Prediction of ADMET Properties of Anti-Breast Cancer Compounds Using Three Machine Learning Algorithms.

In recent years, machine learning methods have been applied successfully in many fields. In this paper, three machine learning algorithms, including partial least squares-discriminant analysis (PLS-DA), adaptive boosting (AdaBoost), and light gradient boosting machine (LGBM), were applied to establish models for predicting the Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET for short) properties, namely Caco-2, CYP3A4, hERG, HOB, MN of anti-breast cancer compounds. To the best of our knowledge, the LGBM algorithm was applied to classify the ADMET property of anti-breast cancer compounds for the first time. We evaluated the established models in the prediction set using accuracy, precision, recall, and F1-score. Compared with the performance of the models established using the three algorithms, the LGBM yielded most satisfactory results (accuracy > 0.87, precision > 0.72, recall > 0.73, and F1-score > 0.73). According to the obtained results, it can be inferred that LGBM can establish reliable models to predict the molecular ADMET properties and provide a useful tool for virtual screening and drug design researchers.

An accurate and reliable analytical strategy for simultaneous determination of target furanocoumarins and flavonoids in cosmetic and pharmaceutical samples by ultra-high performance supercritical fluid chromatography.

Furanocoumarins and flavonoids have various important biological activities and wide application. In the present study, a rapid and reliable supercritical fluid chromatography method was proposed for the separation of 10 target components including 8 furanocoumarins and 2 flavonoids. After detailed condition optimization, the 10 target compounds can be baseline separated on a Trefoil CEL1 (3.0 mm × 150 mm, 2.5 µm) column using gradient elution. A 0.07% (v/v) trifluoroacetic acid in ethanol was determined to be the most proper mobile phase for the separation of target compounds. The column temperature, back pressure, flow rate were set at 36 â„ƒ, 2000 psi, 1.0 mL min-1 to 1.4 mL min-1, respectively. The ten target compounds were analyzed within 24 min using the optimized conditions. Under the optimized conditions, all the target compounds showed good linearity with linear correlation coefficients higher than 0.995, and satisfactory recovery in the range of 83.52-112.92%. All these results showed that the developed ultra-high performance supercritical fluid chromatography method was reliable and effective. Finally, the application of the developed method to cosmetic, Psoraleae fructus and Angelicae dahuricae radix samples were presented. The results highlight the applicability of the ultra-high performance supercritical fluid chromatography method to the analysis of interested compounds in pharmaceutical and cosmetic samples.

A hybrid variable selection and modeling strategy for the determination of target compounds in different spectral datasets.

In this paper, a hybrid technique is proposed to establish quantitative models for the determination of target compounds in different spectral datasets. The proposed hybrid method is the hybridization of interval partial least squares (iPLS) method with gradient descent (GD) algorithm. Here, the novelty of the proposed method is that the iPLS method is applied to variable selection and the GD algorithm is employed to establish quantitative models based on the selected optimal variables. In the application of the hybrid iPLS-GD method, the factors, i.e., the number of the interval for the iPLS method and the learning rate, the number of iterations for the GD method, that affect the quantitative accuracy of the method are optimized and determined. Then three spectral datasets, including the near-infrared spectroscopy (NIR) dataset, nuclear magnetic resonance (1H NMR) dataset and excitation-emission matrix fluorescence (EEM) dataset, are used to test and verify the performance of the iPLS-GD method. We compare the hybrid iPLS-GD method with the PLS and iPLS methods from the aspects of modeling ability and predictive ability. The results demonstrated that the iPLS-GD method can be used as an effective and promising tool for the determination of target compounds in complex samples in practice.

Relationship between Circle of Willis Variations and Cerebral or Cervical Arteries Stenosis Investigated by Computer Tomography Angiography and Multitask Convolutional Neural Network.

Circle of Willis (CoW) is the most critical collateral pathway that supports the redistribution of blood supply in the brain. The variation of CoW is closely correlated with cerebral hemodynamic and cerebral vessel-related diseases. But what is responsible for CoW variation remains unclear. Moreover, the visual evaluation for CoW variation is highly time-consuming. In the present study, based on the computer tomography angiography (CTA) dataset from 255 patients, the correlation between the CoW variations with age, gender, and cerebral or cervical artery stenosis was investigated. A multitask convolutional neural network (CNN) was used to segment cerebral arteries automatically. The results showed the prevalence of variation of the anterior communicating artery (Aco) was higher in the normal senior group than in the normal young group and in females than in males. The changes in the prevalence of variations of individual segments were not demonstrated in the population with stenosis of the afferent and efferent arteries, so the critical factors for variation are related to genetic or physiological factors rather than pathological lesions. Using the multitask CNN model, complete cerebral and cervical arteries could be segmented and reconstructed in 120 seconds, and an average Dice coefficient of 78.2% was achieved. The segmentation accuracy for precommunicating part of anterior cerebral artery and posterior cerebral artery, the posterior communicating arteries, and Aco in CoW was 100%, 99.2%, 94%, and 69%, respectively. Artificial intelligence (AI) can be considered as an adjunct tool for detecting the CoW, particularly related to reducing workload and improving the accuracy of the visual evaluation. The study will serve as a basis for the following research to determine an individual's risk of stroke with the aid of AI.

Surgical stress induces brain-derived neurotrophic factor reduction and postoperative cognitive dysfunction via glucocorticoid receptor phosphorylation in aged mice.

AIMS: This study explored whether surgical stress-induced glucocorticoid receptor (GR) phosphorylation is related to postoperative cognitive dysfunction (POCD) in aged individuals. Inhibition of GR activation could be an effective treatment for POCD. METHODS: A laparotomy was given to C57/BL6 mice in POCD group both 20 and 6 months old. Animals in control group were treated in identical manners except for laparotomy. Cognitive function was evaluated by Morris water maze and elevated plus maze. Western blot and Elisa assay were used to detect related molecules. Mifepristone and roscovitine were treated as inhibitions of GR phosphorylation. RESULTS: The cognitive function was impaired, and brain-derived neurotrophic factor (BDNF) was found reduced in aged POCD group. GR translocation into nucleus and elevated GR phosphorylation were found in prefrontal cortex of aged POCD mice. Cyclin-dependent Kinase 5 (CDK5), kinase for GR phosphorylation also elevated in aged POCD mice. With GR antagonist and CDK5 inhibitor, reduction of BDNF and cognitive dysfunction in aged mice were both rescued. CONCLUSION: These results presented a mechanism that surgical stress-induced GR phosphorylation contributes to POCD in aged individuals. Inhibition of GR activation and phosphorylation might be a potential treatment target of POCD.