IHP-PING-generating integrated human protein-protein interaction networks on-the-fly

SynerGNet: A Graph Neural Network Model to Predict Anticancer Drug Synergy

Drug combination therapy shows promise in cancer treatment by addressing drug resistance, reducing toxicity, and enhancing therapeutic efficacy. However, the intricate and dynamic nature of biological systems makes identifying potential synergistic drugs a costly and time-consuming endeavor. To facilitate the development of combination therapy, techniques employing artificial intelligence have emerged as a transformative solution, providing a sophisticated avenue for advancing existing therapeutic approaches. In this study, we developed SynerGNet, a graph neural network model designed to accurately predict the synergistic effect of drug pairs against cancer cell lines. SynerGNet utilizes cancer-specific featured graphs created by integrating heterogeneous biological features into the human protein-protein interaction network, followed by a reduction process to enhance topological diversity. Leveraging synergy data provided by AZ-DREAM Challenges, the model yields a balanced accuracy of 0.68, significantly outperforming traditional machine learning. Encouragingly, augmenting the training data with carefully constructed synthetic instances improved the balanced accuracy of SynerGNet to 0.73. Finally, the results of an independent validation conducted against DrugCombDB demonstrated that it exhibits a strong performance when applied to unseen data. SynerGNet shows a great potential in detecting drug synergy, positioning itself as a valuable tool that could contribute to the advancement of combination therapy for cancer treatment.

Augmented drug combination dataset to improve the performance of machine learning models predicting synergistic anticancer effects

Combination therapy has gained popularity in cancer treatment as it enhances the treatment efficacy and overcomes drug resistance. Although machine learning (ML) techniques have become an indispensable tool for discovering new drug combinations, the data on drug combination therapy currently available may be insufficient to build high-precision models. We developed a data augmentation protocol to unbiasedly scale up the existing anti-cancer drug synergy dataset. Using a new drug similarity metric, we augmented the synergy data by substituting a compound in a drug combination instance with another molecule that exhibits highly similar pharmacological effects. Using this protocol, we were able to upscale the AZ-DREAM Challenges dataset from 8798 to 6,016,697 drug combinations. Comprehensive performance evaluations show that ML models trained on the augmented data consistently achieve higher accuracy than those trained solely on the original dataset. Our data augmentation protocol provides a systematic and unbiased approach to generating more diverse and larger-scale drug combination datasets, enabling the development of more precise and effective ML models. The protocol presented in this study could serve as a foundation for future research aimed at discovering novel and effective drug combinations for cancer treatment.

Augmented drug combination dataset to improve the performance of machine learning models predicting synergistic anticancer effects

Combination therapy has gained popularity in cancer treatment as it enhances the treatment efficacy and overcomes drug resistance. Although machine learning (ML) techniques have become an indispensable tool for discovering new drug combinations, the data on drug combination therapy currently available may be insufficient to build high-precision models. We developed a data augmentation protocol to unbiasedly scale up the existing anti-cancer drug synergy dataset. Using a new drug similarity metric, we augmented the synergy data by substituting a compound in a drug combination instance with another molecule that exhibits highly similar pharmacological effects. Using this protocol, we were able to upscale the AZ-DREAM Challenges dataset from 8,798 to 6,016,697 drug combinations. Comprehensive performance evaluations show that Random Forest and Gradient Boosting Trees models trained on the augmented data achieve higher accuracy than those trained solely on the original dataset. Our data augmentation protocol provides a systematic and unbiased approach to generating more diverse and larger-scale drug combination datasets, enabling the development of more precise and effective ML models. The protocol presented in this study could serve as a foundation for future research aimed at discovering novel and effective drug combinations for cancer treatment.

Exploring the mechanism of aloe-emodin in the treatment of liver cancer through network pharmacology and cell experiments

Objective: Aloe-emodin (AE) is an anthraquinone compound extracted from the rhizome of the natural plant rhubarb. Initially, it was shown that AE exerts an anti-inflammatory effect. Further studies revealed its antitumor activity against various types of cancer. However, the mechanisms underlying these properties remain unclear. Based on network pharmacology and molecular docking, this study investigated the molecular mechanism of AE in the treatment of hepatocellular carcinoma (HCC), and evaluated its therapeutic effect through in vitro experiments. Methods: CTD, Pharmmapper, SuperPred and TargetNet were the databases to obtain potential drug-related targets. DisGenet, GeneCards, OMIM and TTD were used to identify potential disease-related targets. Intersection genes for drugs and diseases were obtained through the Venn diagram. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of intersecting genes were conducted by the website of Bioinformatics. Intersection genes were introduced into STRING to construct a protein-protein interaction network, while the Cytoscape3.9.1 software was used to visualize and analyze the core targets. AutoDock4.2.6 was utilized to achieve molecular docking between drug and core targets. In vitro experiments investigated the therapeutic effects and related mechanisms of AE. Results: 63 overlapped genes were obtained and GO analysis generated 3,646 entries by these 63 intersecting genes. KEGG analysis mainly involved apoptosis, proteoglycans in cancer, TNF signaling pathway, TP53 signaling pathway, PI3K-AKT signaling pathway, etc. AKT1, EGFR, ESR1, TP53, and SRC have been identified as core targets because the binding energies of them between aloe-emodin were less than -5 kcal/Mol.The mRNA and protein expression, prognosis, mutation status, and immune infiltration related to core targets were further revealed. The involvement of AKT1 and EGFR, as well as the key target of the PI3K-AKT signaling pathway, indicated the importance of this signaling pathway in the treatment of HCC using AE. The results of the Cell Counting Kit-8 assay and flow analysis demonstrated the therapeutic effect of AE. The downregulation of EGFR, PI3KR1, AKT1, and BCL2 in mRNA expression and PI3KR1, AKT,p-AKT in protein expression confirmed our hypothesis. Conclusion: Based on network pharmacology and molecular docking, our study initially showed that AE exerted a therapeutic effect on HCC by modulating multiple signaling pathways. Various analyses confirmed the antiproliferative activity and pro-apoptotic effect of AE on HCC through the PI3K-AKT signaling pathway. This study revealed the therapeutic mechanism of AE in the treatment of HCC through a novel approach, providing a theoretical basis for the clinical application of AE.

Reviewing and assessing existing meta-analysis models and tools

Over the past few years, meta-analysis has become popular among biomedical researchers for detecting biomarkers across multiple cohort studies with increased predictive power. Combining datasets from different sources increases sample size, thus overcoming the issue related to limited sample size from each individual study and boosting the predictive power. This leads to an increased likelihood of more accurately predicting differentially expressed genes/proteins or significant biomarkers underlying the biological condition of interest. Currently, several meta-analysis methods and tools exist, each having its own strengths and limitations. In this paper, we survey existing meta-analysis methods, and assess the performance of different methods based on results from different datasets as well as assessment from prior knowledge of each method. This provides a reference summary of meta-analysis models and tools, which helps to guide end-users on the choice of appropriate models or tools for given types of datasets and enables developers to consider current advances when planning the development of new meta-analysis models and more practical integrative tools.

Investigations of Kidney Dysfunction-Related Gene Variants in Sickle Cell Disease Patients in Cameroon (Sub-Saharan Africa)

BACKGROUND

Renal dysfunctions are associated with increased morbidity and mortality in sickle cell disease (SCD). Early detection and subsequent management of SCD patients at risk for renal failure and dysfunctions are essential, however, predictors that can identify patients at risk of developing renal dysfunction are not fully understood.

METHODS

In this study, we have investigated the association of 31 known kidney dysfunctions-related variants detected in African Americans from multi-ethnic genome wide studies (GWAS) meta-analysis, to kidney-dysfunctions in a group of 413 Cameroonian patients with SCD. Systems level bioinformatics analyses were performed, employing protein-protein interaction networks to further interrogate the putative associations.

RESULTS

Up to 61% of these patients had micro-albuminuria, 2.4% proteinuria, 71% glomerular hyperfiltration, and 5.9% had renal failure. Six variants are significantly associated with the two quantifiable phenotypes of kidney dysfunction (eGFR and crude-albuminuria): A1CF-rs10994860 (P = 0.02020), SYPL2-rs12136063 (P = 0.04208), and APOL1 (G1)-rs73885319 (P = 0.04610) are associated with eGFR; and WNT7A-rs6795744 (P = 0.03730), TMEM60-rs6465825 (P = 0.02340), and APOL1 (G2)-rs71785313 (P = 0.03803) observed to be protective against micro-albuminuria. We identified a protein-protein interaction sub-network containing three of these gene variants: APOL1, SYPL2, and WNT7A, connected to the Nuclear factor NF-kappa-B p105 subunit (NFKB1), revealed to be essential and might indirectly influence extreme phenotypes. Interestingly, clinical variables, including body mass index (BMI), systolic blood pressure, vaso-occlusive crisis (VOC), and haemoglobin (Hb), explain better the kidney phenotypic variations in this SCD population.

CONCLUSION

This study highlights a strong contribution of haematological indices (Hb level), anthropometric variables (BMI, blood pressure), and clinical events (i.e., vaso-occlusive crisis) to kidney dysfunctions in SCD, rather than known genetic factors. Only 6/31 characterised gene-variants are associated with kidney dysfunction phenotypes in SCD samples from Cameroon. The data reveal and emphasise the urgent need to extend GWAS studies in populations of African ancestries living in Africa, and particularly for kidney dysfunctions in SCD.