HOPEXGB: A Consensual Model for Predicting miRNA/lncRNA-Disease Associations Using a Heterogeneous Disease-miRNA-lncRNA Information Network.

Predicting disease-related microRNAs (miRNAs) and long noncoding RNAs (lncRNAs) is crucial to find new biomarkers for the prevention, diagnosis, and treatment of complex human diseases. Computational predictions for miRNA/lncRNA-disease associations are of great practical significance, since traditional experimental detection is expensive and time-consuming. In this paper, we proposed a consensual machine-learning technique-based prediction approach to identify disease-related miRNAs and lncRNAs by high-order proximity preserved embedding (HOPE) and eXtreme Gradient Boosting (XGB), named HOPEXGB. By connecting lncRNA, miRNA, and disease nodes based on their correlations and relationships, we first created a heterogeneous disease-miRNA-lncRNA (DML) information network to achieve an effective fusion of information on similarities, correlations, and interactions among miRNAs, lncRNAs, and diseases. In addition, a more rational negative data set was generated based on the similarities of unknown associations with the known ones, so as to effectively reduce the false negative rate in the data set for model construction. By 10-fold cross-validation, HOPE shows better performance than other graph embedding methods. The final consensual HOPEXGB model yields robust performance with a mean prediction accuracy of 0.9569 and also demonstrates high sensitivity and specificity advantages compared to lncRNA/miRNA-specific predictions. Moreover, it is superior to other existing methods and gives promising performance on the external testing data, indicating that integrating the information on lncRNA-miRNA interactions and the similarities of lncRNAs/miRNAs is beneficial for improving the prediction performance of the model. Finally, case studies on lung, stomach, and breast cancers indicate that HOPEXGB could be a powerful tool for preclinical biomarker detection and bioexperiment preliminary screening for the diagnosis and prognosis of cancers. HOPEXGB is publicly available at https://github.com/airpamper/HOPEXGB.

Exploring biased activation characteristics by molecular dynamics simulation and machine learning for the µ-opioid receptor.

Biased ligands selectively activating specific downstream signaling pathways (termed as biased activation) exhibit significant therapeutic potential. However, the conformational characteristics revealed are very limited for the biased activation, which is not conducive to biased drug development. Motivated by the issue, we combine extensive accelerated molecular dynamics simulations and an interpretable deep learning model to probe the biased activation features for two complex systems constructed by the inactive µOR and two different biased agonists (G-protein-biased agonist TRV130 and ß-arrestin-biased agonist endomorphin2). The results indicate that TRV130 binds deeper into the receptor core compared to endomorphin2, located between W2936.48 and D1142.50, and forms hydrogen bonding with D1142.50, while endomorphin2 binds above W2936.48. The G protein-biased agonist induces greater outward movements of the TM6 intracellular end, forming a typical active conformation, while the ß-arrestin-biased agonist leads to a smaller extent of outward movements of TM6. Compared with TRV130, endomorphin2 causes more pronounced inward movements of the TM7 intracellular end and more complex conformational changes of H8 and ICL1. In addition, important residues determining the two different biased activation states were further identified by using an interpretable deep learning classification model, including some common biased activation residues across Class A GPCRs like some key residues on the TM2 extracellular end, ECL2, TM5 intracellular end, TM6 intracellular end, and TM7 intracellular end, and some specific important residues of ICL3 for µOR. The observations will provide valuable information for understanding the biased activation mechanism for GPCRs.

Acquiring Charge-Transfer-Featured Single-Molecule Ultralong Organic Room Temperature Phosphorescence via Through-Space Electronic Coupling.

Although long-lived triplet charge-transfer (3 CT) state with high energy level has gained significant attention, the development of organic small molecules capable of achieving such states remains a major challenge. Herein, by using the through-space electronic coupling effect, we have developed a compound, namely NIC-DMAC, which has a long-lived 3 CT state at the single-molecule level with a lifetime of 210 ms and a high energy level of up to 2.50 eV. Through a combination of experimental and computational approaches, we have elucidated the photophysical processes of NIC-DMAC, which involve sequential transitions from the first singlet excited state (S1 ) that shows a 1 CT character to the first triplet excited state (T1 ) that exhibits a local excited state feature (3 LE), and then to the second triplet excited state (T2 ) that shows a 3 CT character (i.e., S1 (1 CT)→T1 (3 LE)→T2 (3 CT)). The long lifetime and high energy level of its 3 CT state have enabled NIC-DMAC as an initiator for photocuring in double patterning applications.

Attention Mechanism-Based Graph Neural Network Model for Effective Activity Prediction of SARS-CoV-2 Main Protease Inhibitors: Application to Drug Repurposing as Potential COVID-19 Therapy.

Compared to de novo drug discovery, drug repurposing provides a time-efficient way to treat coronavirus disease 19 (COVID-19) that is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). SARS-CoV-2 main protease (Mpro) has been proved to be an attractive drug target due to its pivotal involvement in viral replication and transcription. Here, we present a graph neural network-based deep-learning (DL) strategy to prioritize the existing drugs for their potential therapeutic effects against SARS-CoV-2 Mpro. Mpro inhibitors were represented as molecular graphs ready for graph attention network (GAT) and graph isomorphism network (GIN) modeling for predicting the inhibitory activities. The result shows that the GAT model outperforms the GIN and other competitive models and yields satisfactory predictions for unseen Mpro inhibitors, confirming its robustness and generalization. The attention mechanism of GAT enables to capture the dominant substructures and thus to realize the interpretability of the model. Finally, we applied the optimal GAT model in conjunction with molecular docking simulations to screen the Drug Repurposing Hub (DRH) database. As a result, 18 drug hits with best consensus prediction scores and binding affinity values were identified as the potential therapeutics against COVID-19. Both the extensive literature searching and evaluations on adsorption, distribution, metabolism, excretion, and toxicity (ADMET) illustrate the premium drug-likeness and pharmacokinetic properties of the drug candidates. Overall, our work not only provides an effective GAT-based DL prediction tool for inhibitory activity of SARS-CoV-2 Mpro inhibitors but also provides theoretical guidelines for drug discovery in the COVID-19 treatment.

General Graph Neural Network-Based Model To Accurately Predict Cocrystal Density and Insight from Data Quality and Feature Representation.

Cocrystal engineering as an effective way to modify solid-state properties has inspired great interest from diverse material fields while cocrystal density is an important property closely correlated with the material function. In order to accurately predict the cocrystal density, we develop a graph neural network (GNN)-based deep learning framework by considering three key factors of machine learning (data quality, feature presentation, and model architecture). The result shows that different stoichiometric ratios of molecules in cocrystals can significantly influence the prediction performances, highlighting the importance of data quality. In addition, the feature complementary is not suitable for augmenting the molecular graph representation in the cocrystal density prediction, suggesting that the complementary strategy needs to consider whether extra features can sufficiently supplement the lacked information in the original representation. Based on these results, 4144 cocrystals with 1:1 stoichiometry ratio are selected as the dataset, supplemented by the data augmentation of exchanging a pair of coformers. The molecular graph is determined to learn feature representation to train the GNN-based model. Global attention is introduced to further optimize the feature space and identify important atoms to realize the interpretability of the model. Benefited from the advantages, our model significantly outperforms three competitive models and exhibits high prediction accuracy for unseen cocrystals, showcasing its robustness and generality. Overall, our work not only provides a general cocrystal density prediction tool for experimental investigations but also provides useful guidelines for the machine learning application. All source codes are freely available at https://github.com/Xiao-Gua00/CCPGraph.

A consensual machine-learning-assisted QSAR model for effective bioactivity prediction of xanthine oxidase inhibitors using molecular fingerprints.

Xanthine oxidase inhibitors (XOIs) have been widely studied due to the promising potential as safe and effective therapeutics in hyperuricemia and gout. Currently, available XOI molecules have been developed from different experiments but they are with the wide structure diversity and significant varying bioactivities. So it is of great practical significance to present a consensual QSAR model for effective bioactivity prediction of XOIs based on a systematic compiling of these XOIs across different experiments. In this work, 249 XOIs belonging to 16 scaffolds were collected and were integrated into a consensual dataset by introducing the concept of IC50 values relative to allopurinol (RIC50). Here, extended connectivity fingerprints (ECFPs) were employed to represent XOI molecules. By performing effective feature selection by machine-learning method, 54 crucial fingerprints were indicated to be valuable for predicting the inhibitory potency (IP) of XOIs. The optimal predictor yields the promising performance by different cross-validation tests. Besides, an external validation of 43 XOIs and a case study on febuxostat also provide satisfactory results, indicating the powerful generalization of our predictor. Here, the predictor was interpreted by shapely additive explanation (SHAP) method which revealed several important substructures by mapping the featured fingerprints to molecular structures. Then, 15 new molecules were designed and predicted by our predictor to show superior IP than febuxostat. Finally, molecular docking simulation was performed to gain a deep insight into molecular binding mode with xanthine oxidase (XO) enzyme, showing that molecules with selenazole moiety, cyano group and isopropyl group tended to yield higher IP. The absorption, distribution, metabolism, excretion and toxicity (ADMET) prediction results further enhanced the potential of these novel XOIs as drug candidates. Overall, this work presents a QSAR model for accurate prediction of IP of XOIs, and is expected to provide new insights for further structure-guided design of novel XOIs.

The landscape of miRNA-related ceRNA networks for marking different renal cell carcinoma subtypes.

We know that different types of cancers usually have different responses to the same treatment. Therefore, it is important to understand the similarities and differences across subtypes of cancers, so as to provide a basis for the individualized treatments. Until now, no comprehensive investigation on competing endogenous RNAs (ceRNAs) has been reported for the three main subtypes of renal cell carcinoma (RCC), so the regulation characteristics of ceRNAs in three subtypes are not well revealed. This paper firstly describes a comparative analysis of ceRNA-ceRNA interaction networks for all the three subtypes of RCC based on differential microRNAs (miRNAs). We comprehensively summarized all miRNA and messenger RNAdata of RCC from 126 matched tumor-normal tissues in The Cancer Genome Atlas, systematically analyzed a total of more than 80 000 ceRNA interactions and highlighted the common and specific properties among them, aiming to identify critical genes to classify them for providing supplementary help in the precise diagnosis of RCC. From three aspects, including common or specific ceRNAs, upregulated or downregulated and classifications across the three subtypes, we highlighted the common and specific properties for the three subtypes and also explored the classification of RCC by combining the specific ceRNAs with differential regulations. Moreover, for the most major subtype of clear cell renal cell carcinoma (KIRC), three critical genes were screened out from KIRC ceRNA network and further demonstrated to be the potential biomarkers of KIRC by performing biological experiments at the transcriptional level.

Molecular Mechanisms of Diverse Activation Stimulated by Different Biased Agonists for the ß2-Adrenergic Receptor.

ß2AR is an important drug target protein involving many diseases. Biased drugs induce specific signaling and provide additional clinical utility to optimize ß2AR-based therapies. However, the biased signaling mechanism has not been elucidated. Motivated by the issue, we chose four agonists with divergent bias (balanced agonist, G-protein-biased agonist, and ß-arrestin-biased agonists) and utilized Gaussian accelerated molecular dynamics simulation coupled with a dynamic network to probe the molecular mechanisms of distinct biased activation induced by the structural differences between the four agonists. Our simulations reveal that the G-protein-biased agonist induces an open conformation with the outward shifts of TM6 and TM7 for the intracellular domain, which will be beneficial to couple G protein. In contrast, the ß-arrestin-biased agonists regulate an occluded conformation with a slightly outward movement of TM6 and an inward shift of TM7, which should favor ß-arrestin signaling. The balanced agonist does not induce an observable outward shift for TM6 but, along with a slight tilt for TM7, leads to an inactive-like conformation. In addition, our results reveal the first time that ICL3 presents specific conformations with different agonists. The G-protein-biased agonist drives ICL3 to open so that the G protein-binding pocket can be available, while the ß-arrestin-biased agonists induce ICL3 to form a closed conformation with a stable local α-helix. MM/PBSA analysis further reveals that the hydroxyl groups in the resorcinol of the G-protein-biased agonist form strong interactions with Y5.38 and S5.42, thus preventing tilting of the TM5 extracellular end. The catechol of the balanced agonist and the ß-arrestin-biased ones induces the rearrangement of two hydrophobic residues F6.52 and W6.48. However, different from the balanced agonist, the ethyl substituent of ß-arrestin-biased agonists forms additional hydrophobic interactions with W6.48 and F6.51 after the rearrangement, which should contribute to the ß-arrestin bias. The shortest pathway analysis further reveals that the three residues Y7.43, N7.45, and N7.49 are crucial for allosterically regulating G-protein-biased signaling, while the two residues W6.48 and F6.44 make an important contribution to regulate ß-arrestin-biased signaling. For the balanced agonist NE, the allosteric regulation pathway simultaneously involves the residue associated with G-protein-biased signaling like S5.46 and the residues related to ß-arrestin-biased signaling like W6.48 and F6.44, thus producing unbiased signaling. The observations could advance our understanding of the biased activation mechanism on class A GPCRs and provide a useful guideline for the design of biased drugs.

Probing Allosteric Regulation Mechanism of W7.35 on Agonist-Induced Activity for µOR by Mutation Simulation.

The residue located at 15 positions before the most conserved residue in TM7 (7.35 of Ballesteros-Weinstein number) plays important roles in ligand binding and the receptor activity for class A GPCRs. Nevertheless, its regulation mechanism has not been clearly clarified in experiments, and some controversies also exist for its impact on µ-opioid receptors (µOR) bound by agonists. Thus, we chose the µ-opioid receptor (µOR) of class A GPCRs as a representative and utilized a microsecond accelerated molecular dynamics simulation (aMD) coupled with a protein structure network (PSN) to explore the effect of W3187.35 on its functional activity induced by the agonist endomorphin2 mainly by a comparison of the wild system and its W7.35A mutant. When endomorphin2 binds to the wild-type µOR, TM6 in µOR moves outward to form an open intracellular conformation that is beneficial to accommodating the ß-arrestin transducer, rather than the G-protein transducer due to the clash with the α5 helix of G-protein, thus acting as a ß-arrestin biased agonist. However, the W318A mutation induces the intracellular part of µOR to form a closed state, which disfavors coupling with either G-protein or ß-arrestin. The allosteric pathway analysis further reveals that the binding of endomorphin2 to the wild-type µOR transmits more activation signals to the ß-arrestin binding site while the W318A mutation induces more structural signals to transmit to common binding residues of the G protein and ß-arrestin. More interestingly, the residue at the 7.35 position regulates the shortest allosteric pathway in indirect ways by influencing the interactions between other ligand-binding residues and endomorphin2. W2936.48 and F2896.44 are important for regulating the different activities of µOR induced either by the agonist or by the mutation. Y3367.53, F3438.50, and D3408.47 play crucial roles in activating the ß-arrestin biased signal induced by the agonist endomorphin2, while L1583.43 and V2866.41 devote important contributions to the change in the activity of endomorphin2 from the ß-arrestin biased agonist to the antagonist upon the W318A mutation.

Biased Activation Mechanism Induced by GPCR Heterodimerization: Observations from µOR/δOR Dimers.

GPCRs regulate multiple intracellular signaling cascades. Biasedly activating one signaling pathway over the others provides additional clinical utility to optimize GPCR-based therapies. GPCR heterodimers possess different functions from their monomeric states, including their selectivity to different transducers. However, the biased signaling mechanism induced by the heterodimerization remains unclear. Motivated by the issue, we select an important GPCR heterodimer (µOR/δOR heterodimer) as a case and use microsecond Gaussian accelerated molecular dynamics simulation coupled with potential of mean force and protein structure network (PSN) to probe mechanisms regarding the heterodimerization-induced constitutive ß-arrestin activity and efficacy change of the agonist DAMGO. The results show that only the lowest energy state of the µOR/δOR heterodimer, which adopts a slightly outward shift of TM6 and an ICL2 conformation close to the receptor core, can selectively accommodate ß-arrestins. PSN further reveals important roles of H8, ICL1, and ICL2 in regulating the constitutive ß-arrestin-biased activity for the apo µOR/δOR heterodimer. In addition, the heterodimerization can allosterically alter the binding mode of DAMGO mainly by means of W7.35. Consequently, DAMGO transmits the structural signal mainly through TM6 and TM7 in the dimer, rather than TM3 similar to the µOR monomer, thus changing the efficacy of DAMGO from a balanced agonist to the ß-arrestin-biased one. On the other side, the binding of DAMGO to the heterodimer can stabilize µOR/δOR heterodimers through a stronger interaction of TM1/TM1 and H8/H8, accordingly enhancing the interaction of µOR with δOR and the binding affinity of the dimer to the ß-arrestin. The agonist DAMGO does not change main compositions of the regulation network from the dimer interface to the transducer binding pocket of the µOR protomer, but induces an increase in the structural communication of the network, which should contribute to the enhanced ß-arrestin coupling. Our observations, for the first time, reveal the molecular mechanism of the biased signaling induced by the heterodimerization for GPCRs, which should be beneficial to more comprehensively understand the GPCR bias signaling.

An Interpretable Convolutional Neural Network Framework for Analyzing Molecular Dynamics Trajectories: a Case Study on Functional States for G-Protein-Coupled Receptors.

Molecular dynamics (MD) simulations have made great contribution to revealing structural and functional mechanisms for many biomolecular systems. However, how to identify functional states and important residues from vast conformation space generated by MD remains challenging; thus an intelligent navigation is highly desired. Despite intelligent advantages of deep learning exhibited in analyzing MD trajectory, its black-box nature limits its application. To address this problem, we explore an interpretable convolutional neural network (CNN)-based deep learning framework to automatically identify diverse active states from the MD trajectory for G-protein-coupled receptors (GPCRs), named the ICNNMD model. To avoid the information loss in representing the conformation structure, the pixel representation is introduced, and then the CNN module is constructed to efficiently extract features followed by a fully connected neural network to realize the classification task. More importantly, we design a local interpretable model-agnostic explanation interpreter for the classification result by local approximation with a linear model, through which important residues underlying distinct active states can be quickly identified. Our model showcases higher than 99% classification accuracy for three important GPCR systems with diverse active states. Notably, some important residues in regulating different biased activities are successfully identified, which are beneficial to elucidating diverse activation mechanisms for GPCRs. Our model can also serve as a general tool to analyze MD trajectory for other biomolecular systems. All source codes are freely available at https://github.com/Jane-Liu97/ICNNMD for aiding MD studies.

Correlated RNN Framework to Quickly Generate Molecules with Desired Properties for Energetic Materials in the Low Data Regime.

Motivated by the challenging of deep learning on the low data regime and the urgent demand for intelligent design on highly energetic materials, we explore a correlated deep learning framework, which consists of three recurrent neural networks (RNNs) correlated by the transfer learning strategy, to efficiently generate new energetic molecules with a high detonation velocity in the case of very limited data available. To avoid the dependence on the external big data set, data augmentation by fragment shuffling of 303 energetic compounds is utilized to produce 500,000 molecules to pretrain RNN, through which the model can learn sufficient structure knowledge. Then the pretrained RNN is fine-tuned by focusing on the 303 energetic compounds to generate 7153 molecules similar to the energetic compounds. In order to more reliably screen the molecules with a high detonation velocity, the SMILE enumeration augmentation coupled with the pretrained knowledge is utilized to build an RNN-based prediction model, through which R2 is boosted from 0.4446 to 0.9572. The comparable performance with the transfer learning strategy based on an existing big database (ChEMBL) to produce the energetic molecules and drug-like ones further supports the effectiveness and generality of our strategy in the low data regime. High-precision quantum mechanics calculations further confirm that 35 new molecules present a higher detonation velocity and lower synthetic accessibility than the classic explosive RDX, along with good thermal stability. In particular, three new molecules are comparable to caged CL-20 in the detonation velocity. All the source codes and the data set are freely available at https://github.com/wangchenghuidream/RNNMGM.

Molecular insights into the allosteric coupling mechanism between an agonist and two different transducers for µ-opioid receptors.

G protein-coupled receptors (GPCRs) as the most important class of pharmacological targets regulate G-protein and ß-arrestin-mediated signaling through allosteric interplay, which are responsible for different biochemical and physiological actions like therapeutic efficacy and side effects. However, the allosteric mechanism underlying preferentially recruiting one transducer versus the other has been poorly understood, limiting drug design. Motivated by this issue, we utilize accelerated molecular dynamics simulation coupled with potential of mean force (PMF), molecular mechanics Poisson Boltzmann surface area (MM/PBSA) and protein structure network (PSN) to study two ternary complex systems of a representative class A GPCR (µ-opioid receptor (µOR)) bound by an agonist and one specific transducer (G-protein and ß-arrestin). The results show that no significant difference exists in the whole structure of µOR between two transducer couplings, but displays transducer-dependent changes in the intracellular binding region of µOR, where the ß-arrestin coupling results in a narrower crevice with TM7 inward movement compared with the G-protein. In addition, both the G-protein and ß-arrestin coupling can increase the binding affinity of the agonist to the receptor. However, the interactions between the agonist and µOR also exhibit transducer-specific changes, in particular for the interaction with ECL2 that plays an important role in recruiting ß-arrestin. The allosteric network analysis further indicates that Y1483.33, F1523.37, F1563.41, N1914.49, T1603.45, Y1062.42, W2936.48, F2896.44, I2485.54 and Y2525.58 play important roles in equally activating G-protein and ß-arrestin. In contrast, M1613.46 and R1653.50 devote important contributions to preferentially recruit G-protein while D1643.49 and R179ICL2 are revealed to be important for selectively activating ß-arrestin. The observations provide useful information for understanding the biased activation mechanism.

A Transfer-Learning-Based Deep Convolutional Neural Network for Predicting Leukemia-Related Phosphorylation Sites from Protein Primary Sequences.

As one of the most important post-translational modifications (PTMs), phosphorylation refers to the binding of a phosphate group with amino acid residues like Ser (S), Thr (T) and Tyr (Y) thus resulting in diverse functions at the molecular level. Abnormal phosphorylation has been proved to be closely related with human diseases. To our knowledge, no research has been reported describing specific disease-associated phosphorylation sites prediction which is of great significance for comprehensive understanding of disease mechanism. In this work, focusing on three types of leukemia, we aim to develop a reliable leukemia-related phosphorylation site prediction models by combing deep convolutional neural network (CNN) with transfer-learning. CNN could automatically discover complex representations of phosphorylation patterns from the raw sequences, and hence it provides a powerful tool for improvement of leukemia-related phosphorylation site prediction. With the largest dataset of myelogenous leukemia, the optimal models for S/T/Y phosphorylation sites give the AUC values of 0.8784, 0.8328 and 0.7716 respectively. When transferred learning on the small size datasets, the models for T-cell and lymphoid leukemia also give the promising performance by common sharing the optimal parameters. Compared with other five machine-learning methods, our CNN models reveal the superior performance. Finally, the leukemia-related pathogenesis analysis and distribution analysis on phosphorylated proteins along with K-means clustering analysis and position-specific conversation profiles on the phosphorylation site all indicate the strong practical feasibility of our easy-to-use CNN models.

Extracts of Knoxia roxburghii (Spreng.) M. A. Rau Induce Apoptosis in Human MCF-7 Breast Cancer Cells via Mitochondrial Pathways.

Knoxia roxburghii (Spreng.) M. A. Rau (KR) is a plant clinically used in traditional Chinese medicine (TCM) for the treatment of cancer. The study objectives were to examine the effects of KR extracts, petroleum ether (PET), ethyl acetate (EtoAc), butanol (n-BuOH), and H2O-soluble fractions (HSF) of the 75% EtOH extraction on A549 (non-small cell lung cancer), HepG2 (liver cancer), HeLa (cervical cancer), MCF-7 (breast cancer), and L02 (normal hepatocyte) cells. It was found that HSF exhibited the strongest cytotoxic activity against MCF-7 cells, and was accompanied by reduced mitochondrial transmembrane potential, increased levels of intra-cellular reactive oxygen species (ROS) and activated caspases, and upregulated pro-apoptotic and downregulated anti-apoptotic proteins. LC-MS analysis further showed that HSF primarily consisted of calycosin, aloe emodin, rein, maackiain, asperuloside, orientin, vicenin-2, and kaempferide, which have been mostly reported for anti-tumor activity in previous studies. In summary, the current study illustrated the effect, mechanism, and the potential major active components of KR against breast cancer.

Maize DNA Methylation in Response to Drought Stress Is Involved in Target Gene Expression and Alternative Splicing.

DNA methylation is important for plant growth, development, and stress response. To understand DNA methylation dynamics in maize roots under water stress (WS), we reanalyzed DNA methylation sequencing data to profile DNA methylation and the gene expression landscape of two inbred lines with different drought sensitivities, as well as two of their derived recombination inbred lines (RILs). Combined with genotyping-by-sequencing, we found that the inheritance pattern of DNA methylation between RILs and parental lines was sequence-dependent. Increased DNA methylation levels were observed under WS and the methylome of drought-tolerant inbred lines were much more stable than that of the drought-sensitive inbred lines. Distinctive differentially methylated genes were found among diverse genetic backgrounds, suggesting that inbred lines with different drought sensitivities may have responded to stress in varying ways. Gene body DNA methylation showed a negative correlation with gene expression but a positive correlation with exon splicing events. Furthermore, a positive correlation of a varying extent was observed between small interfering RNA (siRNA) and DNA methylation, which at different genic regions. The response of siRNAs under WS was consistent with the differential DNA methylation. Taken together, our data can be useful in deciphering the roles of DNA methylation in plant drought-tolerance variations and in emphasizing its function in alternative splicing.

[Overview and prospect of research and development cooperation in traditional medicine between China and Thailand].

Under the background of the Belt and Road Initiative, the exchange of traditional medicine has become inevitable. China and Thailand are amicable neighbors, and the cooperation between the two countries in the field of traditional medicine has become increasingly close in recent years. Nevertheless, on account of the differences in culture, region, politics, economy and so on, the two countries have common features and unique characteristics in the theoretical system of traditional medicine, quality standard control of medicinal materials, research and development and use of medicinal materials. This paper summarizes the similarities and differences as well as the development opportunities of traditional medicine between China and Thailand. The specific content involves the development history, resources, and use of medicinal resources in Thailand, the main achievements and existing problems of modern research of Thai medicine, the spread and development of Chinese medicine in Thailand, and the spread and development of Thai medicine in China. Furthermore, the paper outlines the recent situation of traditional medicine interflow and cooperation between the two countries, and predicts the prospects for cooperation and development of traditional medicine between China and Thailand in the context of the Belt and Road Initiative, especially in the joint research and development and the improvement of quality standards of important medicinal plant varieties commonly used by the two countries and circulated across the border. Through the exchange and mutual learning, we can step up the traditional medicine cooperation between China and Thailand, which will provide advantageous conditions for the safety of medicine use as well as political and social stability between the two countries.

Uncovering the prognostic gene signatures for the improvement of risk stratification in cancers by using deep learning algorithm coupled with wavelet transform.

BACKGROUND: The aim of gene expression-based clinical modelling in tumorigenesis is not only to accurately predict the clinical endpoints, but also to reveal the genome characteristics for downstream analysis for the purpose of understanding the mechanisms of cancers. Most of the conventional machine learning methods involved a gene filtering step, in which tens of thousands of genes were firstly filtered based on the gene expression levels by a statistical method with an arbitrary cutoff. Although gene filtering procedure helps to reduce the feature dimension and avoid overfitting, there is a risk that some pathogenic genes important to the disease will be ignored. RESULTS: In this study, we proposed a novel deep learning approach by combining a convolutional neural network with stationary wavelet transform (SWT-CNN) for stratifying cancer patients and predicting their clinical outcomes without gene filtering based on tumor genomic profiles. The proposed SWT-CNN overperformed the state-of-art algorithms, including support vector machine (SVM) and logistic regression (LR), and produced comparable prediction performance to random forest (RF). Furthermore, for all the cancer types, we firstly proposed a method to weight the genes with the scores, which took advantage of the representative features in the hidden layer of convolutional neural network, and then selected the prognostic genes for the Cox proportional-hazards regression. The results showed that risk stratifications can be effectively improved by using the identified prognostic genes as feature, indicating that the representative features generated by SWT-CNN can well correlate the genes with prognostic risk in cancers and be helpful for selecting the prognostic gene signatures. CONCLUSIONS: Our results indicated that gene expression-based SWT-CNN model can be an excellent tool for stratifying the prognostic risk for cancer patients. In addition, the representative features of SWT-CNN were validated to be useful for evaluating the importance of the genes in the risk stratification and can be further used to identify the prognostic gene signatures.

A Novel Detector Based on Convolution Neural Networks for Multiscale SAR Ship Detection in Complex Background.

Convolution neural network (CNN)-based detectors have shown great performance on ship detections of synthetic aperture radar (SAR) images. However, the performance of current models has not been satisfactory enough for detecting multiscale ships and small-size ones in front of complex backgrounds. To address the problem, we propose a novel SAR ship detector based on CNN, which consist of three subnetworks: the Fusion Feature Extractor Network (FFEN), Region Proposal Network (RPN), and Refine Detection Network (RDN). Instead of using a single feature map, we fuse feature maps in bottom-up and top-down ways and generate proposals from each fused feature map in FFEN. Furthermore, we further merge features generated by the region-of-interest (RoI) pooling layer in RDN. Based on the feature representation strategy, the CNN framework constructed can significantly enhance the location and semantics information for the multiscale ships, in particular for the small ships. On the other hand, the residual block is introduced to increase the network depth, through which the detection precision could be further improved. The public SAR ship dataset (SSDD) and China Gaofen-3 satellite SAR image are used to validate the proposed method. Our method shows excellent performance for detecting the multiscale and small-size ships with respect to some competitive models and exhibits high potential in practical application.

Molecular Mechanism Regarding Allosteric Modulation of Ligand Binding and the Impact of Mutations on Dimerization for CCR5 Homodimer.

In this work, we combined accelerated molecular dynamics (aMD) and conventional molecular dynamics (cMD) simulations coupled with the potential of mean force (PMF), correlation analysis, principal component analysis (PCA), and protein structure network (PSN) to study the effects of dimerization and the mutations of I52V and V150A on the CCR5 homodimer, in order to elucidate the mechanism regarding cooperativity of the ligand binding between two protomers and to address the controversy about the mutation-induced dimer-separation. The results reveal that the dimer with interface involved in TM1, TM2, TM3, and TM4 is stable for the CCR5 homodimer. The dimerization induces an asymmetric impact on the overall structure and the ligand-binding pocket. As a result, the two protomers exhibit an asymmetric binding to the maraviroc (one anti-HIV drug). The binding of one protomer to the drug is enhanced while the other is weakened. The PSN result further reveals the allosteric pathway of the ligand-binding pocket between the two protomers. Six important residues in the pathway were identified, including two residues unreported. The results from PMF, PCA, and the correlation analysis clearly indicate that the two mutations induce strong anticorrelation motions in the interface, finally leading to its separation. The observations from the work could advance our understanding of the structure of the G protein-coupled receptor dimers and implications for their functions.