DrugSK: A Stacked Ensemble Learning Framework for Predicting Drug Combinations of Multiple Diseases.

Combination therapy is an important direction of continuous exploration in the field of medicine, with the core goals of improving treatment efficacy, reducing adverse reactions, and optimizing clinical outcomes. Machine learning technology holds great promise in improving the prediction of drug synergy combinations. However, most studies focus on single disease-oriented collaborative predictive models or involve excessive feature categories, making it challenging to predict the majority of new drugs. To address these challenges, the DrugSK comprehensive model was developed, which utilizes SMILES-BERT to extract structural information from 3492 drugs and trains on reactions from 48,756 drug combinations. DrugSK is an integrated learning model capable of predicting interactions among various drug categories. First, the primary learner is trained from the initial data set. Random forest, support vector machine, and XGboost model are selected as primary learners and logistic regression as secondary learners. A new data set is then "generated" to train level 2 learners, which can be thought of as a prediction for each model. Finally, the results are filtered using logistic regression. Furthermore, the combination of the new antibacterial drug Drafloxacin with other antibacterial agents was tested. The synergistic effect of Drafloxacin and Isavuconazonium in the fight against Candida albicans has been confirmed, providing enlightenment for the clinical treatment of skin infection. DrugSK's prediction is accurate in practical application and can also predict the probability of the outcome. In addition, the tendency of Drafloxacin and antifungal drugs to be synergistic was found. The development of DrugSK will provide a new blueprint for predicting drug combination synergies.

Co-Amorphization of Acemetacin with Basic Amino Acids as Co-Formers for Solubility Improvement and Gastric Ulcer Mitigation.

Acemetacin (ACM) is a new non-steroidal anti-inflammatory drug with anti-inflammatory, analgesic, and antipyretic effects. However, the poor water solubility and gastrointestinal side effects limit its use. Recently, the co-amorphous (CAM) strategy has attracted great interest to improve solubility for poorly water-soluble drugs, and basic amino acids have the potential to protect the gastrointestinal tract. In order to develop a highly efficient and low-toxic ACM formulation, we prepared ACM CAM systems, with basic amino acids (lysine, arginine, and histidine) as co-formers, using a cryo-milling method. The solid-state behaviors of the ACM CAM systems were characterized by polarizing light microscopy, differential scanning calorimetry, and powder X-ray diffraction. Fourier transform infrared spectroscopy and molecular docking were carried out to understand the formation mechanism. Moreover, the gastro-protective effects of ACM CAM systems were evaluated in a rat gastric ulcer model. The results demonstrated that the CAM systems improved the dissolution rates of ACM compared with the neat amorphous counterpart. Furthermore, ACM CAM systems are significantly effective in mitigating the ACM-induced gastric ulcer in rats, and the ulcer inhibition rates were almost 90%. More importantly, this study provided a useful method for mitigating drug-induced gastrointestinal damage and broadened the applications of drug-amino acid CAM systems.

Ensemble learning-based gene signature and risk model for predicting prognosis of triple-negative breast cancer.

Although medical science has been fully developed, due to the high heterogeneity of triple-negative breast cancer (TNBC), it is still difficult to use reasonable and precise treatment. In this study, based on local optimization-feature screening and genomics screening strategy, we screened 25 feature genes. In multiple machine learning algorithms, feature genes have excellent discriminative diagnostic performance among samples composed of multiple large datasets. After screening at the single-cell level, we identified genes expressed substantially in myeloid cells (MCGs) that have a potential association with TNBC. Based on MCGs, we distinguished two types of TNBC patients who showed considerable differences in survival status and immune-related characteristics. Immune-related gene risk scores (IRGRS) were established, and their validity was verified using validation cohorts. A total of 25 feature genes were obtained, among which CXCL9, CXCL10, CCL7, SPHK1, and TREM1 were identified as the result after single-cell level analysis and screening. According to these entries, the cohort was divided into MCA and MCB subtypes, and the two subtypes had significant differences in survival status and tumor-immune microenvironment. After Lasso-Cox screening, IDO1, GNLY, IRF1, CTLA4, and CXCR6 were selected for constructing IRGRS. There were significant differences in drug sensitivity and immunotherapy sensitivity among high-IRGRS and low-IRGRS groups. We revealed the dynamic relationship between TNBC and TIME, identified a potential biomarker called Granulysin (GNLY) related to immunity, and developed a multi-process machine learning package called "MPMLearning 1.0" in Python.

A bibliometric and visualized analysis of liver fibrosis from 2002 to 2022.

Fibrosis of the liver is a degenerative alteration that occurs in the majority of chronic liver disorders. Further progression can lead to cirrhosis, liver failure, and hepatocellular carcinoma, which can seriously affect the health and lives of patients. The field of liver fibrosis research has flourished in the last 20 years, with approximately 9000 articles retrieved from the Web of Science Core Collection database alone. In order to identify future research hotspots and potential paths in a thorough and scientifically reliable manner, it is important to organize and visualize the research on this topic from a holistic and very general perspective. This study used bibliometric analysis with CiteSpace and VOSviewer software to provide a quantitative analysis, hotspot mining, and commentary of articles published in the field of liver fibrosis over the last 20 years. This bibliometric analysis contains a total of 8994 articles with 45667 authors from 6872 institutions in 97 countries, published in 1371 journals and citing 156 309 references. The literature volume has steadily increased over the last 20 years. Research has focused on gastroenterology and hepatology, pharmacology and pharmacy, and medicine, research, and experimental areas. We found that the pathological mechanisms, diagnostic and quantitative methods, etiology, and antifibrotic strategies constitute the knowledge structure of liver fibrosis. Finding mechanisms for liver fibrosis regression, identifying precise noninvasive diagnostic and prognostic biomarkers, and creating efficient liver fibrosis patient treatments are the main goals of current research.

Artificial intelligence-driven prediction of multiple drug interactions.

When a drug is administered to exert its efficacy, it will encounter multiple barriers and go through multiple interactions. Predicting the drug-related multiple interactions is critical for drug development and safety monitoring because it provides foundations for practical, safe compatibility and rational use of multiple drugs. With the progress of artificial intelligence (AI) technology, a variety of novel prediction methods for single interaction have emerged and shown great advantages compared to the traditional, expensive and time-consuming laboratory research. To promote the comprehensive and simultaneous predictions of multiple interactions, we systematically reviewed the application of AI in drug-drug, drug-food (excipients) and drug-microbiome interactions. We began by outlining the model methods, evaluation indicators, algorithms and databases commonly used to build models for three types of drug interactions. The models based on the metabolic enzyme P450, drug similarity and drug targets have empathized among the machine learning models of drug-drug interactions. In particular, we discussed the limitations of current approaches and identified potential areas for future research. It is anticipated the in-depth review will be helpful for the development of the next-generation of systematic prediction models for simultaneous multiple interactions.

A metabolomic study based on accurate mass and isotopic fine structures by dual mode combined-FT-ICR-MS to explore the effects of Rhodiola crenulata extract on Alzheimer disease in rats.

A metabolomic strategy based on accurate mass and isotopic fine structures (IFSs) by dual mode combined-Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) was established to explore the effects of Rhodiola crenulata extract (RCE) on Alzheimer disease (AD) in rats. Experimental AD model was induced in rats by bilateral hippocampal injection of Aß1-42, and Morris water maze task (MWM) was used to evaluate the effects of RCE on AD. Subsequently, the metabolomic study was performed using HPLC-FT-ICR-MS, fraction collector and direct infusion (DI)-FT-ICR-MS to screen and identify the potential biomarkers. A total of 20 metabolites contributing to AD progress were identified, and 17 metabolites of them were restored to the control-like levels after RCE treatment (daily dose: 2.24 g/kg). The metabolic pathway analysis revealed that the disturbed pathways including tryptophan metabolism, sphingolipid metabolism and glycerophospholipid metabolism in AD model rats were regulated after high dose RCE application. It is the first time that the dual mode combined-FT-ICR-MS based metabolomic strategy was applied to biochemically profile the serum metabolic pathways of AD rats affected by RCE. These outcomes provide reliable evidence to illuminate the biochemical mechanisms of AD and facilitate investigation of the therapeutic benefits of RCE in AD treatment. Notably, it indicated that the developed method based on accurate mass and IFSs has sufficient performance for identification of biomarkers in metabolomic studies.

Characterization of global metabolic profile of Rhodiola crenulata after oral administration in rat plasma, urine, bile and feces based on UHPLC-FT-ICR MS.

Rhodiola crenulata has been widely used as a health food, antifatigue and antidepressant in China and many other countries for centuries. However, to date the metabolism of it in vivo still remains unclear. In this study, UHPLC-FT-ICR MS was used to analyze the major components and their metabolites in rats after oral administration of Rhodiola crenulata for the first time. A total of 179 constituents, including 37 prototype compounds and 142 metabolites (89 phase I metabolites and 53 phase II metabolites) were tentatively identified. The metabolic pathways included hydroxylation, deglycosylation, dehydrogenation, glucuronidation and sulphate conjugation. In summary, this study showed an insight into the metabolism of Rhodiola crenulata in vivo, which may provide helpful chemical information for better understanding the multiple functions of it. And also, the developed method could be used as a reliable strategy to study the metabolic profile for other traditional chinese medicines.