AACFlow: an end-to-end model based on attention augmented convolutional neural network and flow-attention mechanism for identification of anticancer peptides.

MOTIVATION: Anticancer peptides (ACPs) have natural cationic properties and can act on the anionic cell membrane of cancer cells to kill cancer cells. Therefore, ACPs have become a potential anticancer drug with good research value and prospect. RESULTS: In this article, we propose AACFlow, an end-to-end model for identification of ACPs based on deep learning. End-to-end models have more room to automatically adjust according to the data, making the overall fit better and reducing error propagation. The combination of attention augmented convolutional neural network (AAConv) and multi-layer convolutional neural network (CNN) forms a deep representation learning module, which is used to obtain global and local information on the sequence. Based on the concept of flow network, multi-head flow-attention mechanism is introduced to mine the deep features of the sequence to improve the efficiency of the model. On the independent test dataset, the ACC, Sn, Sp, and AUC values of AACFlow are 83.9%, 83.0%, 84.8%, and 0.892, respectively, which are 4.9%, 1.5%, 8.0%, and 0.016 higher than those of the baseline model. The MCC value is 67.85%. In addition, we visualize the features extracted by each module to enhance the interpretability of the model. Various experiments show that our model is more competitive in predicting ACPs.

Molecular Insights into the Inhibition and Disaggregation Effects of EGCG on Aß40 and Aß42 Cofibrillation.

The misfolding and aggregation of amyloid-ß (Aß) peptides play a pivotal role in the pathogenesis of Alzheimer's disease (AD). Aß40 and Aß42, the two primary isoforms of Aß, can not only self-aggregate into homogeneous aggregates but also coaggregate to form mixed fibrils. Epigallocatechin-3-gallate (EGCG), a prominent tea polyphenol, has shown the capability to prevent the self-aggregation of Aß40 and Aß42 peptides and disaggregate their homogeneous fibrils. However, its effects on the cofibrillation of Aß40 and Aß42 have not yet been explored. Here, we employed molecular dynamic simulations to investigate the effects of EGCG on the coaggregation of Aß40 and Aß42, as well as on their mixed fibril. Our findings indicated that EGCG effectively inhibits the codimerization of Aß40 and Aß42 primarily by impeding the interchain interaction between the two isoforms. The key binding sites for EGCG on Aß40 and Aß42 are the polar residues and aromatic residues, engaging in hydrogen-bond , π-π, and cation-π interactions with EGCG. Additionally, EGCG disaggregates the Aß40-Aß42 mixed fibril by reducing its long-range interaction through similar binding sites and interactions as those between EGCG and Aß40-Aß42 heterodimers. Our research reveals the comprehensive inhibition and disaggregation effects of EGCG on the cofibrillation of Aß isoforms, which provides further support for the development of EGCG as an effective antiaggregation agent for AD.

The Inhibition Effect of Epigallocatechin-3-Gallate on the Co-Aggregation of Amyloid-ß and Human Islet Amyloid Polypeptide Revealed by Replica Exchange Molecular Dynamics Simulations.

Alzheimer's disease and Type 2 diabetes are two epidemiologically linked diseases which are closely associated with the misfolding and aggregation of amyloid proteins amyloid-ß (Aß) and human islet amyloid polypeptide (hIAPP), respectively. The co-aggregation of the two amyloid proteins is regarded as the fundamental molecular mechanism underlying their pathological association. The green tea extract epigallocatechin-3-gallate (EGCG) has been extensively demonstrated to inhibit the amyloid aggregation of Aß and hIAPP proteins. However, its potential role in amyloid co-aggregation has not been thoroughly investigated. In this study, we employed the enhanced-sampling replica exchange molecular dynamics simulation (REMD) method to investigate the effect of EGCG on the co-aggregation of Aß and hIAPP. We found that EGCG molecules substantially diminish the ß-sheet structures within the amyloid core regions of Aß and hIAPP in their co-aggregates. Through hydrogen-bond, π-π and cation-π interactions targeting polar and aromatic residues of Aß and hIAPP, EGCG effectively attenuates both inter-chain and intra-chain interactions within the co-aggregates. All these findings indicated that EGCG can effectively inhibit the co-aggregation of Aß and hIAPP. Our study expands the potential applications of EGCG as an anti-amyloidosis agent and provides therapeutic options for the pathological association of amyloid misfolding disorders.

Multi-scale characterization and analysis of cellular viscoelastic mechanical phenotypes by atomic force microscopy.

The viscoelasticity of cells serves as a biomarker that reveals changes induced by malignant transformation, which aids the cytological examinations. However, differences in the measurement methods and parameters have prevented the consistent and effective characterization of the viscoelastic phenotype of cells. To address this issue, nanomechanical indentation experiments were conducted using an atomic force microscope (AFM). Multiple indentation methods were applied, and the indentation parameters were gradually varied to measure the viscoelasticity of normal liver cells and cancerous liver cells to create a database. This database was employed to train machine-learning algorithms in order to analyze the differences in the viscoelasticity of different types of cells and as well as to identify the optimal measurement methods and parameters. These findings indicated that the measurement speed significantly influenced viscoelasticity and that the classification difference between the two cell types was most evident at 5 µm/s. In addition, the precision and the area under the receiver operating characteristic curve were comparatively analyzed for various widely employed machine-learning algorithms. Unlike previous studies, this research validated the effectiveness of measurement parameters and methods with the assistance of machine-learning algorithms. Furthermore, the results confirmed that the viscoelasticity obtained from the multiparameter indentation measurement could be effectively used for cell classification. RESEARCH HIGHLIGHTS: This study aimed to analyze the viscoelasticity of liver cancer cells and liver cells. Different nano-indentation methods and parameters were used to measure the viscoelasticity of the two kinds of cells. The neural network algorithm was used to reverse analyze the dataset, and the methods and parameters for accurate classification and identification of cells are successfully found.

Longitudinal Distribution Map of the Active Components and Endophytic Fungi in Angelica sinensis (Oliv.) Diels Root and Their Potential Correlations.

The three distinct medicinal parts of Angelica sinensis (Oliv.) Diels (Ang) roots are the head, body, and tail (ARH, ARB, and ART, respectively). How endophytic fungi shape the differences in metabolic components among these parts remains unclear. We quantified the distribution of active components and endophytic fungi along the ARH, ARB, and ART and their relationships. Based on the metabolic components and their abundances detected via non-target metabolism, the different medicinal parts were distinguishable. The largest number of dominant metabolic components was present in ART. The difference between ART and ARH was the greatest, and ARB was in a transitional state. The dominant active molecules in ART highlight their effects in haemodynamics improvement, antibacterial, anti-inflammatory, and hormone regulation, while ARH and ARB indicated more haemostasis, blood enrichment, neuromodulation, neuroprotection and tranquilisation, hepatoprotection, and antitumour activities than that of ART. The ARHs, ARBs, and ARTs can also be distinguished from each other based on the endophytic fungi at the microbiome level. The most dominant endophytic fungi were distributed in ART; the differences between ART and ARH were the largest, and ARB was in a transition state, which is consistent with the metabolite distributions. Structural equation modelling showed that the endophytic fungi were highly indicative of the metabolic components. Correlation analysis further identified the endophytic fungi significantly positively correlated with important active components, including Condenascus tortuosus, Sodiomyces alcalophilus, and Pleotrichocladium opacum. The bidirectional multivariate interactions between endophytic fungi and the metabolic components shape their spatial variations along the longitudinal direction in the Ang root.

Effect of curcumin on malignant hepatocytes and mitochondria studied using atomic force microscopy.

Mitochondria are emerging as potential targets for the cancer treatment. In this study, the effects of curcumin on the activity, migration, and mitochondrial membrane potential (MMP) of malignant hepatocytes (SMMC-7721 cells) were determined using cell viability, migration, and MMP assays. Changes in the morphology and biomechanics of SMMC-7721 cells and their mitochondria were studied using both optical microscopy and atomic force microscopy (AFM). The cell survival rate, migration and MMP depended on the concentration of curcumin. Optical microscopy studies showed that curcumin altered the cell morphology. AFM studies showed that the changes in the morphology and nanomechanics of SMMC-7721 cells and their mitochondria, were induced by curcumin. As the concentration of curcumin increased, the cell length, width, and adhesion decreased, but the height, roughness and Young's modulus increased. In contrast, the mitochondrial length, width, height and roughness increased, but the adhesion and Young's modulus decreased. There was a close relationship between mitochondria and cells in terms of function, morphology and biomechanics. This study shows the effects of curcumin on SMMC-7721 cells and their mitochondria from biology and biophysics perspectives. The findings aid in comprehensively understanding the interactions between mitochondria and malignant hepatocytes.

iAVPs-ResBi: Identifying antiviral peptides by using deep residual network and bidirectional gated recurrent unit.

Human history is also the history of the fight against viral diseases. From the eradication of viruses to coexistence, advances in biomedicine have led to a more objective understanding of viruses and a corresponding increase in the tools and methods to combat them. More recently, antiviral peptides (AVPs) have been discovered, which due to their superior advantages, have achieved great impact as antiviral drugs. Therefore, it is very necessary to develop a prediction model to accurately identify AVPs. In this paper, we develop the iAVPs-ResBi model using k-spaced amino acid pairs (KSAAP), encoding based on grouped weight (EBGW), enhanced grouped amino acid composition (EGAAC) based on the N5C5 sequence, composition, transition and distribution (CTD) based on physicochemical properties for multi-feature extraction. Then we adopt bidirectional long short-term memory (BiLSTM) to fuse features for obtaining the most differentiated information from multiple original feature sets. Finally, the deep model is built by combining improved residual network and bidirectional gated recurrent unit (BiGRU) to perform classification. The results obtained are better than those of the existing methods, and the accuracies are 95.07, 98.07, 94.29 and 97.50% on the four datasets, which show that iAVPs-ResBi can be used as an effective tool for the identification of antiviral peptides. The datasets and codes are freely available at https://github.com/yunyunliang88/iAVPs-ResBi.

Effects of targeted lung cancer drugs on cardiomyocytes studied by atomic force microscopy.

The epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKIs) has become one of the important targeted drugs for the treatment of non-small cell lung cancer (NSCLC). But the cardiac adverse events (AEs) related to the EGFR-TKI treatment occur frequently. And the cases of TKI-associated cardiac AEs remain poorly understood. In order to study the effects of EGFR-TKIs on cardiomyocytes, atomic force microscopy (AFM) was used to measure and analyze the physical properties of cardiomyocytes under the actions of three drugs (gefitinib, afatinib and osimertinib) with different concentrations. By comparing the height, adhesion, Young's modulus, the amplitude and the time of the contraction and relaxation process, it was found that the changes of the mechanical properties of cells were well correlated with the symptoms of AEs, such as cardiomyocyte hypertrophy, QT prolongation, atrial fibrillation, ejection fraction reductions, and cardiac failure. In addition, osimertinib has the most obvious effect on cardiomyocytes at a low concentration, and gefitinib has the greatest effect with the increase of concentration, while afatinib has the least effect on cardiomyocytes. This provides a new method for screening drugs and exploring the principle of action in the process of cancer treatment at the cellular level.

Genomic properties and clinical outcomes associated with tertiary lymphoid structures in patients with breast cancer.

Tertiary lymphoid structures (TLSs) play a crucial role in determining prognosis and response to immunotherapy in several solid malignancies. Nevertheless, the effect of TLS-associated gene signature based on The Cancer Genome Atlas (TCGA) cohort in patients with breast cancer (BRCA) remains controversial. Based on TCGA-BRCA dataset (n = 866), 9-gene was identified to construct an TLS signature and further analyzed its prognostic value. Then, we explored the relationship of this TLS signature with molecular subtype, immune microenvironment, tumor mutational burden (TMB). High-TLS signature patients had a better overall survival (OS) than low-TLS signature patients, consistent with the results in the METABRIC cohort. Multivariate analysis revealed that TLS signature remained an independent prognostic indicator for OS. In addition, we established a nomogram with the integration of TLS signature and other independent variables to predict individual risk of death. The comprehensive results showed that patients with high TLS signature benefit from immunotherapy; the signature was also correlated with inhibition of cell proliferation pathways, low TP53 mutation rate, high infiltration of B cells, CD8 + T cells, CD4 + T cells, and M1 macrophages. Therefore, TLS signature is a promising biomarker to distinguish the prognosis and immune microenvironment in BRCA.

Detection and classification of hepatocytes and hepatoma cells using atomic force microscopy and machine learning algorithms.

Hepatocellular carcinoma is a high-risk malignant tumor. Hepatoma cells are transformed from normal cells and have unique surface nanofeatures in addition to the characteristics of the original cells. In this paper, atomic force microscopy was used to extract the three-dimensional morphology and mechanical information of HL-7702 human hepatocytes and SMMC-7721 and HepG2 hepatoma cells in culture, such as the elastic modulus and viscoelasticity. The characteristics of different cells were compared and analyzed. Finally, the cell morphology and mechanics information were used for training machine learning algorithms. With the trained model, the detection of cells was realized. The classification accuracy was as high as 94.54%, and the area under the receiver operating characteristic (ROC) curve (AUC) was 0.99. Thus, hepatocytes and hepatoma cells were accurately identified and assessed. We also compared the classification effects of other machine learning algorithms, such as support vector machine and logistic regression. Our method extracts cellular nanofeatures directly from the surface of cells of unknown type for cell classification. Compared with microscope image-based analysis and other methods, this approach can avoid the misjudgment that may occur when different doctors have different levels of experience. Thus, the proposed method provides an objective basis for the early diagnosis of hepatocellular carcinoma. RESEARCH HIGHLIGHTS: The 3D appearance and mechanical characteristics of hepatocellular carcinoma cells are very similar to those of hepatocytes. Application of atomic force microscopy with machine learning algorithm. Collect the data set of nano-characteristic parameters of the cell. The machine learning algorithms is trained by data set, and its classification effect is better than that of a single nano-parameter.

The characterization of antimicrobial nanocomposites based on chitosan, cinnamon essential oil, and TiO2 for fruits preservation.

The freshness and safety of fruits have always been crucial issues in the development of the industry. However, the existing fresh-keeping methods have limited effect, meanwhile, the preservation mechanism of different materials. In this study, Cinnamon essential oil (CEO), TiO2, and chitosan (CS) were compounded to prepare safe and renewable nanocomposites (CS-T-C) for fruit preservation. The results showed that CEO mainly destroyed the bacterial cell wall through penetration, while TiO2 is through destruction. The strawberry coated with CS-T-C showed better hardness, lower weight loss and mildew rate, and the shelf-life at 20℃ was extended for four days compared with the control. And all four nanocomposites were not cytotoxic. In summary, nanocomposites can deal with many problems through different mechanisms to maximize the fresh-keeping effect, and the nanocomposites developed in this work might be a good choice for fruit preservation.

DapNet-HLA: Adaptive dual-attention mechanism network based on deep learning to predict non-classical HLA binding sites.

Human leukocyte antigen (HLA) plays a vital role in immunomodulatory function. Studies have shown that immunotherapy based on non-classical HLA has essential applications in cancer, COVID-19, and allergic diseases. However, there are few deep learning methods to predict non-classical HLA alleles. In this work, an adaptive dual-attention network named DapNet-HLA is established based on existing datasets. Firstly, amino acid sequences are transformed into digital vectors by looking up the table. To overcome the feature sparsity problem caused by unique one-hot encoding, the fused word embedding method is used to map each amino acid to a low-dimensional word vector optimized with the training of the classifier. Then, we use the GCB (group convolution block), SENet attention (squeeze-and-excitation networks), BiLSTM (bidirectional long short-term memory network), and Bahdanau attention mechanism to construct the classifier. The use of SENet can make the weight of the effective feature map high, so that the model can be trained to achieve better results. Attention mechanism is an Encoder-Decoder model used to improve the effectiveness of RNN, LSTM or GRU (gated recurrent neural network). The ablation experiment shows that DapNet-HLA has the best adaptability for five datasets. On the five test datasets, the ACC index and MCC index of DapNet-HLA are 4.89% and 0.0933 higher than the comparison method, respectively. According to the ROC curve and PR curve verified by the 5-fold cross-validation, the AUC value of each fold has a slight fluctuation, which proves the robustness of the DapNet-HLA. The codes and datasets are accessible at https://github.com/JYY625/DapNet-HLA.

Enhanced catalytic activity and stability of composite of cellulose film and nano zero-valent iron on Juncus effusus for activating peroxydisulfate to degrade Rhodamine B dye.

In this study, a novel peroxydisulfate (PDS) activator (CF-nZVI-JE) was prepared via in-situ loading nano zero-valent iron (nZVI) on Juncus effusus (JE) followed with wrapping a layer of cellulose film (CF). The CF-nZVI-JE had the same 3D structure as the JE, being easy to separate from aqueous solution. The loaded nZVI existed single nanoparticles with a size of 60-100 nm except chain-type agglomeration of nanoparticles due to the stabilization of JE fibers. The activation performance of the CF-nZVI-JE for PDS was evaluated with Rhodamine B (Rh B) as a representative pollutant. Under the optimal activating conditions, the degradation rate of Rh B reached 99% within 30 min in the CF-nZVI-JE/PDS system. After five cycles, the degradation rate of Rh B was still over 85%, suggesting that the CF-nZVI-JE had good reusability. More interestingly, SO4·- and ·OH radicals were simultaneously detected in the CF-nZVI-JE/PDS system, but only SO4·- existed in the JE-ZVI/PDS system, suggesting the different activation mechanism. Meanwhile, the introduction of CF not only facilitated to the mineralization of Rh B but also significantly reduced the release amount of iron ions. Hence, the CF-nZVI-JE can be employed as a promising PDS activator for the treatment of organic wastewater.

Glycine betaine inhibits postharvest softening and quality decline of winter jujube fruit by regulating energy and antioxidant metabolism.

Winter jujube fruit easily softens after harvest. To investigate the effects of glycine betaine (N,N,N-trimethylglycine; GB) treatment on the quality of postharvest jujubes, fresh winter jujubes (Zizyphus jujuba Mill. cv. Dongzao) were immersed in 20 mmol·L-1 GB for 20 min. The results showed that GB application can effectively maintain cell wall component content by restraining gene expression and enzyme activities, including PG, CX, PME and ß-Glu. Meanwhile, the activities of antioxidant enzymes (APX, CAT, SOD, POD) and the contents of nonenzymatic antioxidants (MDA, H2O2, ASA, GSH) were enhanced in treated jujubes, thereby reducing the content of ROS. In addition, energy metabolism enzyme activities (H+-ATPase, Ca2+-ATPase, SDH and CCO) and gene expression were also significantly increased, thus maintaining higher energy levels (ATP, ADP, AMP and EC). In summary, GB enhances ATP biosynthesis by increasing energy metabolism. It offers essential energy for the antioxidant metabolism, thus retarding the softening of postharvest jujubes.

Markov State Models Underlying the N-Terminal Premodel of TOPK/PBK.

Lymphokine-activated killer T-cell-originated protein kinase (TOPK) is a potential target for cancer therapy. To explore the micromechanism, we proposed the N-terminal premodel (NTPM) of the TOPK monomer via homology modeling and molecular dynamic simulations and analyzed the conformational dynamics by Markov state model analysis. The electronegative insert (ENI) motif of the NTPM can be opened with a small probability under wild type, regulated by the so-called "N-C" interaction zone consisting of the N-terminal head, the coil between ß3-strand and αC-helix, and the ENI motif. Glutamate substitution at threonine residue 9 or tyrosine residue 74 promotes the closed-open transition, revealing the details of phosphorylation. Allosteric effects induce functionally relevant structural changes, such as increased structural flexibility and active sites, which are thought to be necessary for further activation or binding. These findings provide rational structural templates for designing state-dependent inhibitors and give insight into the molecular regulatory mechanisms of TOPK monomers.

Investigating the evolution process of lung adenocarcinoma via random walk and dynamic network analysis.

Lung adenocarcinoma (LUAD) is a typical disease regarded as having multi-stage progression. However, many existing methods often ignore the critical differences among these stages, thereby limiting their effectiveness for discovering key biological molecules and biological functions as signals at each stage. In this study, we propose a method to discover the evolution between biological molecules and biological functions by investigating the multi-stage biological molecules of LUAD. The method is based on the random walk algorithm and the Monte Carlo method to generate clusters as the modules, which were used as subgraphs of the differentiated biological molecules network in each stage. The connection between modules of adjacent stages is based on the measurement of the Jaccard coefficient. The online gene set enrichment analysis tool (DAVID) was used to obtain biological functions corresponding to the individual important modules. The core evolution network was constructed by combining the aforementioned two networks. Since the networks here are all dynamic, we also propose a strategy to visualize the dynamic information together in one network. Eventually, 12 core modules and 11 core biological functions were found through such evolutionary analyses. Among the core biological functions that we obtained, six functions are related to the disease, the biological function of neutrophil chemotaxis is not directly associated with LUAD but can serve as a predictor, two functions may serve as a predictive signal, and two functions need to be verified through more biological evidence. Compared with two alternative design methods, the method proposed in this study performed more efficiently.

Identification of Colon Cancer-Related RNAs Based on Heterogeneous Networks and Random Walk.

Colon cancer is considered as a complex disease that consists of metastatic seeding in early stages. Such disease is not simply caused by the action of a single RNA, but is associated with disorders of many kinds of RNAs and their regulation relationships. Hence, it is of great significance to study the complex regulatory roles among mRNAs, miRNAs and lncRNAs for further understanding the pathogenic mechanism of colon cancer. In this study, we constructed a heterogeneous network consisting of differentially expressed mRNAs, miRNAs and lncRNAs. This contains three kinds of vertices and six types of edges. All RNAs were re-divided into three categories, which were "related", "irrelevant" and "unlabeled". They were processed by dynamic excitation restart random walk (RW-DIR) for identifying colon cancer-related RNAs. Ten RNAs were finally obtained related to colon cancer, which were hsa-miR-2682-5p, hsa-miR-1277-3p, ANGPTL1, SLC22A18AS, FENDRR, PHLPP2, hsa-miR-302a-5p, APCDD1, MEX3A and hsa-miR-509-3-5p. Numerical experiments have indicated that the proposed network construction framework and the following RW-DIR algorithm are effective for identifying colon cancer-related RNAs, and this kind of analysis framework can also be easily extended to other diseases, effectively narrowing the scope of biological experimental research.

Nanopore-based technologies beyond DNA sequencing.

Inspired by the biological processes of molecular recognition and transportation across membranes, nanopore techniques have evolved in recent decades as ultrasensitive analytical tools for individual molecules. In particular, nanopore-based single-molecule DNA/RNA sequencing has advanced genomic and transcriptomic research due to the portability, lower costs and long reads of these methods. Nanopore applications, however, extend far beyond nucleic acid sequencing. In this Review, we present an overview of the broad applications of nanopores in molecular sensing and sequencing, chemical catalysis and biophysical characterization. We highlight the prospects of applying nanopores for single-protein analysis and sequencing, single-molecule covalent chemistry, clinical sensing applications for single-molecule liquid biopsy, and the use of synthetic biomimetic nanopores as experimental models for natural systems. We suggest that nanopore technologies will continue to be explored to address a number of scientific challenges as control over pore design improves.

Organelle Interaction and Drug Discovery: Towards Correlative Nanoscopy and Molecular Dynamics Simulation.

The inter-organelle interactions, including the cytomembrane, endoplasmic reticulum, mitochondrion, lysosome, dictyosome, and nucleus, play the important roles in maintaining the normal function and homeostasis of cells. Organelle dysfunction can lead to a range of diseases (e.g., Alzheimer's disease (AD), Parkinson's disease (PD), and cancer), and provide a new perspective for drug discovery. With the development of imaging techniques and functional fluorescent probes, a variety of algorithms and strategies have been developed for the ever-improving estimation of subcellular structures, organelle interaction, and organelle-related drug discovery with accounting for the dynamic structures of organelles, such as the nanoscopy technology and molecular dynamics (MD) simulations. Accordingly, this work summarizes a series of state-of-the-art examples of the recent progress in this rapidly changing field and uncovering the drug screening based on the structures and interactions of organelles. Finally, we propose the future outlook for exciting applications of organelle-related drug discovery, with the cooperation of nanoscopy and MD simulations.

Accurate Prediction of Anti-hypertensive Peptides Based on Convolutional Neural Network and Gated Recurrent unit.

Hypertension (HT) is a general disease, and also one of the most ordinary and major causes of cardiovascular disease. Some diseases are caused by high blood pressure, including impairment of heart and kidney function, cerebral hemorrhage and myocardial infarction. Due to the limitations of laboratory methods, bioactive peptides for the treatment of HT need a long time to be identified. Therefore, it is of great immediate significance for the identification of anti-hypertensive peptides (AHTPs). With the prevalence of machine learning, it is suggested to use it as a supplementary method for AHTPs classification. Therefore, we develop a new model to identify AHTPs based on multiple features and deep learning. And the deep model is constructed by combining a convolutional neural network (CNN) and a gated recurrent unit (GRU). The unique convolution structure is used to reduce the feature dimension and running time. The data processed by CNN is input into the recurrent structure GRU, and important information is filtered out through the reset gate and update gate. Finally, the output layer adopts Sigmoid activation function. Firstly, we use Kmer, the deviation between the dipeptide frequency and the expected mean (DDE), encoding based on grouped weight (EBGW), enhanced grouped amino acid composition (EGAAC) and dipeptide binary profile and frequency (DBPF) to extract features. For Kmer, DDE, EBGW and EGAAC, it is widely used in the field of protein research. DBPF is a new feature representation method designed by us. It corresponds dipeptides to binary numbers, and finally obtains a binary coding file and a frequency file. Then these features are spliced together and input into our proposed model for prediction and analysis. After a tenfold cross-validation test, this model has a better competitive advantage than the previous methods, and the accuracy is 96.23% and 99.10%, respectively. From the results, compared with the previous methods, it has been greatly improved. It shows that the combination of convolution calculation and recurrent structure has a positive impact on the classification of AHTPs. The results show that this method is a feasible, efficient and competitive sequence analysis tool for AHTPs. Meanwhile, we design a friendly online prediction tool and it is freely accessible at http://ahtps.zhanglab.site/ .