APEX-pHLA: A novel method for accurate prediction of the binding between exogenous short peptides and HLA class I molecules.

Human leukocyte antigen (HLA) molecules play critically significant role within the realm of immunotherapy due to their capacities to recognize and bind exogenous antigens such as peptides, subsequently delivering them to immune cells. Predicting the binding between peptides and HLA molecules (pHLA) can expedite the screening of immunogenic peptides and facilitate vaccine design. However, traditional experimental methods are time-consuming and inefficient. In this study, an efficient method based on deep learning was developed for predicting peptide-HLA binding, which treated peptide sequences as linguistic entities. It combined the architectures of textCNN and BiLSTM to create a deep neural network model called APEX-pHLA. This model operated without limitations related to HLA class I allele variants and peptide segment lengths, enabling efficient encoding of sequence features for both HLA and peptide segments. On the independent test set, the model achieved Accuracy, ROC_AUC, F1, and MCC is 0.9449, 0.9850, 0.9453, and 0.8899, respectively. Similarly, on an external test set, the results were 0.9803, 0.9574, 0.8835, and 0.7863, respectively. These findings outperformed fifteen methods previously reported in the literature. The accurate prediction capability of the APEX-pHLA model in peptide-HLA binding might provide valuable insights for future HLA vaccine design.

Seroprevalence of Diphtheria and Tetanus Immunoglobulin G among the General Health Population in Guangzhou, China.

A seroepidemiological study was conducted in 2018 to assess diphtheria and tetanus antibodies in Guangzhou, China. Diphtheria and tetanus antibody concentrations were measured with an enzyme-linked immunosorbent assay. A total of 715 subjects were enrolled in the study. The overall diphtheria and tetanus toxoid IgG-specific antibody levels were 0.126 IU/mL (95% CI: 0.115, 0.137) and 0.210 IU/mL (95% CI: 0.185, 0.240), respectively; the overall positivity rate was 61.82% (95% CI: 58.14, 65.39) and 71.61% (95% CI: 68.3, 74.92), respectively. The diphtheria and tetanus antibody concentration was decreased by age and increased by doses. The geometric mean concentrations and positivity rate of diphtheria and tetanus antibodies were lowest and below the essential protection level in people over 14 years of age. Compared to children and adolescents, middle-aged people and the aged are at much higher risk of infection with Corynebacterium diphtheriae and Clostridium tetani. The current diphtheria and tetanus immunization schedule does not provide persistent protection after childhood. There is an urgent need to adjust the current immunization schedule.

Multi_CycGT: A Deep Learning-Based Multimodal Model for Predicting the Membrane Permeability of Cyclic Peptides.

Cyclic peptides are gaining attention for their strong binding affinity, low toxicity, and ability to target "undruggable" proteins; however, their therapeutic potential against intracellular targets is constrained by their limited membrane permeability, and researchers need much time and money to test this property in the laboratory. Herein, we propose an innovative multimodal model called Multi_CycGT, which combines a graph convolutional network (GCN) and a transformer to extract one- and two-dimensional features for predicting cyclic peptide permeability. The extensive benchmarking experiments show that our Multi_CycGT model can attain state-of-the-art performance, with an average accuracy of 0.8206 and an area under the curve of 0.8650, and demonstrates satisfactory generalization ability on several external data sets. To the best of our knowledge, it is the first deep learning-based attempt to predict the membrane permeability of cyclic peptides, which is beneficial in accelerating the design of cyclic peptide active drugs in medicinal chemistry and chemical biology applications.

Efficient Computational Framework for Target-Specific Active Peptide Discovery: A Case Study on IL-17C Targeting Cyclic Peptides.

The development of potentially active peptides for specific targets is critical for the modern pharmaceutical industry's growth. In this study, we present an efficient computational framework for the discovery of active peptides targeting a specific pharmacological target, which combines a conditional variational autoencoder (CVAE) and a classifier named TCPP based on the Transformer and convolutional neural network. In our example scenario, we constructed an active cyclic peptide library targeting interleukin-17C (IL-17C) through a library-based in vitro selection strategy. The CVAE model is trained on the preprocessed peptide data sets to generate potentially active peptides and the TCPP further screens the generated peptides. Ultimately, six candidate peptides predicted by the model were synthesized and assayed for their activity, and four of them exhibited promising binding affinity to IL-17C. Our study provides a one-stop-shop for target-specific active peptide discovery, which is expected to boost up the process of peptide drug development.

Generation of novel Diels-Alder reactions using a generative adversarial network.

Deep learning has enormous potential in the chemical and pharmaceutical fields, and generative adversarial networks (GANs) in particular have exhibited remarkable performance in the field of molecular generation as generative models. However, their application in the field of organic chemistry has been limited; thus, in this study, we attempt to utilize a GAN as a generative model for the generation of Diels-Alder reactions. A MaskGAN model was trained with 14 092 Diels-Alder reactions, and 1441 novel Diels-Alder reactions were generated. Analysis of the generated reactions indicated that the model learned several reaction rules in-depth. Thus, the MaskGAN model can be used to generate organic reactions and aid chemists in the exploration of novel reactions.

Transformer-based multitask learning for reaction prediction under low-resource circumstances.

Recently, effective and rapid deep-learning methods for predicting chemical reactions have significantly aided the research and development of organic chemistry and drug discovery. Owing to the insufficiency of related chemical reaction data, computer-assisted predictions based on low-resource chemical datasets generally have low accuracy despite the exceptional ability of deep learning in retrosynthesis and synthesis. To address this issue, we introduce two types of multitask models: retro-forward reaction prediction transformer (RFRPT) and multiforward reaction prediction transformer (MFRPT). These models integrate multitask learning with the transformer model to predict low-resource reactions in forward reaction prediction and retrosynthesis. Our results demonstrate that introducing multitask learning significantly improves the average top-1 accuracy, and the RFRPT (76.9%) and MFRPT (79.8%) outperform the transformer baseline model (69.9%). These results also demonstrate that a multitask framework can capture sufficient chemical knowledge and effectively mitigate the impact of the deficiency of low-resource data in processing reaction prediction tasks. Both RFRPT and MFRPT methods significantly improve the predictive performance of transformer models, which are powerful methods for eliminating the restriction of limited training data.

Virtual data augmentation method for reaction prediction.

To improve the performance of data-driven reaction prediction models, we propose an intelligent strategy for predicting reaction products using available data and increasing the sample size using fake data augmentation. In this research, fake data sets were created and augmented with raw data for constructing virtual training models. Fake reaction datasets were created by replacing some functional groups, i.e., in the data analysis strategy, the fake data as compounds with modified functional groups to increase the amount of data for reaction prediction. This approach was tested on five different reactions, and the results show improvements over other relevant techniques with increased model predictivity. Furthermore, we evaluated this method in different models, confirming the generality of virtual data augmentation. In summary, virtual data augmentation can be used as an effective measure to solve the problem of insufficient data and significantly improve the performance of reaction prediction.

From theory to experiment: transformer-based generation enables rapid discovery of novel reactions.

Deep learning methods, such as reaction prediction and retrosynthesis analysis, have demonstrated their significance in the chemical field. However, the de novo generation of novel reactions using artificial intelligence technology requires further exploration. Inspired by molecular generation, we proposed a novel task of reaction generation. Herein, Heck reactions were applied to train the transformer model, a state-of-art natural language process model, to generate 4717 reactions after sampling and processing. Then, 2253 novel Heck reactions were confirmed by organizing chemists to judge the generated reactions. More importantly, further organic synthesis experiments were performed to verify the accuracy and feasibility of representative reactions. The total process, from Heck reaction generation to experimental verification, required only 15 days, demonstrating that our model has well-learned reaction rules in-depth and can contribute to novel reaction discovery and chemical space exploration.

Self-Supervised Molecular Pretraining Strategy for Low-Resource Reaction Prediction Scenarios.

In the face of low-resource reaction training samples, we construct a chemical platform for addressing small-scale reaction prediction problems. Using a self-supervised pretraining strategy called MAsked Sequence to Sequence (MASS), the Transformer model can absorb the chemical information of about 1 billion molecules and then fine-tune on a small-scale reaction prediction. To further strengthen the predictive performance of our model, we combine MASS with the reaction transfer learning strategy. Here, we show that the average improved accuracies of the Transformer model can reach 14.07, 24.26, 40.31, and 57.69% in predicting the Baeyer-Villiger, Heck, C-C bond formation, and functional group interconversion reaction data sets, respectively, marking an important step to low-resource reaction prediction.

Reproducing the invention of a named reaction: zero-shot prediction of unseen chemical reactions.

While state-of-art models can predict reactions through the transfer learning of thousands of samples with the same reaction types as those of the reactions to predict, how to prepare such models to predict "unseen" reactions remains an unanswered question. We aimed to study the Transformer model's ability to predict "unseen" reactions through "zero-shot reaction prediction (ZSRP)", a concept derived from zero-shot learning and zero-shot translation. We reproduced the human invention of the Chan-Lam coupling reaction where the inventor was inspired by the Suzuki reaction when improving Barton's bismuth arylation reaction. After being fine-tuned with samples from these two "existing" reactions, the USPTO-trained Transformer could predict "unseen" Chan-Lam coupling reactions with 55.7% top-1 accuracy. Our model could also mimic the later stage of the history of this reaction, where the initial case of this reaction was generalized to more reactants and reagents via "one-shot/few-shot reaction prediction (OSRP/FSRP)" approaches.

A graph-convolutional neural network for addressing small-scale reaction prediction.

We describe a graph-convolutional neural network (GCN) model, the reaction prediction capabilities of which are as potent as those of the transformer model based on sufficient data, and we adopt the Baeyer-Villiger oxidation reaction to explore their performance differences based on limited data. The top-1 accuracy of the GCN model (90.4%) is higher than that of the transformer model (58.4%).

[The molecular epidemiology characteristics of norovirus in environment and clinical samples in Guangzhou from 2009 to 2011].

OBJECTIVE: To investigate the molecular epidemiological characteristics of norovirus in Guangzhou from 2009 to 2011. METHODS: A total of 183 water samples, 1162 seafood samples and 1066 diarrhea stool specimens were collected from January 2010 to May 2011, June 2009 to June 2011 and July 2009 to December 2010 respectively in Guangzhou. Norovirus was detected by real time reverse transcript-PCR (qRT-PCR). The partial polymerase gene was amplified from norovirus positive samples, then sequenced and compared with the sequences of norovirus in GenBank. The phylogenetic tree was created. RESULTS: The positive rate was 19.67% (36/183), 8.26% (96/1162) and 37.05% (395/1066) in water samples, seafood and diarrhea patients respectively. Noroviruses from positive samples could be divided into 10 representative strains, in which 7 representative strains of genotype of 208 samples was type G2-4. The sequences from water, seafood and stool specimens were highly homologous with the similarity of 94% - 100%. CONCLUSION: In Guangzhou, the predominant Norovirus genotype was G2-4 and the positive rate of samples was high.

Molecular epidemiological and virological study of dengue virus infections in Guangzhou, China, during 2001-2010.

BACKGROUND: Dengue virus (DENV) infection is the most prevalent arthropod-borne viral infection in tropical and subtropical regions worldwide. Guangzhou has the ideal environment for DENV transmission and DENV epidemics have been reported in this region for more than 30 years. METHODS: Information for DENV infection cases in Guangzhou from 2001 to 2010 were collected and analyzed. The DENV strains were cultured and isolated from patients' sera. Viral RNA was extracted from cell culture supernatants. cDNA was synthesized by reverse transcription PCR. Phylogenetic trees of four DENV serotypes were constructed respectively. RESULTS: In total, 2478 DENV infection cases were reported; 2143 of these (86.43%) occurred during 3 months of the year: August, September and October. Of these, 2398 were local cases (96.77%) and 80 were imported cases (3.23%). Among the imported cases, 69 (86.25%) were from Southeast Asian countries. From the 90 isolated strains, 66.67%, 3.33%, 14.44%, and 15.56% belonged to DENV serotypes 1, 2, 3, and 4, respectively. DENV-1 was predominant in most of the years, including during 2 outbreaks in 2002 and 2006; however, none of the strains or genotypes identified in this study were found to be predominant. Interestingly, DENV strains from different years had different origins. Moreover, the strains from each year belonged to different serotypes and/or genotypes. CONCLUSIONS: Southeast Asia countries were found to be the possible source of DENV in Guangzhou. These findings suggest that there is increasing diversity in DENV strains in Guangzhou, which could increase the risk of DENV outbreaks in the near future.

Antibody dynamics of 2009 influenza A (H1N1) virus in infected patients and vaccinated people in China.

BACKGROUND: To evaluate the risk of the recurrence and the efficiency of the vaccination, we followed-up antibody responses in patients with the 2009 pandemic H1N1 influenza and persons who received the pandemic H1N1 vaccine in Guangzhou China. METHODS: We collected serum samples from 129 patients and 86 vaccinated persons at day 0, 15, 30, 180 after the disease onset or the vaccination, respectively. Antibody titers in these serum samples were determined by haemagglutination inhibition (HI) assay using a local isolated virus strain A/Guangdong Liwan/SWL1538/2009(H1N1). RESULTS: HI antibody positive rate of the patients increased significantly from 0% to 60% at day 15 (χ(2) = 78, P<0.001) and 100% at day 30 (χ(2) = 23, P<0.001), but decreased significantly to 52% at day 180 (χ(2) = 38, P<0.001), while that of vaccinated subjects increased from 0% to 78% at day 15 (χ(2) = 110, P<0.001) and 81% at day 30 (χ(2) = 0.32, P = 0.57), but decreased significantly to 34% at day 180 (χ(2) = 39, P<0.001). Geometric mean titers (GMT) of HI antibodies in positive samples from the patients did not change significantly between day 15 and day 30 (T = 0.92, P = 0.36), but it decreased significantly from 80 at day 30 to 52 at day 180 (T = 4.5, P<0.001). GMT of vaccinated persons increased significantly from 100 at day 15 to 193 at day 30 (T = 4.5, P<0.001), but deceased significantly to 74 at day 180 (T = 5.1, P<0.001). Compared to the patients, the vaccinated subjects showed lower seroconversion rate (χ(2) = 11, P<0.001; χ(2) = 5.9, P = 0.015), but higher GMT (T = 6.0, P<0.001; T = 3.6, P = 0.001) at day 30 and day 180, respectively. CONCLUSION: Vaccination of 2009 influenza A (H1N1) was effective. However, about half or more recovered patients and vaccinated persons might have lost sufficient immunity against the recurrence of the viral infection after half a year. Vaccination or re-vaccination may be necessary for prevention of the recurrence.

Epidemiological investigation of an outbreak of pandemic influenza A (H1N1) 2009 in a boarding school: serological analysis of 1570 cases.

BACKGROUND: A large number of 2009 pandemic influenza A (H1N1) infections were localized in school populations. OBJECTIVES: To describe the epidemiology, clinical features and risk factors associated with an outbreak that occurred at a vocational boarding school in Guangzhou, P.R. China. STUDY DESIGN: Data were collected prospectively and retrospectively through the use of on-site doctors and a post-outbreak survey and blood collection. The survey was used to confirm symptoms, and to investigate a series of flu-related factors such as dormitory conditions, health habits, vaccine history and population contact history. Blood samples were taken for serological analysis. Pandemic H1N1 infection was initially confirmed by a real-time RT-PCR assay. Following the identification of the outbreak by the Guangzhou CDC on September 4, cases were diagnosed symptomatically and retrospectively by serological analysis using the hemagglutination inhibition assay and a neutralization assay. RESULTS: The infection rate was 32% (505/1570) and the attack rate was 22.2% (349/1570). The asymptomatic infection rate was 9.9% (156/1570). Sharing a classroom (OR=2.17, 95% CI: 1.62-2.91) and dormitory space (OR=2.32, 95% CI: 1.84-2.93) was associated with higher rates of infection. Opening windows for ventilation was the only control measure that significantly protected against infection. CONCLUSION: Social isolation and quarantine should be used to prevent the spread of infection. Ventilation and a control of air flow between classrooms and dorms should be implemented as possible. School closures may be effective if implemented early.

[Study on the inhibitory effect of RNA interference on replication of dengue virus].

To investigate the inhibitory effect of RNA interference (RNAi) on dengue virus I (DENV-1) replication. Small interfering RNA (siRNA) against the PreM gene of dengue virus was synthesized and transfected into C6/36 cells with liposome, which was then attacked by DENV-1 virus. The antiviral effect of siRNA was evaluated by cytopathic effect (CPE), the cell survival rate measured by MTT, and virus RNA quantified by real-time RT-PCR. The results showed that after 7 days post infection of dengue virus, the transfected C6/36 cells showed less CPE. The cell survival rate of the transfected C6/36 cells increased by 2.26 fold, and the amount of virus RNA in the transfected cells was reduced by about 97.54% as well. These findings indicated that the siRNA could effectively inhibit dengue virus RNA replication, and protect C6/36 cells from viral attack, indicating its potential role in prevention and treatment of dengue fever.

Inhibitory effect of small interfering RNA on dengue virus replication in mosquito cells.

BACKGROUND: Dengue viruses (DENs) are the wildest transmitted mosquito-borne pathogens throughout tropical and sub-tropical regions worldwide. Infection with DENs can cause severe flu-like illness and potentially fatal hemorrhagic fever. Although RNA interference triggered by long-length dsRNA was considered a potent antiviral pathway in the mosquito, only limited studies of the value of small interfering RNA (siRNA) have been conducted. RESULTS: A 21 nt siRNA targeting the membrane glycoprotein precursor gene of DEN-1 was synthesized and transfected into mosquito C6/36 cells followed by challenge with DEN. The stability of the siRNA in cells was monitored by flow cytometry. The antiviral effect of siRNA was evaluated by measurement of cell survival rate using the MTT method and viral RNA was quantitated with real-time RT-PCR. The presence of cells containing siRNA at 0.25, 1, 3, 5, 7 days after transfection were 66.0%, 52.1%, 32.0%, 13.5% and 8.9%, respectively. After 7 days incubation with DEN, there was reduced cytopathic effect, increased cell survival rate (76.9 ± 4.5% vs 23.6 ± 14.6%) and reduced viral RNA copies (Ct value 19.91 ± 0.63 vs 14.56 ± 0.39) detected in transfected C6/36 cells. CONCLUSIONS: Our data showed that synthetic siRNA against the DEN-1 membrane glycoprotein precursor gene effectively inhibited DEN-1 viral RNA replication and increased C6/36 cell survival rate. siRNA may offer a potential new strategy for prevention and treatment of DEN infection.

[Relation between clinical detection rates of the novel influenza virus A/H1N1 RNA in confirmed patients in Guangzhou and the disease course].

OBJECTIVE: To study the relation of the detection rates of the novel influenza virus A/H1N1 RNA in clinically confirmed patients in the 2009 pandemic with the age distribution of the patients and the disease course. METHODS: A total of 151 clinical patients with H1N1 infection were enrolled in this study, from whom 833 dynamic throat swab samples were obtained for detecting the H1N1 RNA using real-time PCR. A statistical analysis of the age distribution was performed among the patients with different disease courses. Chi-square for trend test was used to study the correlation between the detection rates of H1N1 RNA and the time of disease onset. RESULTS: The majority of patients were young with their ages ranging from 10 to 20 years (57.26%) and 20 to 30 years (22.18%). Chi-square for trend test revealed that the positivity rates of the throat swabs in the patients decreased with the prolongation of the disease course (chi(2)=9.784, P=0.002). CONCLUSION: Most of the H1N1 patients are young within the age range of 10-30 years, and the longest disease course can exceed 10 days. The positivity rates of throat swabs from the H1N1 patients decreases with the prolongation of the disease course.

[Risk assessment of H5N1 human infection after an outbreak of avian influenza in water fowl].

OBJECTIVE: To evaluate the risk of human infection after the outbreak of avian influenza H5N1 in animals, and probe the possibility for virus transmission. METHODS: By means of field epidemiological study, molecular epidemiology, serology and emergency surveillance, persons who had ever closely contacted with sick or dead poultry were observed. While, the RT-PCR and gene sequencing method were used to detect H5 nucleic acid from environmental swabs from 4 epidemic spots, and hemagglutination inhibition assay was also used to detect H5 antibody. RESULTS: Of 22 environmental swabs detected from 4 epidemic spots, one was positive for H5 nucleic acid, and the homogeneity was 95.9% as compared with H5N1 virus A/China/GD01/2006 (H5N1) found in Guangzhou in 2006 by gene sequence analysis. 62 environmental swabs from live poultry stalls of food markets near epidemic spot were detected negative. Six of 68 blood samples of contacts were positive for H9 antibody, and all were negative for H5 antibody. 68 throat swabs of contacts were detected negative for H5 nucleic acid. No close contact was found abnormal after 7 days medical observation. 337 influenza-like cases were reported in emergency surveillance, and no suspicious case was found. CONCLUSION: The current outbreak of H5N1 avian influenza in water fowls has not yet caused further transmission, and human avian influenza case has not been observed. It indicates that the ability of H5N1 virus to transmit to human is not strong yet, and the risk of human infection for H5N1 is still low.