Large-scale comparison of machine learning methods for profiling prediction of kinase inhibitors.

Conventional machine learning (ML) and deep learning (DL) play a key role in the selectivity prediction of kinase inhibitors. A number of models based on available datasets can be used to predict the kinase profile of compounds, but there is still controversy about the advantages and disadvantages of ML and DL for such tasks. In this study, we constructed a comprehensive benchmark dataset of kinase inhibitors, involving in 141,086 unique compounds and 216,823 well-defined bioassay data points for 354 kinases. We then systematically compared the performance of 12 ML and DL methods on the kinase profiling prediction task. Extensive experimental results reveal that (1) Descriptor-based ML models generally slightly outperform fingerprint-based ML models in terms of predictive performance. RF as an ensemble learning approach displays the overall best predictive performance. (2) Single-task graph-based DL models are generally inferior to conventional descriptor- and fingerprint-based ML models, however, the corresponding multi-task models generally improves the average accuracy of kinase profile prediction. For example, the multi-task FP-GNN model outperforms the conventional descriptor- and fingerprint-based ML models with an average AUC of 0.807. (3) Fusion models based on voting and stacking methods can further improve the performance of the kinase profiling prediction task, specifically, RF::AtomPairs + FP2 + RDKitDes fusion model performs best with the highest average AUC value of 0.825 on the test sets. These findings provide useful information for guiding choices of the ML and DL methods for the kinase profiling prediction tasks. Finally, an online platform called KIPP ( https://kipp.idruglab.cn ) and python software are developed based on the best models to support the kinase profiling prediction, as well as various kinase inhibitor identification tasks including virtual screening, compound repositioning and target fishing.

EMD-based gray combined forecasting model - Application to long-term forecasting of wind power generation.

Wind power is the most promising renewable energy source after hydropower because of its mature technology and low price, and has great potential for carbon emission reduction. Long-term forecasts of its power generation can help power companies to develop operational plans, grid configuration and power dispatch, and can also provide a basis for the government to formulate energy and environmental policies. However, due to the characteristics of China's monsoon climate and wind power industry development, wind power generation data are characterized by nonlinear cycles and small data volume, which makes accurate prediction more difficult. To this end, this paper develops a new prediction model and applies it to the long-term prediction of wind power generation in China, and proposes some targeted policy recommendations based on the prediction results to promote the development of China's wind power industry.

Synergy-Compensation Effect of Ferroelectric Polarization and Cationic Vacancy Collaboratively Promoting CO2 Photoreduction.

Photocatalytic CO2 reduction is severely limited by the rapid recombination of photo-generated charges and insufficient reactive sites. Creating electric field and defects are effective strategies to inhibit charge recombination and enrich catalytic sites, respectively. Herein, a coupled strategy of ferroelectric poling and cationic vacancy is developed to achieve high-performance CO2 photoreduction on ferroelectric Bi2 MoO6 , and their interesting synergy-compensation relationship is first disclosed. Corona poling increases the remnant polarization of Bi2 MoO6 to enhance the intrinsic electric field for promoting charge separation, while it decreases the CO2 adsorption. The introduced Mo vacancy (VMo ) facilitates the adsorption and activation of CO2 , and surface charge separation by creating local electric field. Unfortunately, VMo largely reduces the remnant polarization intensity. Coupling poling and VMo not only integrate their advantages, resulting in an approximately sevenfold increased surface charge transfer efficiency, but also compensate for their shortcomings, for example, VMo largely alleviates the negative effects of ferroelectric poling on CO2 adsorption. In the absence of co-catalyst or sacrificial agent, the poled Bi2 MoO6 with VMo exhibits a superior CO2 -to-CO evolution rate of 19.75 µmol g-1 h-1 , ≈8.4 times higher than the Bi2 MoO6 nanosheets. This work provides new ideas for exploring the role of polarization and defects in photocatalysis.

Targeting the PI3K/AKT/mTOR Signaling Pathway in the Treatment of Human Diseases: Current Status, Trends, and Solutions.

The phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway is one of the most important intracellular pathways involved in cell proliferation, growth, differentiation, and survival. Therefore, this route is a prospective biological target for treating various human diseases, such as tumors, neurodegenerative diseases, pulmonary fibrosis, and diabetes. An increasing number of clinical studies emphasize the necessity of developing novel molecules targeting the PI3K/AKT/mTOR pathway. This review focuses on recent advances in ATP-competitive inhibitors, allosteric inhibitors, covalent inhibitors, and proteolysis-targeting chimeras against the PI3K/AKT/mTOR pathway, and highlights possible solutions for overcoming the toxicities and acquired drug resistance of currently available drugs. We also provide recommendations for the future design and development of promising drugs targeting this pathway.

Predicting Pyrolysis of a Wide Variety of Petroleum Coke Using an Independent Parallel Reaction Model and a Backpropagation Neural Network.

In this work, the pyrolysis behavior and gaseous products of petroleum coke were investigated by nonisothermal thermogravimetric analysis (TGA) and thermogravimetry-mass spectrometry (TG-MS). Then, the pyrolysis kinetics of six kinds of petroleum coke (Fushun (FS), Fuyu (FY), Wuhan (WH), Zhenhai (ZH), Qilu (QL), and Shijiazhuang (SJZ)) were determined by an independent parallel reaction (IPR) model, and the kinetic parameters (activation energy and preexponential factor) were obtained. In addition, an efficient backpropagation neural network (BPNN) was developed to predict the thermal data of six kinds of petroleum coke. The BPNN-predicted thermal data were used to calculate the kinetic parameters based on the IPR model, and the results were compared with the ones calculated using experimental data. The results showed that the pyrolysis process of six kinds of petroleum coke was divided into three stages, of which stage II (250-900 °C) had the significant mass loss, corresponding to the devolatilization of petroleum coke. MS fragmented ion intensity analysis indicated that the main pyrolysis products were methane CH x (m/z = 13, 14, 15, and 16), aliphatic hydrocarbon C3H5, H2, CO, CO2, and H2O. The thermal data predicted by the IPR, BPNN, and BPNN-IPR (BPNN combined with IPR) models were in good agreement with the experimental data. Most importantly, it was concluded that the BPNN-predicted data can be further applied to calculate the kinetic parameters using the IPR kinetic model.

Ligand- and structure-based identification of novel CDK9 inhibitors for the potential treatment of leukemia.

Cyclin-dependent kinase 9 (CDK9) plays a vital role in controlling cell transcription and has been an attractive target for cancer treatment. Herein, ten predictive models derived from 1330 unique molecules against CDK9 were constructed based on molecular fingerprints and graphs using two conventional machine learning and four deep learning methods. The evaluation results showed that FP-GNN deep learning architecture performed best for CDK9 inhibitors prediction with the highest BA and F1 values of 0.681 and 0.912 for testing set. We then performed virtual screening to identify new CDK9 inhibitors by incorporating the optimal established predictive model and molecular docking. Five compounds were identified to show broad anticancer activity against various cancer cell lines through bioassays. For example, C9 exhibited antiproliferative activities against HeLa, MOLM-13 and MDA-MB-231 with IC50 values of 2.53, 3.92 and 11.65 µM. Kinase inhibition assay results demonstrated that these compounds displayed submicromolar (214 âˆ¼ 504 nM) inhibitory activities against CDK9. Further cellular mechanism evaluation revealed that C9 suppressed the activity of CDK9 and interfered with the expression of Mcl-1 and cleaved PARP in MOLM-13 cells, resulting in the induction of cellular apoptosis. In addition, C9 displayed a good stability in rat liver microsomes, artificial gastrointestinal fluid and plasm. An online platform (called DEEPCDK9Pred) was developed based on the FP-GNN models to predict or design new CDK9 inhibitors. Collectively, our findings demonstrated that FP-GNN algorithm can achieve accurate prediction of CDK9 inhibitors and the subsequent discovery of C9 as a new potential CDK9 inhibitor deserves further structural modification for the treatment of leukemia.

Thermal Debinding Kinetics of Gelcast Ceramic Parts via a Modified Independent Parallel Reaction Model in Comparison with the Multiple Normally Distributed Activation Energy Model.

This work aims to provide useful insights into the thermal debinding kinetics of gelcast ceramic parts, especially for debinding kinetics prediction involving heat preservation. Debinding experiments were conducted in a differential thermogravimetric analyzer at five heating rates (5, 8, 10, 15, and 20 °C/min) in the temperature range of 35-900 °C under an air atmosphere. The conversion (α) and pyrolysis rate (dα/dT) data were simulated using a modified independent parallel reaction (IPR) model and a multiple normally distributed activation energy model (M-DAEM). Their validity was assessed and compared by checking the agreement between the experimental results and the prediction capability. The results showed that both the modified IPR model and M-DAEM had high predictability for thermal debinding kinetics under linear heating conditions. The fitting quality parameters (Fit) were less than 1.406 and 1.01%, respectively. The activation energies (E i , i = 1, 2, 3, 4, and 5) calculated by the M-DAEM ranged from 153.312 to 217.171 kJ/mol. The relationships between E i of pseudo components 1 to 5 calculated by the modified IPR model were a function of the conversion rate. The E i values were E 1(α) = 116.750 + 11.153α - 26.772α2 + 4.362α3 kJ/mol, E 2(α) = 139.595 - 66.162α + 75.702α2 - 38.041α3 kJ/mol, E 3(α) = 190.854 + 135.755α - 214.801α2 + 116.093α3 kJ/mol, E 4(α) = 64.068 + 280.086α - 380.270α2 + 264.724α3 kJ/mol, and E 5(α) = 188.257 - 77.086α + 74.129α2 - 48.669α3 kJ/mol, respectively. However, it is noteworthy that the α and dα/dT curves predicted by the modified IPR model with a deviation of less than 8% were better than those predicted by the M-DAEM for the linear thermal debinding process with the holding stage. Accordingly, it is believed that the proposed modified IPR model is suitable for describing the thermal debinding kinetics involving the heat preservation of gelcast green parts.

Synergistic Polarization Engineering on Bulk and Surface for Boosting CO2 Photoreduction.

Sluggish charge kinetics and low CO2 affinity seriously inhibit CO2 photoreduction. Herein, the synchronous promotion of charge separation and CO2 affinity of Bi4 Ti3 O12 is realized by coupling corona poling and surface I-grafting. Corona poling enhances ferroelectric polarization of Bi4 Ti3 O12 by aligning the domains direction, which profoundly promotes charge transfer along opposite directions across bulk. Surface I-grafting forms a surface local electric field for further separating charge carriers and provides abundant active sites to enhance CO2 adsorption. The two modifications cooperatively further increase the ferroelectric polarization of Bi4 Ti3 O12 , which maximize the separation efficiency of photogenerated charges, resulting in an enhanced CO production rate of 15.1 µmol g-1 h-1 (nearly 9 times) with no sacrificial agents or cocatalysts. This work discloses that ferroelectric polarization and surface ion grafting can promote CO2 photoreduction in a synergistic way.

New lathyrane diterpenoids with anti-inflammatory activity isolated from the roots of Jatropha curcas L.

ETHNOPHARMACOLOGICAL RELEVANCE: Jatropha curcas L. (Euphorbiaceae), as a drought resistant shrub mainly cultivated in tropical and subtropical areas worldwide, is widely used as traditional medicine to cure arthritis, dysentery, abscess and pneumonia in Asian, African and South American folklores. The methanolic extracts of the roots have been revealed the anti-inflammatory activity in vivo and vitro. AIM OF STUDY: This research aimed to provide promising anti-inflammatory candidates from the roots of J. curcas. In addition, RNA-Seq was conducted to give targeted genes involved in the anti-inflammatory action. MATERIALS AND METHODS: The diterpenoids were isolated from the CH2Cl2 fraction of the methanolic extract from the roots of J. curcas by column chromatography (CC): silica gel, Sephadex LH-20, ODS, semi-preparative reversed-phase high-performance liquid chromatography (HPLC). The structures were identified based on HR-ESI-MS and 1D, 2D-NMR spectroscopic analysis. Their anti-inflammatory effects were tested on lipopolysaccharide (LPS, 500 ng/mL)-stimulated murine RAW264.7 macrophages. Furthermore, we conducted transcriptome-wide RNA sequencing to profile gene expression alterations in LPS-induced RAW264.7 cells upon treatment with jatrocurcasenone I (4) and analyzed the underlying genes targeted by this compound. RESULTS: Six diterpenoids were obtained from J. curcas, and four of them were identified to be new lathyrane diterpenoids: jatrocurcasenones F-I (1-4). Compounds 3 and 4 exhibited potent inhibitory activities against LPS-induced nitric oxide (NO) production in RAW264.7 cells with IC50 values of 11.28 µM and 7.71 µM, respectively. Western blotting analysis showed that the expression of inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) were suppressed with the supplementation of 3 and 4. The results of RNA-seq showed that 4 (20 µM) exhibited regulation on the 587 differentially expressed genes (DEGs) induced by LPS (500 ng/mL). Transcriptome-wide RNA sequencing indicated that the protective activity of 4 supplementation was most likely driven by modulating expression levels of IL-1α, IL-1ß, IL-1f6, IL-6, IL-1rn, IL-27, Ccl2, Ccl5, Ccl7, Ccl9, Ccl22, Cxcl10, Tnfsf12, Tnfsf15, Lta, Trim25, Bcl2a1a, Dusp1, Dusp2, Ptgs2, Edn1 and Nr4a1. CONCLUSIONS: This study offered four new lathyrane diterpenoids, of them, jatrocurcasenone I (4) showed significant anti-inflammatory activity. RNA-Seq suggested that jatrocurcasenone I (4) could be a candidate drug for the prevention inflammation-mediated diseases by modulating 24 candidate DEGs.

B-O Bonds in Ultrathin Boron Nitride Nanosheets to Promote Photocatalytic Carbon Dioxide Conversion.

Limited by the chemical inertness of CO2 and the high dissociation energy of the C═O bond, photocatalytic CO2 conversion is highly challenging. Herein, we prepare ultrathin oxygen-modified h-BN (O/BN) nanosheets containing B-O bonds. On the O/BN surface, CO2 can be chemically captured and is bonded with the B-O bond, leading to the formation of an O-B-O bond. This new chemical bond acting as an electron-delivery channel strengthens the interaction between CO2 and the surface. Thus, the reactants can continuously obtain electrons from the surface through this channel. Therefore, the majority of gaseous CO2 directly converts into carbon active species that are detected by in situ DRIFTS over O/BN. Moreover, the activated energies of CO2 conversion are significantly reduced with the introduction of the B-O bond evidenced by DFT calculations. As a result, O/BN nanosheets present an enhanced photocatalytic CO2 conversion performance with the H2 and CO generation rates of 3.3 and 12.5 µmol g-1 h-1, respectively. This work could help in realizing the effects of nonmetal chemical bonds in the CO2 photoreduction reaction for designing efficient photocatalysts.

Multistage kinetic analysis of DMAA/MBAM polymer removal from gelcast ceramic parts using a multi-stage parallel reaction model and model-free method.

This work aims to develop an effective method for investigating the multistage debinding kinetics and reaction mechanisms of removing N,N-dimethylacrylamide/N,N'-methylenebisacrylamide (DMAA/MBAM) polymer from gelcast ceramic parts. Thermogravimetry (TG) and pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) experiments were performed to investigate the thermal degradation characteristics and the main compounds produced during the pyrolysis of DMAA/MBAM polymer within green components. A multi-stage parallel reaction model (M-PRM) was proposed to separate the overlapping peaks in the dα/dT curves. The kinetic parameters (activation energy E and pre-exponential factor k 0) of each substage were calculated using model-free methods (Flynn-Wall-Ozawa, Starink, Friedman and Kissinger-Akahira-Sunose) and an activation energy variable model. In addition, the most appropriate kinetic mechanism function f(α) of each substage was analyzed and discussed via Málek's procedure and the Sesták-Berggren (SB) model. The results showed that the DMAA/MBAM polymer burnout in green components can be divided into three substages through a three-stage parallel reaction model (3-PRM). The values of E (Friedman method) for substages 1 to 3 were E(α) = 139.862 - 110.481α + 156.161α 2 - 88.714α 3 kJ mol-1, E(α) = 160.791 + 152.496α - 236.906α 2 + 163.724α 3 kJ mol-1 and E(α) = 72.132 + 452.830α - 669.039α 2 + 507.015α 3 kJ mol-1, respectively. The average values of E showed an increasing tendency from substages 1 to 3, and a kinetic compensation effect was also observed between the E and k 0 in each substage. The kinetic mechanism analysis revealed that the reaction mechanisms for substages 1 to 3 were f(α) = (1 - α)0.668 α 3.049(-ln(1 - α))-3.874, f(α) = (1 - α)0.700 α 3.177(-ln(1 - α))-3.962 and f(α) = (1 - α)1.049 α -0.161(-ln(1 - α))0.518, respectively. It is expected that the research results can be extended to investigate the multiplex debinding of binders or polymers for various colloidal molding techniques.

Dimericursones A and B: two unprecedented hexacyclic dimeric diterpenoids from the root barks of Jatropha curcas.

Dimericursones A and B (1 and 2), two unprecedented hexacyclic dimeric diterpenoids, were obtained from the root barks of Jatropha curcas. Their structures were elucidated by extensive spectroscopic analysis, electronic circular dichroism calculations, and single-crystal X-ray diffraction. Dimericursone B (2) showed significant inhibition on nitric oxide production of lipopolysaccharide-induced RAW264.7 macrophages with IC50 values of 5.65 µM.