analyze the sensing properties, the wavelength and amplitude modulation inquiry method is adopted. The results show that the maximum wavelength sensitivity (WS), amplitude sensitivity (AS), minimum resolution (R), and optimal FOM are 35,000 nm/RIU, 438.08R I U -1, 2.86×10-6 R I U, and 165.16R I U -1, respectively. In addition, the sensor can detect analyte RIs between 1.00 and 1.36 for gas and liquid analytes simultaneously. Owing to the simple structure, low cost, and ambient-condition monitoring, the sensor has large potential in a myriad of applications including sewage treatment, food safety, humoral regulation, environmental and biological monitoring, and medical diagnosis.

Water masses influence the variation of microbial communities in the Yangtze River Estuary and its adjacent waters.

The Yangtze River estuary (YRE) are strongly influenced by the Kuroshio and terrigenous input from rivers, leading to the formation of distinct water masses, however, there remains a limited understanding of the full extent of this influence. Here the variation of water masses and bacterial communities of 58 seawater samples from the YRE and its adjacent waters were investigated. Our findings suggested that there were 5 water masses in the studied area: Black stream (BS), coastal water in the East China Sea (CW), nearshore mixed water (NM), mixed water in the middle and deep layers of the East China Sea (MM), and deep water blocks in the middle of the East China Sea (DM). The CW mass harbors the highest alpha diversity across all layers, whereas the NM mass exhibits higher diversity in the surface layer but lower in the middle layers. Proteobacteria was the most abundant taxa in all water masses, apart from that, in the surface layer masses, Cyanobacterium, Bacteroidota, and Actinobacteriota were the highest proportion in CW, while Bacteroidota and Actinobacteriota were the highest proportion in NM and BS; in the middle layer, Bacteroidota and Actinobacteriota were dominant phylum in CW and BS masses, but Cyanobacterium was main phylum in NM mass; in the bottom layer, Bacteroidota and Actinobacteriota were the dominant phylum in CW, while Marininimicrobia was the dominated phylum in DM and MM masses. Network analysis suggests water masses have obvious influence on community topological characteristics, moreover, community assembly across masses also differ greatly. Taken together, these results emphasized the significant impact of water masses on the bacterial composition, topological characteristics and assembly process, which may provide a theoretical foundation for predicting alterations in microbial communities within estuarine ecosystems under the influence of water masses.

Impact of bioelectricity on DNRA process and microbial community composition within cathodic biofilms in dual-chambered bioelectrode microbial fuel cell (MFC).

The synchronous bioelectricity generation and dissimilatory nitrate reduction to ammonium (DNRA) pathway in Klebsiella variicola C1 was investigated. The presence of bioelectricity facilitated cell growth on the anodic biofilms, consequently enhancing the nitrate removal efficiency decreasing total nitrogen levels and causing a negligible accumulation of NO2- in the supernatant. Genomic analysis revealed that K. variicola C1 possessed a complete DNRA pathway and largely annotated electron shuttles. The up-regulated expression of genes narG and nirB, encoding nitrite oxidoreductase and nitrite reductase respectively, was closely associated with increased extracellular electron transfer (EET). High-throughput sequencing analysis was employed to investigate the impact of bioelectricity on microbial community composition within cathodic biofilms. Results indicated that Halomonas, Marinobacter and Prolixibacteraceae were enriched at the cathode electrodes. In conclusion, the integration of a DNRA strain with MFC facilitated the efficient removal of wastewater containing high concentrations of NO3- and enabled the environmentally friendly recovery of NH4+.

TripHLApan: predicting HLA molecules binding peptides based on triple coding matrix and transfer learning.

Human leukocyte antigen (HLA) recognizes foreign threats and triggers immune responses by presenting peptides to T cells. Computationally modeling the binding patterns between peptide and HLA is very important for the development of tumor vaccines. However, it is still a big challenge to accurately predict HLA molecules binding peptides. In this paper, we develop a new model TripHLApan for predicting HLA molecules binding peptides by integrating triple coding matrix, BiGRU + Attention models, and transfer learning strategy. We have found the main interaction site regions between HLA molecules and peptides, as well as the correlation between HLA encoding and binding motifs. Based on the discovery, we make the preprocessing and coding closer to the natural biological process. Besides, due to the input being based on multiple types of features and the attention module focused on the BiGRU hidden layer, TripHLApan has learned more sequence level binding information. The application of transfer learning strategies ensures the accuracy of prediction results under special lengths (peptides in length 8) and model scalability with the data explosion. Compared with the current optimal models, TripHLApan exhibits strong predictive performance in various prediction environments with different positive and negative sample ratios. In addition, we validate the superiority and scalability of TripHLApan's predictive performance using additional latest data sets, ablation experiments and binding reconstitution ability in the samples of a melanoma patient. The results show that TripHLApan is a powerful tool for predicting the binding of HLA-I and HLA-II molecular peptides for the synthesis of tumor vaccines. TripHLApan is publicly available at https://github.com/CSUBioGroup/TripHLApan.git.

Genomic survey and expression analysis of cellulose synthase superfamily and COBRA-like gene family in Zanthoxylum bungeanum stipule thorns.

The Cellulose Synthase gene (CS) superfamily and COBRA-like (COBL) gene family are essential for synthesizing cellulose and hemicellulose, which play a crucial role in cell wall biosynthesis and the hardening of plant tissues. Our study identified 126 ZbCS and 31 ZbCOBL genes from the Zanthoxylum bungeanum (Zb) genome. Phylogenetic analysis and conservative domain analysis unfolded that ZbCS and ZbCOBL genes were divided into seven and two subfamilies, respectively. Gene duplication data suggested that more than 75% of these genes had tandem and fragment duplications. Codon usage patterns analysis indicated that the ZbCS and ZbCOBL genes prefer ending with A/T base, with weak codon preference. Furthermore, seven key ZbCS and five key ZbCOBL genes were identified based on the content of cellulose and hemicellulose and the expression characteristics of ZbCS and ZbCOBL genes in various stages of stipule thorns. Altogether, these results improve the understanding of CS and COBL genes and provide valuable reference data for cultivating Zb with soft thorns. Supplementary Information: The online version contains supplementary material available at 10.1007/s12298-024-01432-x.

Recombinant bivalent subunit vaccine combining truncated VP4 from P[7] and P[23] induces protective immunity against prevalent porcine rotaviruses.

Porcine rotaviruses (PoRVs) cause severe economic losses in the swine industry. P[7] and P[23] are the predominant genotypes circulating on farms, but no vaccine is yet available. Here, we developed a bivalent subunit PoRV vaccine using truncated versions (VP4*) of the VP4 proteins from P[7] and P[23]. The vaccination of mice with the bivalent subunit vaccine elicited more robust neutralizing antibodies (NAbs) and cellular immune responses than its components, even at high doses. The bivalent subunit vaccine and inactivated bivalent vaccine prepared from strains PoRVs G9P[7] and G9P[23] were used to examine their protective efficacy in sows and suckling piglets after passive immunization. The immunized sows showed significantly elevated NAbs in the serum and colostrum, and the suckling piglets acquired high levels of sIgA antibodies from the colostrum. Challenging subunit-vaccinated or inactivated-vaccinated piglets with homologous virulent strains did not induce diarrhea, except in one or two piglets, which had mild diarrhea. Immunization with the bivalent subunit vaccine and inactivated vaccine also alleviated the microscopic lesions in the intestinal tissues caused by the challenge with the corresponding homologous virulent strain. However, all the piglets in the challenged group displayed mild to watery diarrhea and high levels of viral shedding, whereas the feces and intestines of the piglets in the bivalent subunit vaccine and inactivated vaccine groups had lower viral loads. In summary, our data show for the first time that a bivalent subunit vaccine combining VP4*P[7] and VP4*P[23] effectively protects piglets against the diarrhea caused by homologous virulent strains.IMPORTANCEPoRVs are the main causes of diarrhea in piglets worldwide. The multisegmented genome of PoRVs allows the reassortment of VP4 and VP7 genes from different RV species and strains. The P[7] and P[23] are the predominant genotypes circulating in pig farms, but no vaccine is available at present in China. Subunit vaccines, as nonreplicating vaccines, are an option to cope with variable genotypes. Here, we have developed a bivalent subunit candidate vaccine based on a truncated VP4 protein, which induced robust humoral and cellular immune responses and protected piglets against challenge with homologous PoRV. It also appears to be safe. These data show that the truncated VP4-protein-based subunit vaccine is a promising candidate for the prevention of PoRV diarrhea.

Methylated derivatives of polycyclic aromatic hydrocarbons in road dust, green belt soil and parking lot dust: occurrence, spatial distribution and emission sources.

Convenient transportation facilities not only bring the higher standard of living to big cities, but also bring some environmental pollution problems. In order to understand the presence and sources of methylated polycyclic aromatic hydrocarbons (Me-PAHs) in environmental samples and their association with total organic carbon (TOC), 49 Me-PAHs were analyzed in road dust, green belt soil and parking lot dust samples in Harbin. The results showed that the ranges of the total Me-PAHs (ΣMe-PAHs) content in road dust were 221-5826 ng/g in autumn and 697-7302 ng/g in spring, and those in green belt soil were 170-2509 ng/g and 155-9215 ng/g in autumn and spring, respectively. And ΣMe-PAHs content in parking lot dust ranged from 269 to 2515 ng/g in surface parking lots and from 778 to 10,052 ng/g in underground parking lots. In these samples, the composition profile of Me-PAHs was dominated by 4-ring Me-PAHs. The results of diagnostic ratios and principal component analysis (PCA) indicated that petrogenic and pyrogenic sources were the main sources of Me-PAHs in the samples. Spearman correlation analysis showed that there was no correlation for Me-PAHs in road dust and green belt soil on the same road. Furthermore, there was a significant positive relationship (0.12 ≤ R2 ≤ 0.67, P < 0.05) between Me-PAHs concentrations and the TOC content. This study demonstrated the presence of Me-PAHs with high concentrations in the road environment samples of Harbin.

Inhalation of 2, 4-di-tert-butylphenol-Loaded micelles suppresses respiratory syncytial virus infection in mice.

Human respiratory syncytial virus (RSV) is a common cause of respiratory infections in infants, young children, and elderly people. However, there are no effective treatments or vaccines available in most countries. In this study, we explored the anti-RSV potential of 2, 4-Di-tert-butylphenol (2, 4-DTBP), a compound derived from Houttuynia cordata Thunb. To overcome the poor solubility of 2, 4-DTBP, we encapsulated it in polymeric micelles and delivered it by inhalation. We found that 2, 4-DTBP-loaded micelles inhibited RSV infection in vitro and improved survival, lung pathology, and viral clearance in RSV-infected mice. Our results suggested that 2, 4-DTBP-loaded micelle is a promising novel therapeutic agent for RSV infection.

De novo diploid genome assembly using long noisy reads.

The high sequencing error rate has impeded the application of long noisy reads for diploid genome assembly. Most existing assemblers failed to generate high-quality phased assemblies using long noisy reads. Here, we present PECAT, a Phased Error Correction and Assembly Tool, for reconstructing diploid genomes from long noisy reads. We design a haplotype-aware error correction method that can retain heterozygote alleles while correcting sequencing errors. We combine a corrected read SNP caller and a raw read SNP caller to further improve the identification of inconsistent overlaps in the string graph. We use a grouping method to assign reads to different haplotype groups. PECAT efficiently assembles diploid genomes using Nanopore R9, PacBio CLR or Nanopore R10 reads only. PECAT generates more contiguous haplotype-specific contigs compared to other assemblers. Especially, PECAT achieves nearly haplotype-resolved assembly on B. taurus (Bison×Simmental) using Nanopore R9 reads and phase block NG50 with 59.4/58.0 Mb for HG002 using Nanopore R10 reads.

Deep Imputation Bi-stochastic Graph Regularized Matrix Factorization for Clustering Single-cell RNA-sequencing Data.

By generating massive gene transcriptome data and analyzing transcriptomic variations at the cell level, single-cell RNA-sequencing (scRNA-seq) technology has provided new way to explore cellular heterogeneity and functionality. Clustering scRNA-seq data could discover the hidden diversity and complexity of cell populations, which can aid to the identification of the disease mechanisms and biomarkers. In this paper, a novel method (DSINMF) is presented for single cell RNA sequencing data by using deep matrix factorization. Our proposed method comprises four steps: first, the feature selection is utilized to remove irrelevant features. Then, the dropout imputation is used to handle missing value problem. Further, the dimension reduction is employed to preserve data characteristics and reduce noise effects. Finally, the deep matrix factorization with bi-stochastic graph regularization is used to obtain cluster results from scRNA-seq data. We compare DSINMF with other state-of-the-art algorithms on nine datasets and the results show our method outperformances than other methods.

DTKGIN: Predicting drug-target interactions based on knowledge graph and intent graph.

Knowledge graph intent graph attention mechanism Predicting drug-target interactions (DTIs) plays a crucial role in drug discovery and drug development. Considering the high cost and risk of biological experiments, developing computational approaches to explore the interactions between drugs and targets can effectively reduce the time and cost of drug development. Recently, many methods have made significant progress in predicting DTIs. However, existing approaches still suffer from the high sparsity of DTI datasets and the cold start problem. In this paper, we develop a new model to predict drug-target interactions via a knowledge graph and intent graph named DTKGIN. Our method can effectively capture biological environment information for targets and drugs by mining their associated relations in the knowledge graph and considering drug-target interactions at a fine-grained level in the intent graph. DTKGIN learns the representation of drugs and targets from the knowledge graph and the intent graph. Then the probabilities of interactions between drugs and targets are obtained through the inner product of the representation of drugs and targets. Experimental results show that our proposed method outperforms other state-of-the-art methods in 10-fold cross-validation, especially in cold-start experimental settings. Furthermore, the case studies demonstrate the effectiveness of DTKGIN in predicting potential drug-target interactions. The code is available on GitHub: https://github.com/Royluoyi123/DTKGIN.

Establishing Salvia miltiorrhiza-Derived Exosome-like Nanoparticles and Elucidating Their Role in Angiogenesis.

Exosomes are multifunctional, cell-derived nanoscale membrane vesicles. Exosomes derived from certain mammalian cells have been developed as angiogenesis promoters for the treatment of myocardial ischemia-reperfusion injury, as they possess the capability to enhance endothelial cell proliferation, migration, and angiogenesis. However, the low yield of exosomes derived from mammalian cells limits their clinical applications. Therefore, we chose to extract exosome-like nanoparticles from the traditional Chinese medicine Salvia miltiorrhiza, which has been shown to promote angiogenesis. Salvia miltiorrhiza-derived exosome-like nanoparticles offer advantages, such as being economical, easily obtainable, and high-yielding, and have an ideal particle size, Zeta potential, exosome-like morphology, and stability. Salvia miltiorrhiza-derived exosome-like nanoparticles can enhance the cell viability of Human Umbilical Vein Endothelial Cells and can promote cell migration and improve the neovascularization of the cardiac tissues of myocardial ischemia-reperfusion injury, indicating their potential as angiogenesis promoters for the treatment of myocardial ischemia-reperfusion injury.

Indirubin-3'-monoxime exhibits potent antiviral and anti-inflammatory effects against human adenoviruses in vitro and in vivo.

Human adenovirus (HAdV) infection is a major cause of respiratory disease, yet no antiviral drugs have been approved for its treatment. Herein, we evaluated the antiviral and anti-inflammatory effects of cyclin-dependent protein kinase (CDK) inhibitor indirubin-3'-monoxime (IM) against HAdV infection in cells and a transgenic mouse model. After evaluating its cytotoxicity, cytopathic effect reduction, antiviral replication kinetics, and viral yield reduction assays were performed to assess the anti-HAdV activity of IM. Quantitative real-time polymerase chain reaction (qPCR), quantitative reverse transcription PCR (qRT-PCR), and western blotting were used to assess the effects of IM on HAdV DNA replication, transcription, and protein expression, respectively. IM significantly inhibited HAdV DNA replication as well as E1A and Hexon transcription, in addition to significantly suppressing the phosphorylation of the RNA polymerase II C-terminal domain (CTD). IM mitigated body weight loss, reduced viral burden, and lung injury, decreasing cytokine and chemokine secretion to a greater extent than cidofovir. Altogether, IM inhibits HAdV replication by downregulating CTD phosphorylation to suppress viral infection and corresponding innate immune reactions as a promising therapeutic agent.

Aniline antioxidants in road dust, parking lot dust, and green-belt soil in Harbin, a megacity in China: Occurrence, profile, and seasonal variation.

Aniline antioxidants (ANs) are widely used as industrial chemicals in products composed of rubber. ANs originate mainly from vehicles, where tire wear particles end up in dust and soil after being deposited on roads. Nowadays, limited information is available on the fate and seasonal variation of ANs in the road environment. In this study, we investigated the occurrence of 32 ANs in dust and soil from different road environments, including road dust, garage dust, parking lot dust, and green-belt soil. The total concentrations of ANs were 369 ng g-1 in road dust, 712 ng g-1 in garage dust, and 687 ng g-1 in parking lot dust. These concentrations are several times higher than that in green-belt soil (128 ng g-1). The highest concentrations of N-(1,3-dimethylbutyl)-N'-phenyl-1,4-phenylenediamine (6PPD) were found in dust and soil. Furthermore, notable seasonal differences were observed, with significantly higher concentrations of ANs in autumn than those in spring. In the main urban area, roads with high traffic volume exhibited higher concentrations of ANs than those with low traffic volume, indicating that ANs were produced by vehicle-related sources.

MA-PEP: A novel anticancer peptide prediction framework with multimodal feature fusion based on attention mechanism.

AntiCancer Peptides (ACPs) have emerged as promising therapeutic agents for cancer treatment. The time-consuming and costly nature of wet-lab discriminatory methods has spurred the development of various machine learning and deep learning-based ACP classification methods. Nonetheless, current methods encountered challenges in efficiently integrating features from various peptide modalities, thereby limiting a more comprehensive understanding of ACPs and further restricting the improvement of prediction model performance. In this study, we introduce a novel ACP prediction method, MA-PEP, which leverages multiple attention mechanisms for feature enhancement and fusion to improve ACP prediction. By integrating the enhanced molecular-level chemical features and sequence information of peptides, MA-PEP demonstrates superior prediction performance across several benchmark datasets, highlighting its efficacy in ACP prediction. Moreover, the visual analysis and case studies further demonstrate MA-PEP's reliable feature extraction capability and its promise in the realm of ACP exploration. The code and datasets for MA-PEP are available at https://github.com/liangxiaodata/MA-PEP.

GraphPath: a graph attention model for molecular stratification with interpretability based on the pathway-pathway interaction network.

MOTIVATION: Studying the molecular heterogeneity of cancer is essential for achieving personalized therapy. At the same time, understanding the biological processes that drive cancer development can lead to the identification of valuable therapeutic targets. Therefore, achieving accurate and interpretable clinical predictions requires paramount attention to thoroughly characterizing patients at both the molecular and biological pathway levels. RESULTS: Here, we present GraphPath, a biological knowledge-driven graph neural network with multi-head self-attention mechanism that implements the pathway-pathway interaction network. We train GraphPath to classify the cancer status of patients with prostate cancer based on their multi-omics profiling. Experiment results show that our method outperforms P-NET and other baseline methods. Besides, two external cohorts are used to validate that the model can be generalized to unseen samples with adequate predictive performance. We reduce the dimensionality of latent pathway embeddings and visualize corresponding classes to further demonstrate the optimal performance of the model. Additionally, since GraphPath's predictions are interpretable, we identify target cancer-associated pathways that significantly contribute to the model's predictions. Such a robust and interpretable model has the potential to greatly enhance our understanding of cancer's biological mechanisms and accelerate the development of targeted therapies. AVAILABILITY AND IMPLEMENTATION: https://github.com/amazingma/GraphPath.