Few-shot learning is becoming more and more popular in many fields,especially in the computer vision field.This inspires us to introduce few-shot learning to the genomic field,which faces a typical few-shot problem be...Few-shot learning is becoming more and more popular in many fields,especially in the computer vision field.This inspires us to introduce few-shot learning to the genomic field,which faces a typical few-shot problem because some tasks only have a limited number of samples with high-dimensions.The goal of this study was to investigate the few-shot disease sub-type prediction problem and identify patient subgroups through training on small data.Accurate disease subtype classification allows clinicians to efficiently deliver investigations and interventions in clinical practice.We propose the SW-Net,which simulates the clinical process of extracting the shared knowledge from a range of interrelated tasks and generalizes it to unseen data.Our model is built upon a simple baseline,and we modified it for genomic data.Supportbased initialization for the classifier and transductive fine-tuning techniques were applied in our model to improve prediction accuracy,and an Entropy regularization term on the query set was appended to reduce over-fitting.Moreover,to address the high dimension and high noise issue,we future extended a feature selection module to adaptively select important features and a sample weighting module to prioritize high-confidence samples.Experiments on simulated data and The Cancer Genome Atlas meta-dataset show that our new baseline model gets higher prediction accuracy compared to other competing algorithms.展开更多
This research demonstrates the production of mesoporous activated carbon from sargassum fusiforme via physical activation with carbon dioxide.Central composite design was applied to conduct the experiments at differen...This research demonstrates the production of mesoporous activated carbon from sargassum fusiforme via physical activation with carbon dioxide.Central composite design was applied to conduct the experiments at different levels by altering three operating parameters.Activation temperature(766-934℃),CO2 flow rate(0.8-2.8 L·min^-1)and activation time(5-55 min)were the variables examined in this study.The effect of parameters on the specific surface area,total pore volume and burn-out rate of activated carbon was studied,and the influential parameters of methylene blue adsorption value were identified employing analysis of variance.The optimum conditions for maximum methylene blue adsorption value were:activation temperature=900℃,activation time=29.05 min and CO2 flow rate=1.8 L·min(-1).The activated carbon produced under optimum conditions was characterized by BET,FTIR and SEM.The adsorption behavior on congo red was studied.The effect of parameters on the adsorbent dosage,temperature,PH and initial congo red concentration was investigated.The adsorption properties of the activated carbon were investigated by kinetics.The equilibrium removal rate and maximum adsorption capacity reaches up to 94.72%,234 mg·g^-1,respectively when initial congo red concentration is 200 mg·L^-1 under adsorbent dosage(0.8 g·L^-1),temperature(30℃),PH7.展开更多
Radiogenic isotopic dating and Lu–Hf isotopic composition using laser ablation-inductively coupled plasma-mass spectrometry(LA-ICP-MS)of the Wude basalt in Yunnan province from the Emeishan large igneous province(ELI...Radiogenic isotopic dating and Lu–Hf isotopic composition using laser ablation-inductively coupled plasma-mass spectrometry(LA-ICP-MS)of the Wude basalt in Yunnan province from the Emeishan large igneous province(ELIP)yielded timing of formation and post-eruption tectonothermal event.Holistic lithogeochemistry and elements mapping of basaltic rocks were further reevaluated to provide insights into crustal contamination and formation of the ELIP.A zircon U–Pb age of 251.3±2.0 Ma of the Wude basalt recorded the youngest volcanic eruption event and was consistent with the age span of 251-263 Ma for the emplacement of the ELIP.Such zircons hadε_(Hf)(t)values ranging from7.3 to+2.2,identical to those of magmatic zircons from the intrusive rocks of the ELIP,suggesting that crust-mantle interaction occurred during magmatic emplacement,or crust-mantle mixing existed in the deep source region prior to deep melting.The apatite U–Pb age at 53.6±3.4 Ma recorded an early Eocene magmatic superimposition of a regional tectonothermal event,corresponding to the Indian–Eurasian plate collision.Negative Nb,Ta,Ti and P anomalies of the Emeishan basalt may reflect crustal contamination.The uneven Nb/La and Th/Ta values distribution throughout the ELIP supported a mantle plume model origin.Therefore,the ELIP was formed as a result of a mantle plume which was later superimposed by a regional tectonothermal event attributed to the Indian–Eurasian plate collision during early Eocene.展开更多
Using the ionic liquid[emim][Tf2N]as a physical solvent,it was found by Aspen Plus simulation that it was possible to attempt to capture CO2 from the flue gas discharged from the coal-fired unit of the power plant.Usi...Using the ionic liquid[emim][Tf2N]as a physical solvent,it was found by Aspen Plus simulation that it was possible to attempt to capture CO2 from the flue gas discharged from the coal-fired unit of the power plant.Using the combination of model calculation and experimental determination,the density,isostatic heat capacity,viscosity,vapor pressure,thermal conductivity,surface tension and solubility of[emim][Tf2N]were obtained.Based on the NRTL model,the Henry coefficient and NRTL binary interaction parameters of CO2 dissolved in[emim][Tf2N]were obtained by correlating[emim][Tf2N]with the gas–liquid equilibrium data of CO2.Firstly,the calculated relevant data is imported into Aspen Plus,and the whole process model of the ionic liquid absorption process is established.Then the absorption process is optimized according to the temperature distribution in the absorption tower to obtain a new absorption process.Finally,the density,constant pressure heat capacity,surface tension,thermal conductivity,and viscosity of[emim][Tf2N]were changed to investigate the effect of ionic liquid properties on process energy consumption,solvent circulation and heat exchanger design.The results showed that based on the composition of the inlet gas stream to the absorbers,CO2 with a capture rate of 90%and a mass purity higher than 99.5%was captured.These results indicate that the[emim][Tf2N]could be used as a physical solvent for CO2 capture from coal-fired units.In addition,the results will provide a theoretical basis for the design of new ionic liquids for CO2 capture.展开更多
基金supported by the Macao Science and Technology Development Funds Grands No.0158/2019/A3 from the Macao Special Administrative Region of the People’s Republic of China.
文摘Few-shot learning is becoming more and more popular in many fields,especially in the computer vision field.This inspires us to introduce few-shot learning to the genomic field,which faces a typical few-shot problem because some tasks only have a limited number of samples with high-dimensions.The goal of this study was to investigate the few-shot disease sub-type prediction problem and identify patient subgroups through training on small data.Accurate disease subtype classification allows clinicians to efficiently deliver investigations and interventions in clinical practice.We propose the SW-Net,which simulates the clinical process of extracting the shared knowledge from a range of interrelated tasks and generalizes it to unseen data.Our model is built upon a simple baseline,and we modified it for genomic data.Supportbased initialization for the classifier and transductive fine-tuning techniques were applied in our model to improve prediction accuracy,and an Entropy regularization term on the query set was appended to reduce over-fitting.Moreover,to address the high dimension and high noise issue,we future extended a feature selection module to adaptively select important features and a sample weighting module to prioritize high-confidence samples.Experiments on simulated data and The Cancer Genome Atlas meta-dataset show that our new baseline model gets higher prediction accuracy compared to other competing algorithms.
基金financially supported by the Zhejiang Provincial Natural Science Foundation of China(LY16B060014)the Program for the Joint Research Fund for Overseas Chinese,Hong Kong and Macao Scholars of National Natural Science Foundation of China(Grant No.21628601)the Innovation and Development of Marine Economy Demonstration。
文摘This research demonstrates the production of mesoporous activated carbon from sargassum fusiforme via physical activation with carbon dioxide.Central composite design was applied to conduct the experiments at different levels by altering three operating parameters.Activation temperature(766-934℃),CO2 flow rate(0.8-2.8 L·min^-1)and activation time(5-55 min)were the variables examined in this study.The effect of parameters on the specific surface area,total pore volume and burn-out rate of activated carbon was studied,and the influential parameters of methylene blue adsorption value were identified employing analysis of variance.The optimum conditions for maximum methylene blue adsorption value were:activation temperature=900℃,activation time=29.05 min and CO2 flow rate=1.8 L·min(-1).The activated carbon produced under optimum conditions was characterized by BET,FTIR and SEM.The adsorption behavior on congo red was studied.The effect of parameters on the adsorbent dosage,temperature,PH and initial congo red concentration was investigated.The adsorption properties of the activated carbon were investigated by kinetics.The equilibrium removal rate and maximum adsorption capacity reaches up to 94.72%,234 mg·g^-1,respectively when initial congo red concentration is 200 mg·L^-1 under adsorbent dosage(0.8 g·L^-1),temperature(30℃),PH7.
基金the National Natural Science Foundation of China(91962106,41702069)the National Key Research Program(2019YFA0708603)+4 种基金the Fundamental Research Funds for the Central Universities,China(2652018125)the 111 Project of the Ministry of Science and Technology,China(BP0719021)the China International Science and Technology Cooperation Award and the Overseas Experts Exchange Project of the Ministry of Science and Technology,China(G20190001257)the Beijing Nova Program(No.Z201100006820097)the National Natural Science Foundation of China.
文摘Radiogenic isotopic dating and Lu–Hf isotopic composition using laser ablation-inductively coupled plasma-mass spectrometry(LA-ICP-MS)of the Wude basalt in Yunnan province from the Emeishan large igneous province(ELIP)yielded timing of formation and post-eruption tectonothermal event.Holistic lithogeochemistry and elements mapping of basaltic rocks were further reevaluated to provide insights into crustal contamination and formation of the ELIP.A zircon U–Pb age of 251.3±2.0 Ma of the Wude basalt recorded the youngest volcanic eruption event and was consistent with the age span of 251-263 Ma for the emplacement of the ELIP.Such zircons hadε_(Hf)(t)values ranging from7.3 to+2.2,identical to those of magmatic zircons from the intrusive rocks of the ELIP,suggesting that crust-mantle interaction occurred during magmatic emplacement,or crust-mantle mixing existed in the deep source region prior to deep melting.The apatite U–Pb age at 53.6±3.4 Ma recorded an early Eocene magmatic superimposition of a regional tectonothermal event,corresponding to the Indian–Eurasian plate collision.Negative Nb,Ta,Ti and P anomalies of the Emeishan basalt may reflect crustal contamination.The uneven Nb/La and Th/Ta values distribution throughout the ELIP supported a mantle plume model origin.Therefore,the ELIP was formed as a result of a mantle plume which was later superimposed by a regional tectonothermal event attributed to the Indian–Eurasian plate collision during early Eocene.
基金financially supported by the Zhejiang Provincial Natural Science Foundation of China(LY16B060014)State Key Laboratory of Chemical Engineering(No.SKL-ChE-08A01)the Innovation and Development of Marine Economy Demonstration。
文摘Using the ionic liquid[emim][Tf2N]as a physical solvent,it was found by Aspen Plus simulation that it was possible to attempt to capture CO2 from the flue gas discharged from the coal-fired unit of the power plant.Using the combination of model calculation and experimental determination,the density,isostatic heat capacity,viscosity,vapor pressure,thermal conductivity,surface tension and solubility of[emim][Tf2N]were obtained.Based on the NRTL model,the Henry coefficient and NRTL binary interaction parameters of CO2 dissolved in[emim][Tf2N]were obtained by correlating[emim][Tf2N]with the gas–liquid equilibrium data of CO2.Firstly,the calculated relevant data is imported into Aspen Plus,and the whole process model of the ionic liquid absorption process is established.Then the absorption process is optimized according to the temperature distribution in the absorption tower to obtain a new absorption process.Finally,the density,constant pressure heat capacity,surface tension,thermal conductivity,and viscosity of[emim][Tf2N]were changed to investigate the effect of ionic liquid properties on process energy consumption,solvent circulation and heat exchanger design.The results showed that based on the composition of the inlet gas stream to the absorbers,CO2 with a capture rate of 90%and a mass purity higher than 99.5%was captured.These results indicate that the[emim][Tf2N]could be used as a physical solvent for CO2 capture from coal-fired units.In addition,the results will provide a theoretical basis for the design of new ionic liquids for CO2 capture.
