Simultaneous Prediction of Interaction Sites on the Protein and Peptide Sides of Complexes through Multilayer Graph Convolutional Networks.

Identifying the binding residues of protein-peptide complexes is essential for understanding protein function mechanisms and exploring drug discovery. Recently, many computational methods have been developed to predict the interaction sites of either protein or peptide. However, to our knowledge, no prediction method can simultaneously identify the interaction sites on both the protein and peptide sides. Here, we propose a deep graph convolutional network (GCN)-based method called GraphPPepIS to predict the interaction sites of protein-peptide complexes using protein and peptide structural information. We also propose a companion method, SeqPPepIS, for assisting with the lack of structural information and the flexibility of peptides. SepPPepIS replaces the peptide structural features in GraphPPepIS by learning features from peptide sequences. We performed a comprehensive evaluation of the benchmark data sets, and the results show that our two methods outperform state-of-the-art methods on the accurate interaction sites of both protein and peptide sides. We show that our methods can help improve protein-peptide docking. For docking data sets, our methods maintain robust performance in identifying binding sites, thereby enhancing the prediction of peptide binding poses. Finally, we visualized the analysis of protein and peptide graph embedding to demonstrate the learning ability of graph convolution in predicting interaction sites, which was mainly obtained through the shared parameters of a protein graph and peptide graph.

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.

Extracting biomedical relation from cross-sentence text using syntactic dependency graph attention network.

In biomedical literature, cross-sentence texts can usually express rich knowledge, and extracting the interaction relation between entities from cross-sentence texts is of great significance to biomedical research. However, compared with single sentence, cross-sentence text has a longer sequence length, so the research on cross-sentence text information extraction should focus more on learning the context dependency structural information. Nowadays, it is still a challenge to handle global dependencies and structural information of long sequences effectively, and graph-oriented modeling methods have received more and more attention recently. In this paper, we propose a new graph attention network guided by syntactic dependency relationship (SR-GAT) for extracting biomedical relation from the cross-sentence text. It allows each node to pay attention to other nodes in its neighborhood, regardless of the sequence length. The attention weight between nodes is given by a syntactic relation graph probability network (SR-GPR), which encodes the syntactic dependency between nodes and guides the graph attention mechanism to learn information about the dependency structure. The learned feature representation retains information about the node-to-node syntactic dependency, and can further discover global dependencies effectively. The experimental results demonstrate on a publicly available biomedical dataset that, our method achieves state-of-the-art performance while requiring significantly less computational resources. Specifically, in the "drug-mutation" relation extraction task, our method achieves an advanced accuracy of 93.78% for binary classification and 92.14% for multi-classification. In the "drug-gene-mutation" relation extraction task, our method achieves an advanced accuracy of 93.22% for binary classification and 92.28% for multi-classification. Across all relation extraction tasks, our method improves accuracy by an average of 0.49% compared to the existing best model. Furthermore, our method achieved an accuracy of 69.5% in text classification, surpassing most existing models, demonstrating its robustness in generalization across different domains without additional fine-tuning.

Research on RNA secondary structure predicting via bidirectional recurrent neural network.

BACKGROUND: RNA secondary structure prediction is an important research content in the field of biological information. Predicting RNA secondary structure with pseudoknots has been proved to be an NP-hard problem. Traditional machine learning methods can not effectively apply protein sequence information with different sequence lengths to the prediction process due to the constraint of the self model when predicting the RNA secondary structure. In addition, there is a large difference between the number of paired bases and the number of unpaired bases in the RNA sequences, which means the problem of positive and negative sample imbalance is easy to make the model fall into a local optimum. To solve the above problems, this paper proposes a variable-length dynamic bidirectional Gated Recurrent Unit(VLDB GRU) model. The model can accept sequences with different lengths through the introduction of flag vector. The model can also make full use of the base information before and after the predicted base and can avoid losing part of the information due to truncation. Introducing a weight vector to predict the RNA training set by dynamically adjusting each base loss function solves the problem of balanced sample imbalance. RESULTS: The algorithm proposed in this paper is compared with the existing algorithms on five representative subsets of the data set RNA STRAND. The experimental results show that the accuracy and Matthews correlation coefficient of the method are improved by 4.7% and 11.4%, respectively. CONCLUSIONS: The flag vector introduced allows the model to effectively use the information before and after the protein sequence; the introduced weight vector solves the problem of unbalanced sample balance. Compared with other algorithms, the LVDB GRU algorithm proposed in this paper has the best detection results.

Iron availability is a key factor for freshwater cyanobacterial survival against saline stress.

Estuaries are among the most productive ecosystems and dynamic environments on Earth. Varying salinity is the most important challenge for phytoplankton survival in estuaries. In order to investigate the role of iron nutrition on phytoplankton survival under salinity stress, a freshwater cyanobacterial strain was cultivated in media added with different proportions of seawater (measured with siderophore activities), and supplied with gel-immobilized ferrihydrite as iron source. Results showed that the strain grew well in media with 0% seawater supplied with ferrihydrite as iron source. Surprisingly, the biomasses in media with 50% seawater, with more newly excreted siderophore, were similar to those with 0% seawater, but better than those with 6.25%, 12.5% and 25% seawater. Smaller iron isotopic discriminations between the cyanobacterial cells associated iron and dissolved iron were observed in media with 0% and 50% seawater suggested that higher fractions of iron uptake from aqueous dissolved iron reservoir by these comparatively larger biomasses. In summary, this study proved that iron availability plays a key role in cyanobacterial survival under varying salinity stress, and suggested that siderophores introduced by seawater may accelerate iron dissolution, increase iron availability, and make cyanobacterial cells overcome the adverse effects of high-salinity, and indicated that siderophore excretion is a kind of survival strategy for phytoplankton in face of salinity stress.

Research on predicting 2D-HP protein folding using reinforcement learning with full state space.

BACKGROUND: Protein structure prediction has always been an important issue in bioinformatics. Prediction of the two-dimensional structure of proteins based on the hydrophobic polarity model is a typical non-deterministic polynomial hard problem. Currently reported hydrophobic polarity model optimization methods, greedy method, brute-force method, and genetic algorithm usually cannot converge robustly to the lowest energy conformations. Reinforcement learning with the advantages of continuous Markov optimal decision-making and maximizing global cumulative return is especially suitable for solving global optimization problems of biological sequences. RESULTS: In this study, we proposed a novel hydrophobic polarity model optimization method derived from reinforcement learning which structured the full state space, and designed an energy-based reward function and a rigid overlap detection rule. To validate the performance, sixteen sequences were selected from the classical data set. The results indicated that reinforcement learning with full states successfully converged to the lowest energy conformations against all sequences, while the reinforcement learning with partial states folded 50% sequences to the lowest energy conformations. Reinforcement learning with full states hits the lowest energy on an average 5 times, which is 40 and 100% higher than the three and zero hit by the greedy algorithm and reinforcement learning with partial states respectively in the last 100 episodes. CONCLUSIONS: Our results indicate that reinforcement learning with full states is a powerful method for predicting two-dimensional hydrophobic-polarity protein structure. It has obvious competitive advantages compared with greedy algorithm and reinforcement learning with partial states.

Ranking near-native candidate protein structures via random forest classification.

BACKGROUND: In ab initio protein-structure predictions, a large set of structural decoys are often generated, with the requirement to select best five or three candidates from the decoys. The clustered central structures with the most number of neighbors are frequently regarded as the near-native protein structures with the lowest free energy; however, limitations in clustering methods and three-dimensional structural-distance assessments make identifying exact order of the best five or three near-native candidate structures difficult. RESULTS: To address this issue, we propose a method that re-ranks the candidate structures via random forest classification using intra- and inter-cluster features from the results of the clustering. Comparative analysis indicated that our method was better able to identify the order of the candidate structures as comparing with current methods SPICKR, Calibur, and Durandal. The results confirmed that the identification of the first model were closer to the native structure in 12 of 43 cases versus four for SPICKER, and the same as the native structure in up to 27 of 43 cases versus 14 for Calibur and up to eight of 43 cases versus two for Durandal. CONCLUSIONS: In this study, we presented an improved method based on random forest classification to transform the problem of re-ranking the candidate structures by an binary classification. Our results indicate that this method is a powerful method for the problem and the effect of this method is better than other methods.

Predicting RNA secondary structure via adaptive deep recurrent neural networks with energy-based filter.

BACKGROUND: RNA secondary structure prediction is an important issue in structural bioinformatics, and RNA pseudoknotted secondary structure prediction represents an NP-hard problem. Recently, many different machine-learning methods, Markov models, and neural networks have been employed for this problem, with encouraging results regarding their predictive accuracy; however, their performances are usually limited by the requirements of the learning model and over-fitting, which requires use of a fixed number of training features. Because most natural biological sequences have variable lengths, the sequences have to be truncated before the features are employed by the learning model, which not only leads to the loss of information but also destroys biological-sequence integrity. RESULTS: To address this problem, we propose an adaptive sequence length based on deep-learning model and integrate an energy-based filter to remove the over-fitting base pairs. CONCLUSIONS: Comparative experiments conducted on an authoritative dataset RNA STRAND (RNA secondary STRucture and statistical Analysis Database) revealed a 12% higher accuracy relative to three currently used methods.

Novel Core-Interlayer-Shell DOX/ZnPc Co-loaded MSNs@ pH-Sensitive CaP@PEGylated Liposome for Enhanced Synergetic Chemo-Photodynamic Therapy.

PURPOSE: This work was intended to develop novel doxorubicin (DOX)/zinc (II) phthalocyanine (ZnPc) co-loaded mesoporous silica (MSNs)@ calcium phosphate (CaP)@PEGylated liposome nanoparticles (NPs) that could efficiently achieve collaborative anticancer therapy by the combination of photodynamic therapy (PDT) and chemotherapy. The interlayer of CaP could be utilized to achieve pH-triggered controllable drug release, promote the cellular uptake, and induce cell apoptosis to further enhance the anticancer effects. METHODS: MSNs were first synthesized as core particles in which the pores were diffusion-filled with DOX, then the cores were coated by CaP followed by the liposome encapsulation with ZnPc to form the final DOX/ZnPc co-loaded MSNs@CaP@PEGylated liposome. RESULTS: A core-interlayer-shell MSNs@CaP@PEGylated liposomes was developed as a multifunctional theranostic nanoplatform. In vitro experiment indicated that CaP could not only achieve pH-triggered controllable drug release, promote the cellular uptake of the NPs, but also generate high osmotic pressure in the endo/lysosomes to induce cell apoptosis. Besides, the chemotherapy using DOX and PDT effect was achieved by the photosensitizer ZnPc. Furthermore, the MSNs@CaP@PEGylated liposomes showed outstanding tumor-targeting ability by enhanced permeability and retention (EPR) effect. CONCLUSIONS: The novel prepared MSNs@CaP@PEGylated liposomes could serve as a promising multifunctional theranostic nanoplatform in anticancer treatment by synergic chemo-PDT and superior tumor-targeting ability.

Effect of rs1063843 in the CAMKK2 gene on the dorsolateral prefrontal cortex.

Recently, a single nucleotide polymorphism (SNP) in the CAMKK2 gene (rs1063843) was found to be associated with lower expression of the gene in the dorsolateral prefrontal cortex (DLPFC) and with schizophrenia (SCZ) and deficits in working memory and executive function. However, the brain mechanism underlying this association is poorly understood. A functional magnetic resonance imaging (fMRI) study (N = 84 healthy volunteers) involving multiple cognitive tasks, including a Stroop task (to measure attentional executive control), an N-back task (to measure working memory), and a delay discounting task (to measure decision making) to identify the brain regions affected by rs1063843 was performed. Across all three tasks, it was found that carriers of the risk allele consistently exhibited increased activation of the left DLPFC. In addition, the risk allele carriers also exhibited increased activation of the right DLPFC and the left cerebellum during the Stroop task and of the left caudate nucleus during the N-back task. These findings helped to elucidate the role of CAMKK2 in cognitive functions and in the etiology of SCZ. Hum Brain Mapp 37:2398-2406, 2016. © 2016 Wiley Periodicals, Inc.

The 37-kDa laminin receptor precursor regulates the malignancy of human glioma cells.

Glioma is one of the most common brain tumors and one of the most aggressive cancers. Although extensive progress has been made regarding to the diagnosis and treatment, the mortality in glioma patients is still high. Therefore, finding new therapeutic targets to the glioma is critical to the advancement in cancer treatment. Recently, the 37-kDa laminin receptor precursor (37LRP) was reported to play important roles in occurrence of some types of cancer, indicating that this molecule may function as a key regulator in the tumor migration and metastasis. However, there is still no report to elucidate the correlation between 37LRP expression and glioma genesis and development. In this study, we found the higher expression of 37LRP in the glioma cells compared with the normal brain cells. We also indicated that the downregulation of 37LRP could affect the glioma biomarker expression and also weaken the proliferative, migratory, and metastatic capacity of glioma cells in vitro. Furthermore, 37LRP silencing inhibited the glioma tumor growth in vivo. Collectively, these data demonstrated that 37LRP regulates the metastasis of glioma cells in vitro and tumor growth in vivo, suggesting that 37LRP may function as a potential molecular target in the glioma treatment.

The ANK3 gene and facial affect processing: An ERP study.

ANK3 is one of the most promising candidate genes for bipolar disorder (BD). A polymorphism (rs10994336) within the ANK3 gene has been associated with BD in at least three genome-wide association studies of BD [McGuffin et al., 2003; Kieseppä, 2004; Edvardsen et al., 2008]. Because facial affect processing is disrupted in patients with BD, the current study aimed to explore whether the BD risk alleles are associated with the N170, an early event-related potential (ERP) component related to facial affect processing. We collected data from two independent samples of healthy individuals (Ns = 83 and 82, respectively) to test the association between rs10994336 and an early event-related potential (ERP) component (N170) that is sensitive to facial affect processing. Repeated-measures analysis of covariance in both samples consistently revealed significant main effects of rs10994336 genotype (Sample I: F (1, 72) = 7.24, P = 0.009; Sample II: F (1, 69) = 11.81, P = 0.001), but no significant interaction of genotype × electrodes (Ps > 0.05) or genotype × emotional conditions (Ps > 0.05). These results suggested that rs10994336 was linked to early ERP component reflecting facial structural encoding during facial affect processing. These results shed new light on the brain mechanism of this risk SNP and associated disorders such as BD. © 2016 Wiley Periodicals, Inc.

Orthogonally functionalized nanoscale micelles for active targeted codelivery of methotrexate and mitomycin C with synergistic anticancer effect.

The design of nanoscale drug delivery systems for the targeted codelivery of multiple therapeutic drugs still remains a formidable challenge (ACS Nano, 2013, 7, 9558-9570; ACS Nano, 2013, 7, 9518-9525). In this article, both mitomycin C (MMC) and methotrexate (MTX) loaded DSPE-PEG micelles (MTX-M-MMC) were prepared by self-assembly using the dialysis technique, in which MMC-soybean phosphatidylcholine complex (drug-phospholipid complex) was encapsulated within MTX-functionalized DSPE-PEG micelles. MTX-M-MMC could coordinate an early phase active targeting effect with a late-phase synergistic anticancer effect and enable a multiple-responsive controlled release of both drugs (MMC was released in a pH-dependent pattern, while MTX was released in a protease-dependent pattern). Furthermore, MTX-M-MMC could codeliver both drugs to significantly enhance the cellular uptake, intracellular delivery, cytotoxicity, and apoptosis in vitro and improve the tumor accumulation and penetration and anticancer effect in vivo compared with either both free drugs treatment or individual free drug treatment. To our knowledge, this work provided the first example of the systemically administrated, orthogonally functionalized, and self-assisted nanoscale micelles for targeted combination cancer chemotherapy. The highly convergent therapeutic strategy opened the door to more simplified, efficient, and flexible nanoscale drug delivery systems.

Bacillus-shape design of polymer based drug delivery systems with janus-faced function for synergistic targeted drug delivery and more effective cancer therapy.

The particle shape of the drug delivery systems had a strong impact on their in vitro and in vivo performance, but there was limited availability of techniques to produce the specific shaped drug carriers. In this article, the novel methotrexate (MTX) decorated MPEG-PLA nanobacillus (MPEG-PLA-MTX NB) was prepared by the self-assembly technique followed by the extrusion through SPG membrane with high N2 pressure for targeted drug delivery, in which Janus-like MTX was not only used as a specific anticancer drug but could also be served as a tumor-targeting ligand. The MPEG-PLA-MTX NBs demonstrated much higher in vitro and in vivo targeting efficiency compared to the MPEG-PLA-MTX nanospheres (MPEG-PLA-MTX NSs) and MPEG-PLA nanospheres (MPEG-PLA NSs). In addition, the MPEG-PLA-MTX NBs also displayed much more excellent in vitro and in vivo antitumor activity than the MPEG-PLA-MTX NSs and free MTX injection. To our knowledge, this work provided the first example of the integration of the shape design (which mediated an early phase tumor accumulation and a late-phase cell internalization) and Janus-faced function (which mediated an early phase active targeting effect and a late-phase anticancer effect) on the basis of nanoscaled drug delivery systems. The highly convergent and cooperative drug delivery strategy opens the door to more drug delivery systems with new shapes and functions for cancer therapy.

The prevalence and determinants of drug-resistance-associated mutations in the HIV-1-infected MSM population of Henan Province in China.

To estimate the prevalence of human immunodeficiency virus (HIV) drug resistance (DR) in a population of men who have sex with men (MSM) from Henan Province of China and to identify the DR-associated HIV-1 mutations in these MSM. The HIV-positive status of the MSM subjects in this study was confirmed using ELISA and Western blotting. The MSM subjects were classified into non-treatment group (n = 106) and treatment group (n = 313). CD4(+) T-lymphocyte counts were obtained by flow cytometry, and viral load was measured by branched DNA (bDNA) signal amplification assay. HIV-1 genotypic resistance tests were performed by sequence analysis of the HIV-1 protease and reverse transcriptase genes. In the non-treatment group, 15 patients (14.2 %) displayed DR to non-nucleoside reverse transcriptase inhibitor (NNRTI). In the treatment group, the failure rate of viral suppression was 38.33 % and the DR rate was 33.2 %, which was higher than the rate observed in the non-treatment group (P < 0.05). The incidence of mutations corresponding to NNRTI resistance was significantly higher than the incidence of mutations corresponding to nucleoside reverse transcriptase inhibitor (NRTI) resistance (32.9 % vs. 26.5 %) in the cohort. After antiretroviral therapy (ART), the frequencies of K103N, G190A, Y181C, and V106A mutations were highly elevated. Logistic regression analysis results showed that duration of treatment, poor treatment compliance, drug abuse and homosexual orientation are the major risk factors for DR in this MSM population (all P < 0.05). Our results showed that DR-associated mutations in the HIV-1-infected MSM population increased significantly after ART. Furthermore, duration of treatment, poor treatment compliance, drug abuse and homosexual orientation were identified as the risk factors for DR in the MSM population from Henan Province in China.

Improved packing of protein side chains with parallel ant colonies.

INTRODUCTION: The accurate packing of protein side chains is important for many computational biology problems, such as ab initio protein structure prediction, homology modelling, and protein design and ligand docking applications. Many of existing solutions are modelled as a computational optimisation problem. As well as the design of search algorithms, most solutions suffer from an inaccurate energy function for judging whether a prediction is good or bad. Even if the search has found the lowest energy, there is no certainty of obtaining the protein structures with correct side chains. METHODS: We present a side-chain modelling method, pacoPacker, which uses a parallel ant colony optimisation strategy based on sharing a single pheromone matrix. This parallel approach combines different sources of energy functions and generates protein side-chain conformations with the lowest energies jointly determined by the various energy functions. We further optimised the selected rotamers to construct subrotamer by rotamer minimisation, which reasonably improved the discreteness of the rotamer library. RESULTS: We focused on improving the accuracy of side-chain conformation prediction. For a testing set of 442 proteins, 87.19% of X1 and 77.11% of X12 angles were predicted correctly within 40° of the X-ray positions. We compared the accuracy of pacoPacker with state-of-the-art methods, such as CIS-RR and SCWRL4. We analysed the results from different perspectives, in terms of protein chain and individual residues. In this comprehensive benchmark testing, 51.5% of proteins within a length of 400 amino acids predicted by pacoPacker were superior to the results of CIS-RR and SCWRL4 simultaneously. Finally, we also showed the advantage of using the subrotamers strategy. All results confirmed that our parallel approach is competitive to state-of-the-art solutions for packing side chains. CONCLUSIONS: This parallel approach combines various sources of searching intelligence and energy functions to pack protein side chains. It provides a frame-work for combining different inaccuracy/usefulness objective functions by designing parallel heuristic search algorithms.

Mitomycin C-soybean phosphatidylcholine complex-loaded self-assembled PEG-lipid-PLA hybrid nanoparticles for targeted drug delivery and dual-controlled drug release.

Most present drug-phospholipid delivery systems were based on a water-insoluble drug-phospholipid complex but rarely water-soluble drug-phospholipid complex. Mitomycin C (MMC) is a water-soluble anticancer drug extensively used in first-line chemotherapy but is limited by its poor aqueous stability in vitro, rapid elimination from the body, and lack of target specificity. In this article, we report the MMC-soybean phosphatidylcholine complex-loaded PEG-lipid-PLA hybrid nanoparticles (NPs) with Folate (FA) functionalization (FA-PEG-PE-PLA NPs@MMC-SPC) for targeted drug delivery and dual-controlled drug release. FA-PEG-PE-PLA NPs@MMC-SPC comprise a hydrophobic core (PLA) loaded with MMC-SPC, an amphiphilic lipid interface layer (PE), a hydrophilic shell (PEG), and a targeting ligand (FA) on the surface, with a spherical shape, a nanoscaled particle size, and high drug encapsulation efficiency of almost 95%. The advantage of the new drug delivery systems is the early phase controlled drug release by the drug-phospholipid complex and the late-phase controlled drug release by the pH-sensitive polymer-lipid hybrid NPs. In vitro cytotoxicity and hemolysis assays demonstrated that the drug carriers were cytocompatible and hemocompatible. The pharmacokinetics study in rats showed that FA-PEG-PE-PLA NPs@MMC-SPC significantly prolonged the blood circulation time compared to that of the free MMC. More importantly, FA-PEG-PE-PLA NPs@MMC-SPC presented the enhanced cell uptake/cytotoxicity in vitro and superior tumor accumulation/therapeutic efficacy in vivo while reducing the systemic toxicity. A significant accumulation of MMC in the nuclei as the site of MMC action achieved in FA-PEG-PE-PLA NPs@MMC-SPC made them ideal for MMC drug delivery. This study may provide an effective strategy for the design and development of the water-soluble drug-phospholipid complex-based targeted drug delivery and sustained/controlled drug release.

Therapeutic effect of folate-targeted and PEGylated phytosomes loaded with a mitomycin C-soybean phosphatidyhlcholine complex.

A mitomycin C (MMC)-soybean phosphatidyhlcholine complex loaded in phytosomes was previously reported for the purpose of developing a MMC drug delivery system (Mol. Pharmaceutics 2013, 10, 90-101), but this approach was limited by rapid elimination from the body and lack of target specificity. In this article, to overcome these limitations, MMC-soybean phosphatidyhlcholine complex-loaded phytosomes (MMC-loaded phytosomes) as drug carriers were surface-functionalized with folate-PEG (FA-PEG) to achieve reduced toxicity and a superior MMC-mediated therapeutic effect. For this purpose, FA was conjugated to DSPE-PEG-NH2, and the resultant DSPE-PEG-FA was introduced into the lipid moiety of the phytosomes via a postinsertion technique. The prepared FA-PEG-functionalized MMC-loaded phytosomes (FA-PEG-MMC-loaded phytosomes) have a particle size of 201.9 ± 2.4 nm, a PDI of 0.143 ± 0.010, a zeta potential of -27.50 ± 1.67 mV, a spherical shape, and sustained drug release. The remarkable features of FA-PEG-MMC-loaded phytosomes included increased cellular uptake in HeLa cells and higher accumulation in H22 tumor-bearing mice over that of the PEG-MMC-loaded phytosomes. Furthermore, FA-PEG-MMC-loaded phytosomes were associated with enhanced cytotoxic activity in vitro and an improved antitumor effect in vivo compared to that resulting from free MMC injection. These results suggest that FA-PEG-MMC-loaded phytosomes may be useful drug delivery systems for widening the therapeutic window of MMC in clinical trials.

GTAMP-DTA: Graph transformer combined with attention mechanism for drug-target binding affinity prediction.

Drug target affinity prediction (DTA) is critical to the success of drug development. While numerous machine learning methods have been developed for this task, there remains a necessity to further enhance the accuracy and reliability of predictions. Considerable bias in drug target binding prediction may result due to missing structural information or missing information. In addition, current methods focus only on simulating individual non-covalent interactions between drugs and proteins, thereby neglecting the intricate interplay among different drugs and their interactions with proteins. GTAMP-DTA combines special Attention mechanisms, assigning each atom or amino acid an attention vector. Interactions between drug forms and protein forms were considered to capture information about their interactions. And fusion transformer was used to learn protein characterization from raw amino acid sequences, which were then merged with molecular map features extracted from SMILES. A self-supervised pre-trained embedding that uses pre-trained transformers to encode drug and protein attributes is introduced in order to address the lack of labeled data. Experimental results demonstrate that our model outperforms state-of-the-art methods on both the Davis and KIBA datasets. Additionally, the model's performance undergoes evaluation using three distinct pooling layers (max-pooling, mean-pooling, sum-pooling) along with variations of the attention mechanism. GTAMP-DTA shows significant performance improvements compared to other methods.

Identification of DNA-protein binding residues through integration of Transformer encoder and Bi-directional Long Short-Term Memory.

DNA-protein binding is crucial for the normal development and function of organisms. The significance of accurately identifying DNA-protein binding sites lies in its role in disease prevention and the development of innovative approaches to disease treatment. In the present study, we introduce a precise and robust identifier for DNA-protein binding residues. In the context of protein representation, we combine the evolutionary information of the protein, represented by its position-specific scoring matrix, with the spatial information of the protein's secondary structure, enriching the overall informational content. This approach initially employs a combination of Bi-directional Long Short-Term Memory and Transformer encoder to jointly extract the interdependencies among residues within the protein sequence. Subsequently, convolutional operations are applied to the resulting feature matrix to capture local features of the residues. Experimental results on the benchmark dataset demonstrate that our method exhibits a higher level of competitiveness when compared to contemporary classifiers. Specifically, our method achieved an MCC of 0.349, SP of 96.50%, SN of 44.03% and ACC of 94.59% on the PDNA-41 dataset.