RPEMHC: improved prediction of MHC-peptide binding affinity by a deep learning approach based on residue-residue pair encoding.

MOTIVATION: Binding of peptides to major histocompatibility complex (MHC) molecules plays a crucial role in triggering T cell recognition mechanisms essential for immune response. Accurate prediction of MHC-peptide binding is vital for the development of cancer therapeutic vaccines. While recent deep learning-based methods have achieved significant performance in predicting MHC-peptide binding affinity, most of them separately encode MHC molecules and peptides as inputs, potentially overlooking critical interaction information between the two. RESULTS: In this work, we propose RPEMHC, a new deep learning approach based on residue-residue pair encoding to predict the binding affinity between peptides and MHC, which encode an MHC molecule and a peptide as a residue-residue pair map. We evaluate the performance of RPEMHC on various MHC-II-related datasets for MHC-peptide binding prediction, demonstrating that RPEMHC achieves better or comparable performance against other state-of-the-art baselines. Moreover, we further construct experiments on MHC-I-related datasets, and experimental results demonstrate that our method can work on both two MHC classes. These extensive validations have manifested that RPEMHC is an effective tool for studying MHC-peptide interactions and can potentially facilitate the vaccine development. AVAILABILITY: The source code of the method along with trained models is freely available at https://github.com/lennylv/RPEMHC.

CAPLA: improved prediction of protein-ligand binding affinity by a deep learning approach based on a cross-attention mechanism.

MOTIVATION: Accurate and rapid prediction of protein-ligand binding affinity is a great challenge currently encountered in drug discovery. Recent advances have manifested a promising alternative in applying deep learning-based computational approaches for accurately quantifying binding affinity. The structure complementarity between protein-binding pocket and ligand has a great effect on the binding strength between a protein and a ligand, but most of existing deep learning approaches usually extracted the features of pocket and ligand by these two detached modules. RESULTS: In this work, a new deep learning approach based on the cross-attention mechanism named CAPLA was developed for improved prediction of protein-ligand binding affinity by learning features from sequence-level information of both protein and ligand. Specifically, CAPLA employs the cross-attention mechanism to capture the mutual effect of protein-binding pocket and ligand. We evaluated the performance of our proposed CAPLA on comprehensive benchmarking experiments on binding affinity prediction, demonstrating the superior performance of CAPLA over state-of-the-art baseline approaches. Moreover, we provided the interpretability for CAPLA to uncover critical functional residues that contribute most to the binding affinity through the analysis of the attention scores generated by the cross-attention mechanism. Consequently, these results indicate that CAPLA is an effective approach for binding affinity prediction and may contribute to useful help for further consequent applications. AVAILABILITY AND IMPLEMENTATION: The source code of the method along with trained models is freely available at https://github.com/lennylv/CAPLA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

DeepMPSF: A Deep Learning Network for Predicting General Protein Phosphorylation Sites Based on Multiple Protein Sequence Features.

Phosphorylation, as one of the most important post-translational modifications, plays a key role in various cellular physiological processes and disease occurrences. In recent years, computer technology has been gradually applied to the prediction of protein phosphorylation sites. However, most existing methods rely on simple protein sequence features that provide limited contextual information. To overcome this limitation, we propose DeepMPSF, a phosphorylation site prediction model based on multiple protein sequence features. There are two types of features: sequence semantic features, which comprise protein residue type information and relative position information within protein sequence, and protein background biophysical features, which include global semantic information containing more comprehensive protein background information obtained from pretrained models. To extract these features, DeepMPSF employs two separate subnetworks: the S71SFE module and the BBFE module, which automatically extract high-level semantic features. Our model incorporates a learning strategy for handling imbalanced datasets through ensemble learning during training and prediction. DeepMPSF is trained and evaluated on a well-established dataset of human proteins. Comparing the analysis with other benchmark methods reveals that DeepMPSF outperforms in predicting both S/T residues and Y residues. In particular, DeepMPSF showed excellent generalization performance in cross-species blind test performance, with an average improvement of 5.63%/5.72%, 22.28%/25.94%, 20.11%/17.49%, and 26.40%/28.33% for Mus musculus/Rattus norvegicus test sets in area under curves (AUCs) of ROC curve, AUC of the PR curve, F1-score, and MCC metrics, respectively. Furthermore, it also shows excellent performance in the latest updated case of natural proteins with functional phosphorylation sites. Through an ablation study and visual analysis, we uncover that the design of different feature modules significantly contributes to the accurate classification of DeepMPSF, which provides valuable insights for predicting phosphorylation sites and offers effective support for future downstream research.

Multisource Attention-Mechanism-Based Encoder-Decoder Model for Predicting Drug-Drug Interaction Events.

Many computational methods have been proposed to predict drug-drug interactions (DDIs), which can occur when combining drugs to treat various diseases, but most mainly utilize single-source features of drugs, which is inadequate for drug representation. To fill this gap, we propose two attention-mechanism-based encoder-decoder models that incorporate multisource information: one is MAEDDI, which can predict DDIs, and the other is MAEDDIE, which can make further DDI-associated event predictions for drug pairs with DDIs. To better express the drug feature, we used three encoding methods to encode the drugs, integrating the self-attention mechanism, cross-attention mechanism, and graph attention network to construct a multisource feature fusion network. Experiments showed that both MAEDDI and MAEDDIE performed better than some state-of-the-art methods in various validation attempts at different experimental tasks. The visualization analysis showed that the semantic features of drug pairs learned from our models had a good drug representation. In practice, MAEDDIE successfully screened 43 DDI events on favipiravir, an influenza antiviral drug, with a success rate of nearly 50%. Our model achieved competitive results, mainly owing to the design of sequence-based, structural, biochemical, and statistical multisource features. Moreover, different encoders constructed based on different features learn the interrelationship information between drug pairs, and the different representations of these drug pairs are incorporated to predict the target problem. All of these encoders were designed to better characterize the complex DDI relationships, allowing us to achieve high generalization in DDI and DDI-associated event predations.

Induced pluripotent stem cells attenuate chronic allogeneic vasculopathy in an integrin beta-1-dependent manner.

This study aimed to determine the mechanism of isogeneic-induced pluripotent stem cells (iPSCs) homing to vascular transplants and their therapeutic effect on chronic allogeneic vasculopathy. We found that integrin ß1 (Intgß1) was the dominant integrin ß unit in iPSCs that mediates the adhesion of circulatory and endothelial cells (ECs). Intgß1 knockout or Intgß1-siRNAs inhibit iPSC adhesion and migration across activated endothelial monolayers. The therapeutic effects of the following were examined: iPSCs, Intgß1-knockout iPSCs, iPSCs transfected with Intgß1-siRNAs or nontargeting siRNAs, iPSC-derived ECs, iPSC-derived ECs simultaneously overexpressing Intgα4 and Intgß1, iPSCs precultured in endothelial medium for 3 days (endothelial-prone stem cells), primary aortic ECs, mouse embryonic fibroblasts, and phosphate-buffered saline (control). The cells were administered every 3 days for a period of 8 weeks. iPSCs, iPSCs transfected with nontargeting siRNAs, and endothelial-prone stem cells selectively homed on the luminal surface of the allografts, differentiated into ECs, and decreased neointimal proliferation. Through a single administration, we found that iPSCs trafficked to allograft lesions, differentiated into ECs within 1 week, and survived for 4-8 weeks. The therapeutic effect of a single administration was moderate. Thus, Intgß1 and pluripotency are essential for iPSCs to treat allogeneic vasculopathy.

Neutrophil CD64 Index as a superior biomarker for early diagnosis of infection in febrile patients in the hematology department.

BACKGROUND: In the hematology department, the availability of biomarkers for early detection of infection is difficult to obtain. The present study aimed to compare the diagnostic values of neutrophil CD64 Index, procalcitonin (PCT), interleukin-6 (IL-6) and C-reactive protein (CRP) and to determine whether the combined analysis of these biomarkers offer stronger predictive power in the diagnosis for the infection of febrile patients. METHODS: Neutrophil CD64 Index, PCT, IL-6 and CRP levels were determined in 356 febrile patients in the hematology ward from May 2013 to May 2015. Sensitivity, specificity, positive and negative likelihood ratios, positive and negative predictive values, receiver operating characteristic (ROC) areas under the curve (AUC), and logistic regression analysis were determined to evaluate the diagnostic values of these biomarkers. RESULTS: The levels of the four biomarkers were higher in the infection patients (p<0.001), and the PCT and IL-6 were higher in the patients with positive microbial blood culture (p<0.01). The neutrophil CD64 Index, PCT, IL-6, CRP had AUCs of 0.95, 0.83, 0.75 and 0.73, respectively. The best cut-off value of the neutrophil CD64 Index to detect infections was 5.06, with high specificity (87.5%) and sensitivity (88.4%). Furthermore, neutrophil CD64 Index, PCT and IL-6 offered the best combination of diagnosis with sensitivity of 93.9% and an AUC of 0.95. In addition, the neutrophil CD64 Index may have a special value to assist the physician to diagnose infection in the neutropenic patients with fever. CONCLUSIONS: The neutrophil CD64 Index is useful for early identification of infections in febrile patients in the hematology department. The combined analysis of the CD64 Index, PCT and IL-6 could further improve its sensitivity.

MultiModRLBP: A Deep Learning Approach for Multi-Modal RNA-Small Molecule Ligand Binding Sites Prediction.

This study aims to tackle the intricate challenge of predicting RNA-small molecule binding sites to explore the potential value in the field of RNA drug targets. To address this challenge, we propose the MultiModRLBP method, which integrates multi-modal features using deep learning algorithms. These features include 3D structural properties at the nucleotide base level of the RNA molecule, relational graphs based on overall RNA structure, and rich RNA semantic information. In our investigation, we gathered 851 interactions between RNA and small molecule ligand from the RNAglib dataset and RLBind training set. Unlike conventional training sets, this collection broadened its scope by including RNA complexes that have the same RNA sequence but change their respective binding sites due to structural differences or the presence of different ligands. This enhancement enables the MultiModRLBP model to more accurately capture subtle changes at the structural level, ultimately improving its ability to discern nuances among similar RNA conformations. Furthermore, we evaluated MultiModRLBP on two classic test sets, Test18 and Test3, highlighting its performance disparities on small molecules based on metal and non-metal ions. Additionally, we conducted a structural sensitivity analysis on specific complex categories, considering RNA instances with varying degrees of structural changes and whether they share the same ligands. The research results indicate that MultiModRLBP outperforms the current state-of-the-art methods on multiple classic test sets, particularly excelling in predicting binding sites for non-metal ions and instances where the binding sites are widely distributed along the sequence. MultiModRLBP also can be used as a potential tool when the RNA structure is perturbed or the RNA experimental tertiary structure is not available. Most importantly, MultiModRLBP exhibits the capability to distinguish binding characteristics of RNA that are structurally diverse yet exhibit sequence similarity. These advancements hold promise in reducing the costs associated with the development of RNA-targeted drugs.

Facile fabrication of graphene encapsulating 3d transition metal nanoparticles as highly active and anti-poisoning catalysts for selective hydrogenation of nitroaromatics.

Graphene encapsulating 3d transition metal nanoparticles (Ni, Co, Fe@G) are successfully fabricated through pyrolysis of complexes which are simply prepared via "acid-base reactions" between metal hydroxides and carboxylic acid such as citric acid. In particular, the Ni@G catalyst exhibits outstanding catalytic activity and selectivity (>99%) toward the reduction of various nitroaromatics under mild conditions (1 MPa H2, 60 °C), even in the presence of poisons (CO and thiophene etc.). This "acid-base reactions" based method provides a facile and scalable approach to prepare graphene encapsulating 3d transition metals with wide ranges of applications.

Weighted bootstrapping: a correction method for assessing the robustness of phylogenetic trees.

BACKGROUND: Non-parametric bootstrapping is a widely-used statistical procedure for assessing confidence of model parameters based on the empirical distribution of the observed data 1 and, as such, it has become a common method for assessing tree confidence in phylogenetics 2. Traditional non-parametric bootstrapping does not weigh each tree inferred from resampled (i.e., pseudo-replicated) sequences. Hence, the quality of these trees is not taken into account when computing bootstrap scores associated with the clades of the original phylogeny. As a consequence, traditionally, the trees with different bootstrap support or those providing a different fit to the corresponding pseudo-replicated sequences (the fit quality can be expressed through the LS, ML or parsimony score) contribute in the same way to the computation of the bootstrap support of the original phylogeny. RESULTS: In this article, we discuss the idea of applying weighted bootstrapping to phylogenetic reconstruction by weighting each phylogeny inferred from resampled sequences. Tree weights can be based either on the least-squares (LS) tree estimate or on the average secondary bootstrap score (SBS) associated with each resampled tree. Secondary bootstrapping consists of the estimation of bootstrap scores of the trees inferred from resampled data. The LS and SBS-based bootstrapping procedures were designed to take into account the quality of each "pseudo-replicated" phylogeny in the final tree estimation. A simulation study was carried out to evaluate the performances of the five weighting strategies which are as follows: LS and SBS-based bootstrapping, LS and SBS-based bootstrapping with data normalization and the traditional unweighted bootstrapping. CONCLUSIONS: The simulations conducted with two real data sets and the five weighting strategies suggest that the SBS-based bootstrapping with the data normalization usually exhibits larger bootstrap scores and a higher robustness compared to the four other competing strategies, including the traditional bootstrapping. The high robustness of the normalized SBS could be particularly useful in situations where observed sequences have been affected by noise or have undergone massive insertion or deletion events. The results provided by the four other strategies were very similar regardless the noise level, thus also demonstrating the stability of the traditional bootstrapping method.

Occupational Mercury Exposure at a Thermometer Facility - Jiangsu Province, 2019.

WHAT IS ALREADY KNOWN ABOUT THIS TOPIC?: Mercury is still used in the manufacture of some thermometers in China. This may pose health risks if exposure is not properly prevented and controlled. WHAT IS ADDED BY THIS REPORT?: An onsite investigation of a workplace at a thermometer facility in Jiangsu Province in 2019 found heavily elevated airborne and urinary mercury levels among a massive number of workers exposed to mercury. Traditional and obsolete technology as well as inadequate protection measures for occupational hazards caused this high level of exposure. WHAT ARE THE IMPLICATIONS FOR PUBLIC HEALTH PRACTICE?: Employers at thermometer producing facilities need to adopt effective protection measures and implement strict management. Monitoring exposure, adopting better engineering controls, diligent cleaning, and providing recommended personal protective equipment (PPE) along with training to their workers properly can alleviate mercury exposure at their facilities. In addition, transitioning to mercury-free thermometers would eliminate the risk of mercury exposure.

In situ evolution of the active phase on stainless steel mesh toward a cost-effective bifunctional electrode for energy storage and conversion.

Developing low-cost, efficient and stable electrode materials is a major challenge of energy storage and conversion. Here, we report a facile, cost-effective and scaled-up self-sacrificing strategy for transforming commercial stainless steel into highly active and ultrastable electrodes for supercapacitors and the hydrogen evolution reaction. The modified stainless steel displays superior electrochemical activity as well as excellent cycling durability.

An Efficient Metal-Free Catalyst for Oxidative Dehydrogenation Reaction: Activated Carbon Decorated with Few-Layer Graphene.

Activated carbon (AC) has been widely used in the catalysis field because of its low cost, scalable production, high specific surface area, and abundant exposed edge. Because of the amorphous structure, traditional AC is unstable in presence of O2 at high temperature, which hinders the application of AC catalysts in oxidative dehydrogenation (ODH) of alkanes. Here, partially graphitic AC decorated with few-layer graphene is facilely fabricated by simple high-temperature calcination. The graphitic transformation significantly enhances the antioxidation property, long-term stability of AC during the ODH reaction, and especially dramatically increases the graphitic edge areas in which the active ketonic carbonyl groups are selectively formed in ODH reactions. A high reactivity with 41.5 % selectivity and 13.2 % yield to C4 alkenes were obtained at 450 °C over the optimized catalyst, which is superior to all the previously reported carbon catalysts and shows a great potential for industrial application.

Relationship between JAK2V617F mutation, allele burden and coagulation function in Ph-negative myeloproliferative neoplasms.

OBJECTIVES: Our aim was to explore the relationship between JAK2V617F mutation allele burden and hematological parameters especially in coagulation function in Chinese population. METHODS: This study included 133 Ph-negative myeloproliferative neoplasms (MPNs) patients between 2013 and 2016. All the clinical and experimental data of patients were collected at the time of the diagnosis without any prior treatment, including blood parameters, coagulation function, splenomegaly, vascular events and chromosome karyotype. PCR and qPCR were used to detect JAK2V617F mutation and JAK2V617F mutation allele burden. RESULTS: In polycythemia vera patients, a positive correlation between the allele burden of JAK2V617F mutation and PLT counts was found; in essential thrombocythemia (ET) patients, WBC counts, RBC counts, HB, and HCT were higher in mutated patients than in wild-type patients. Furthermore, PT-INR was higher in ET and PMF mutated patients. In addition, a positive correlation between the allele burden of JAK2V617F mutation and activated partial thromboplastin time (APTT) was observed in JAK2V617F mutated ET patients. CONCLUSIONS: Higher hematologic parameters including counts of WBC, RBC, and PLT are closely associated with JAK2V617F mutation and its burden in Ph-negative MPNs; importantly, PT-INR, APTT are also related to JAK2V617F mutation and allele burden. Thus, our data indicate that JAK2V617F mutation allele burden might not only represent the burden of MPN but also alter the coagulation function.

Prognostic value of RDW in cancers: a systematic review and meta-analysis.

Red blood cell distribution width (RDW), a parameter that used to differentiate the type of anemia for several decades, recent studies suggest it was a prognostic factor in various types of cancer patients. However, the prognostic value of RDW in cancer patients remains controversial. Here, we performed a meta-analysis and systematic review to evaluate the prognostic value of RDW in cancer patients. Relevant studies were picked out from the databases of Web of Science, Embase, Pubmed and Cochrane Library. A total of 16 papers with 4267 patients were included in this meta-analysis, and the combined results indicated that elevated RDW was associated with poor over survival (OS) (HR = 1.47, 95%CI:1.29-1.66), poor cancer-specific survival (CSS) (HR = 1.46, 95%CI:1.08-1.85), poor disease-free survival (DFS) (HR = 1.91, 95%CI:1.27-2.56), poor event-free survival (EFS) (HR = 2.98, 95%CI:0.57-5.39) and poor progress-free survival (PFS) (HR = 3.21, 95%CI:0.33-6.75) after treatment. Furthermore, the similar results were observed in subgroup analysis stratified by cancer type, cutoff value of RDW, sample size and ethnicity. In conclusion, this meta-analysis demonstrated that RDW may be a potential prognostic marker in patients with cancer, and high RDW may also be associated with poor outcomes.

Transduction of interleukin-10 through renal artery attenuates vascular neointimal proliferation and infiltration of immune cells in rat renal allograft.

Renal transplantation is the treatment of choice for end-stage renal failure. Although acute rejection is not a major issue anymore, chronic rejection, especially vascular rejection, is still a major factor that might lead to allograft dysfunction on the long term. The role of the local immune-regulating cytokine interleukin-10 (IL-10) in chronic renal allograft is unclear. Many clinical observations showed that local IL-10 level was negatively related to kidney allograft function. It is unknown this negative relationship was the result of immunostimulatory property or insufficient immunosuppression property of local IL-10. We performed ex vivo transduction before transplantation through artery of the renal allograft using adeno-associated viral vectors carrying IL-10 gene. Twelve weeks after transplantation, we found intrarenal IL-10 gene transduction significantly inhibited arterial neointimal proliferation, the number of occluded intrarenal artery, interstitial fibrosis, peritubular capillary congestion and glomerular inflammation in renal allografts compared to control allografts receiving PBS or vectors carrying YFP. IL-10 transduction increased serum IL-10 level at 4 weeks but not at 8 and 12 weeks. Renal IL-10 level increased while serum creatinine decreased significantly in IL-10 group at 12 weeks compared to PBS or YFP controls. Immunohistochemical staining showed unchanged total T cells (CD3) and B cells (CD45R/B220), decreased cytotoxic T cells (CD8), macrophages (CD68) and increased CD4+ and FoxP3+ cells in IL-10 group. In summary, intrarenal IL-10 inhibited the allograft rejection while modulated immune response.

[Preparation of a reversed-phase capillary monolithic column and its application in the separation of polypeptide mixtures].

Capillary monolithic columns (75 microm i.d.) were prepared by the copolymerization of lauryl methacrylate as the basic monomer, ethylene dimethacrylate as the cross-linking agent and 1-dodecyl alcohol, 1,4-butanediol and dimethyl sulfoxide as the porogenic mixture. The synthetic stationary phases had better mechanical properties and chemical stabilities. A series of characterization and evaluations were performed on the capillary monolithic columns including the scanning electron microscope (SEM) images, the influences of pressure and the effects on the separation of peptide mixtures by changing the proportions of the porogen solution and cross-linking agent. The final prescription ciontained 15% (w/w) monomer, 15% (w/ w) cross-linking agent, and 70% (w/w) porogenic agent. Then the solution was heated at 70 1C for 24 h. The test of relationship between column length and back pressure showed that the capillary monolithic columns prepared have superior permeability, so a longer column can be used to improve the effects of separation. The prepared capillary monolithic columns are fitted on the nano-scale high performance liquid chromatography for the separation of tryptic digests of bovine serum albumin (BSA) and human plasma samples, and better results have been obtained.