Few-Shot Drug Synergy Prediction With a Prior-Guided Hypernetwork Architecture.

Predicting drug synergy is critical to tailoring feasible drug combination treatment regimens for cancer patients. However, most of the existing computational methods only focus on data-rich cell lines, and hardly work on data-poor cell lines. To this end, here we proposed a novel few-shot drug synergy prediction method (called HyperSynergy) for data-poor cell lines by designing a prior-guided Hypernetwork architecture, in which the meta-generative network based on the task embedding of each cell line generates cell line dependent parameters for the drug synergy prediction network. In HyperSynergy model, we designed a deep Bayesian variational inference model to infer the prior distribution over the task embedding to quickly update the task embedding with a few labeled drug synergy samples, and presented a three-stage learning strategy to train HyperSynergy for quickly updating the prior distribution by a few labeled drug synergy samples of each data-poor cell line. Moreover, we proved theoretically that HyperSynergy aims to maximize the lower bound of log-likelihood of the marginal distribution over each data-poor cell line. The experimental results show that our HyperSynergy outperforms other state-of-the-art methods not only on data-poor cell lines with a few samples (e.g., 10, 5, 0), but also on data-rich cell lines.

A social theory-enhanced graph representation learning framework for multitask prediction of drug-drug interactions.

Current machine learning-based methods have achieved inspiring predictions in the scenarios of mono-type and multi-type drug-drug interactions (DDIs), but they all ignore enhancive and depressive pharmacological changes triggered by DDIs. In addition, these pharmacological changes are asymmetric since the roles of two drugs in an interaction are different. More importantly, these pharmacological changes imply significant topological patterns among DDIs. To address the above issues, we first leverage Balance theory and Status theory in social networks to reveal the topological patterns among directed pharmacological DDIs, which are modeled as a signed and directed network. Then, we design a novel graph representation learning model named SGRL-DDI (social theory-enhanced graph representation learning for DDI) to realize the multitask prediction of DDIs. SGRL-DDI model can capture the task-joint information by integrating relation graph convolutional networks with Balance and Status patterns. Moreover, we utilize task-specific deep neural networks to perform two tasks, including the prediction of enhancive/depressive DDIs and the prediction of directed DDIs. Based on DDI entries collected from DrugBank, the superiority of our model is demonstrated by the comparison with other state-of-the-art methods. Furthermore, the ablation study verifies that Balance and Status patterns help characterize directed pharmacological DDIs, and that the joint of two tasks provides better DDI representations than individual tasks. Last, we demonstrate the practical effectiveness of our model by a version-dependent test, where 88.47 and 81.38% DDI out of newly added entries provided by the latest release of DrugBank are validated in two predicting tasks respectively.

Prediction of Drug-Drug Interaction Using an Attention-Based Graph Neural Network on Drug Molecular Graphs.

The treatment of complex diseases by using multiple drugs has become popular. However, drug-drug interactions (DDI) may give rise to the risk of unanticipated adverse effects and even unknown toxicity. Therefore, for polypharmacy safety it is crucial to identify DDIs and explore their underlying mechanisms. The detection of DDI in the wet lab is expensive and time-consuming, due to the need for experimental research over a large volume of drug combinations. Although many computational methods have been developed to predict DDIs, most of these are incapable of predicting potential DDIs between drugs within the DDI network and new drugs from outside the DDI network. In addition, they are not designed to explore the underlying mechanisms of DDIs and lack interpretative capacity. Thus, here we propose a novel method of GNN-DDI to predict potential DDIs by constructing a five-layer graph attention network to identify k-hops low-dimensional feature representations for each drug from its chemical molecular graph, concatenating all identified features of each drug pair, and inputting them into a MLP predictor to obtain the final DDI prediction score. The experimental results demonstrate that our GNN-DDI is suitable for each of two DDI predicting scenarios, namely the potential DDIs among known drugs in the DDI network and those between drugs within the DDI network and new drugs from outside DDI network. The case study indicates that our method can explore the specific drug substructures that lead to the potential DDIs, which helps to improve interpretability and discover the underlying interaction mechanisms of drug pairs.

Directed graph attention networks for predicting asymmetric drug-drug interactions.

It is tough to detect unexpected drug-drug interactions (DDIs) in poly-drug treatments because of high costs and clinical limitations. Computational approaches, such as deep learning-based approaches, are promising to screen potential DDIs among numerous drug pairs. Nevertheless, existing approaches neglect the asymmetric roles of two drugs in interaction. Such an asymmetry is crucial to poly-drug treatments since it determines drug priority in co-prescription. This paper designs a directed graph attention network (DGAT-DDI) to predict asymmetric DDIs. First, its encoder learns the embeddings of the source role, the target role and the self-roles of a drug. The source role embedding represents how a drug influences other drugs in DDIs. In contrast, the target role embedding represents how it is influenced by others. The self-role embedding encodes its chemical structure in a role-specific manner. Besides, two role-specific items, aggressiveness and impressionability, capture how the number of interaction partners of a drug affects its interaction tendency. Furthermore, the predictor of DGAT-DDI discriminates direction-specific interactions by the combination between two proximities and the above two role-specific items. The proximities measure the similarity between source/target embeddings and self-role embeddings. In the designated experiments, the comparison with state-of-the-art deep learning models demonstrates the superiority of DGAT-DDI across a direction-specific predicting task and a direction-blinded predicting task. An ablation study reveals how well each component of DGAT-DDI contributes to its ability. Moreover, a case study of finding novel DDIs confirms its practical ability, where 7 out of the top 10 candidates are validated in DrugBank.

deepMDDI: A deep graph convolutional network framework for multi-label prediction of drug-drug interactions.

It is crucial to identify DDIs and explore their underlying mechanism (e.g., DDIs types) for polypharmacy safety. However, the detection of DDIs in assays is still time-consuming and costly, due to the need for experimental search over a large space of drug combinations. Thus, many computational methods have been developed to predict DDIs, most of them focusing on whether a drug interacts with another or not. And a few deep learning-based methods address a more realistic screening task for identifying various DDI types, but they assume a DDI only triggers one pharmacological effect, while a DDI can trigger more types of pharmacological effects. Thus, here we proposed a novel end-to-end deep learning-based method (called deepMDDI) for the Multi-label prediction of Drug-Drug Interactions. deepMDDI contains an encoder derived from relational graph convolutional networks and a tensor-like decoder to uniformly model interactions. deepMDDI is not only efficient for DDI transductive prediction, but also inductive prediction. The experimental results show that our model is superior to other state-of-the-art deep learning-based methods. We also validated the power of deepMDDI in the DDIs multi-label prediction and found several new valid DDIs in the case study. In conclusion, deepMDDI is beneficial to uncover the mechanism and regularity of DDIs.

Network controllability-based algorithm to target personalized driver genes for discovering combinatorial drugs of individual patients.

Multiple driver genes in individual patient samples may cause resistance to individual drugs in precision medicine. However, current computational methods have not studied how to fill the gap between personalized driver gene identification and combinatorial drug discovery for individual patients. Here, we developed a novel structural network controllability-based personalized driver genes and combinatorial drug identification algorithm (CPGD), aiming to identify combinatorial drugs for an individual patient by targeting personalized driver genes from network controllability perspective. On two benchmark disease datasets (i.e. breast cancer and lung cancer datasets), performance of CPGD is superior to that of other state-of-the-art driver gene-focus methods in terms of discovery rate among prior-known clinical efficacious combinatorial drugs. Especially on breast cancer dataset, CPGD evaluated synergistic effect of pairwise drug combinations by measuring synergistic effect of their corresponding personalized driver gene modules, which are affected by a given targeting personalized driver gene set of drugs. The results showed that CPGD performs better than existing synergistic combinatorial strategies in identifying clinical efficacious paired combinatorial drugs. Furthermore, CPGD enhanced cancer subtyping by computationally providing personalized side effect signatures for individual patients. In addition, CPGD identified 90 drug combinations candidates from SARS-COV2 dataset as potential drug repurposing candidates for recently spreading COVID-19.

DPDDI: a deep predictor for drug-drug interactions.

BACKGROUND: The treatment of complex diseases by taking multiple drugs becomes increasingly popular. However, drug-drug interactions (DDIs) may give rise to the risk of unanticipated adverse effects and even unknown toxicity. DDI detection in the wet lab is expensive and time-consuming. Thus, it is highly desired to develop the computational methods for predicting DDIs. Generally, most of the existing computational methods predict DDIs by extracting the chemical and biological features of drugs from diverse drug-related properties, however some drug properties are costly to obtain and not available in many cases. RESULTS: In this work, we presented a novel method (namely DPDDI) to predict DDIs by extracting the network structure features of drugs from DDI network with graph convolution network (GCN), and the deep neural network (DNN) model as a predictor. GCN learns the low-dimensional feature representations of drugs by capturing the topological relationship of drugs in DDI network. DNN predictor concatenates the latent feature vectors of any two drugs as the feature vector of the corresponding drug pairs to train a DNN for predicting the potential drug-drug interactions. Experiment results show that, the newly proposed DPDDI method outperforms four other state-of-the-art methods; the GCN-derived latent features include more DDI information than other features derived from chemical, biological or anatomical properties of drugs; and the concatenation feature aggregation operator is better than two other feature aggregation operators (i.e., inner product and summation). The results in case studies confirm that DPDDI achieves reasonable performance in predicting new DDIs. CONCLUSION: We proposed an effective and robust method DPDDI to predict the potential DDIs by utilizing the DDI network information without considering the drug properties (i.e., drug chemical and biological properties). The method should also be useful in other DDI-related scenarios, such as the detection of unexpected side effects, and the guidance of drug combination.

Short hairpin RNA-mediated knockdown of nuclear factor erythroid 2-like 3 exhibits tumor-suppressing effects in hepatocellular carcinoma cells.

BACKGROUND: Hepatocellular carcinoma (HCC) is one of the most common malignant tumors with high mortality-to-incidence ratios. Nuclear factor erythroid 2-like 3 (NFE2L3), also known as NRF3, is a member of the cap 'n' collar basic-region leucine zipper family of transcription factors. NFE2L3 is involved in the regulation of various biological processes, whereas its role in HCC has not been elucidated. AIM: To explore the expression and biological function of NFE2L3 in HCC. METHODS: We analyzed the expression of NFE2L3 in HCC tissues and its correlation with clinicopathological parameters based on The Cancer Genome Atlas (TCGA) data portal. Short hairpin RNA (shRNA) interference technology was utilized to knock down NFE2L3 in vitro. Cell apoptosis, clone formation, proliferation, migration, and invasion assays were used to identify the biological effects of NFE2L3 in BEL-7404 and SMMC-7721 cells. The expression of epithelial-mesenchymal transition (EMT) markers was examined by Western blot analysis. RESULTS: TCGA analysis showed that NFE2L3 expression was significantly positively correlated with tumor grade, T stage, and pathologic stage. The qPCR and Western blot results showed that both the mRNA and protein levels of NFE2L3 were significantly decreased after shRNA-mediated knockdown in BEL-7404 and SMMC-7721 cells. The shRNA-mediated knockdown of NFE2L3 could induce apoptosis and inhibit the clone formation and cell proliferation of SMMC-7721 and BEL-7404 cells. NFE2L3 knockdown also significantly suppressed the migration, invasion, and EMT of the two cell lines. CONCLUSION: Our study showed that shRNA-mediated knockdown of NFE2L3 exhibited tumor-suppressing effects in HCC cells.

Increased apolipoprotein M induced by lack of scavenger receptor BI is not activated via HDL-mediated cholesterol uptake in hepatocytes.

BACKGROUND: Scavenger receptor BI (SR-BI) is a classic high-density lipoprotein (HDL) receptor, which mediates selective lipid uptake from HDL cholesterol esters (HDL-C). Apolipoprotein M (ApoM), as a component of HDL particles, could influence preß-HDL formation and cholesterol efflux. The aim of this study was to determine whether SR-BI deficiency influenced the expression of ApoM. METHODS: Blood samples and liver tissues were collected from SR-BI gene knockout mice, and serum lipid parameters, including total cholesterol (TC), triglyceride (TG), high and low-density lipoprotein cholesterol (HDL-C and LDL-C) and ApoM were measured. Hepatic ApoM and ApoAI mRNA levels were also determined. In addition, BLT-1, an inhibitor of SR-BI, was added to HepG2 cells cultured with cholesterol and HDL, under serum or serum-free conditions. The mRNA and protein expression levels of ApoM were detected by RT-PCR and western blot. RESULTS: We found that increased serum ApoM protein levels corresponded with high hepatic ApoM mRNA levels in both male and female SR-BI-/- mice. Besides, serum TC and HDL-C were also significantly increased. Treatment of HepG2 hepatoma cells with SR-BI specific inhibitor, BLT-1, could up-regulate ApoM expression in serum-containing medium but not in serum-free medium, even in the presence of HDL-C and cholesterol. CONCLUSIONS: Results suggested that SR-BI deficiency promoted ApoM expression, but the increased ApoM might be independent from HDL-mediated cholesterol uptake in hepatocytes.

Apolipoprotein M Protects Against Lipopolysaccharide-Induced Acute Lung Injury via Sphingosine-1-Phosphate Signaling.

It had been demonstrated that apolipoprotein M (apoM) is an important carrier of sphingosine-1-phosphate (S1P) in blood, and the S1P has critical roles in the pathogenesis of sepsis-induced acute lung injury (ALI). In the present study, we investigated whether apoM has beneficial effects in a mouse model after lipopolysaccharide (LPS)-induced ALI. Forty-eight mice were divided into two groups: male C57BL/6 wild-type (apoM+/+) group (n = 24) and apoM gene-deficient (apoM-/-) group (n = 24) and then randomly subdivided into four subgroups (n = 6 each) according to different intraperitoneal (i.p.) injection: control group, W146 group, LPS group, and LPS + W146 group. Serum levels of interleukin-1 beta (IL-1ß) and mRNA levels of IL-1ß, interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α), lung histology, wet/dry weight ratio, and immunohistochemistry were measured at 3 h after the baseline and compared in each group. Our results clearly demonstrated that IL-1ß mRNA levels and other inflammatory biomarkers were significantly increased in the lungs of LPS-induced ALI apoM-/- mice compared to those of the apoM+/+ mice. Moreover, when apoM+/+ mice were treated with W146, a S1P receptor (S1PR1) antagonist, these inflammatory biomarkers could be significantly upregulated by LPS-induced ALI. Therefore, it suggests that apoM-S1P-S1PR1 signaling might underlie the pathogenesis of ALI and apoM could have physiological benefits to alleviate LPS-induced ALI.

17ß-estradiol regulates the expression of apolipoprotein M through estrogen receptor α-specific binding motif in its promoter.

BACKGROUND: We have previously demonstrated that estrogen could significantly enhance expression of apolipoprotein M (apoM), whereas the molecular basis of its mechanism is not fully elucidated yet. To further investigate the mechanism behind the estrogen induced up-regulation of apoM expression. RESULTS: Our results demonstrated either free 17ß-estradiol (E2) or membrane-impermeable bovine serum albumin-conjugated E2 (E2-BSA) could modulate human apoM gene expression via the estrogen receptor alpha (ER-α) pathway in the HepG2 cells. Moreover, experiments with the luciferase activity analysis of truncated apoM promoters could demonstrate that a regulatory region (from-1580 to -1575 bp (-GGTCA-)) upstream of the transcriptional start site of apoM gene was essential for the basal transcriptional activity that regulated by the ER-α. With the applications of an electrophoresis mobility shift assay and a chromatin immunoprecipitation assay, we could successfully identify a specific ER-α binding element in the apoM promoter region. CONCULSION: In summary, the present study indicates that 17ß-estradiol induced up-regulation of apoM in HepG2 cells is through an ER-α-dependent pathway involving ER-α binding element in the promoter of the apoM gene.

[Spatial variability of soil nutirients and their relations with rice yield at village scale in karst mountainous area of Guizhou Province, Southwest China].

By using geo-statistics methods, an investigation was conducted on the spatial variability of soil pH, organic matter, total N, P, and K, and available N, P, K, Cu, Zn, Fe, and Mn and their relations with rice yield in Wenxing Village, Anshun City of Guizhou Province, Southwest China. The C0/(C0 + C) ratios for the soil pH, total P, and available N, P, K, and Cu and the rice yield components were lower than 25%, indicating their strong spatial correlations, while the C0/ (C0 + C) ratios for the soil organic matter, total N and K, and available Zn, Fe, and Mn and the rice yield were 25%-75%, showing a medium spatial correlation. Of all the soil nutrients, the soil available K had the closest relation with rice yield (r = 0.4669, P < 0.0001). The direct path coefficients of the soil available N, K and P to the effective panicle and thousand-grain mass were positive, in line with the partial correlation analysis. The Kriging interpolation showed that the soil organic matter, total N, and available N, K, Cu, and Zn contents presented a decreasing trend from the southwest to northeast, but the rice yield was higher in the northwest and southeast of the Wenxing Village.