Acacia hybrids offer a great potential for paper industry in Southeast Asia due to their fast growth and ability to grow on abandoned or marginal lands. Breeding Acacia hybrids with desirable traits can be achieved th...Acacia hybrids offer a great potential for paper industry in Southeast Asia due to their fast growth and ability to grow on abandoned or marginal lands. Breeding Acacia hybrids with desirable traits can be achieved through marker assisted selection(MAS) breeding. To develop a MAS program requires development of linkage maps and QTL analysis. Two mapping populations were developed through interspecific hybridization for linkage mapping and QTL analysis. All seeds per pod were cultured initially to improve hybrid yield as quality and density of linkage mapping is affected by the size of the mapping population. Progenies from two mapping populations were field planted for phenotypic and genotypic evaluation at three locations in Malaysia,(1) Forest Research Institute Malaysia field station at Segamat, Johor,(2) Borneo Tree Seeds and Seedlings Supplies Sdn, Bhd.(BTS) field trial site at Bintulu, Sarawak, and(3) Asiaprima RCF field trial site at Lancang, Pahang. During field planting, mislabeling was reported at Segamat, Johor, and a similar problem was suspected for Bintulu, Sarawak. Early screening with two isozymes effectively selected hybrid progenies, and these hybrids were subsequently further confirmed by using species-specific SNPs. During field planting, clonal mislabeling was reported and later confirmed by using a small set of STMS markers. A large set of SNPs were also used to screen all ramets in both populations. A total of 65.36% mislabeled ramets were encountered in the wood density population and 60.34% in the fibre length mapping population. No interpopulation pollen contamination was detected because all ramets found their match within the same population in question.However, mislabeling was detected among ramets of the same population. Mislabeled individuals were identified and grouped as they originated from 93 pods for wood density and 53 pods for fibre length mapping populations.On average 2 meiotically unique seeds per pod(179 seeds/93 pods) for wood density and 3 meiotically unique seeds per p展开更多
There is an increasing demand for salmonid authentication due to the globalization of the salmonid trade.DNA barcoding and mini-DNA barcoding are widely used for identifying fish species based on a fragment of the mit...There is an increasing demand for salmonid authentication due to the globalization of the salmonid trade.DNA barcoding and mini-DNA barcoding are widely used for identifying fish species based on a fragment of the mitochondrial cytochrome c oxidase subunit I(COI)sequence.In this study,rainbow trout(Oncorhynchus mykiss),steelhead trout(O.mykiss),and Atlantic salmon(Salmo salar)collected from two salmonid aquaculture bases in China were authenticated by DNA barcoding(about 650 bp)and mini-DNA barcoding(127 bp)to evaluate the accuracy of the two methods in the identification of different salmonid species.The results revealed that both methods could effectively distinguish O.mykiss and S.salar with 100%accuracy.However,the two methods failed to separate rainbow trout(O.mykiss)and steelhead trout(O.mykiss),which are the same species but cultured in different water environments.Moreover,salmonid samples from three main distribution channels in the Qingdao area(traditional supermarkets,online supermarkets,and sushi bars)were identified by the two methods.Substitution of S.salar with O.mykiss was discovered,and the 27.78%overall substitution rate of salmonids in the Qingdao area was higher than those in other regions reported in previous studies.In addition,the mislabeling rates of salmonids from traditional supermarkets,online supermarkets,and sushi bars were compared in this study.The mislabeling rate was significantly greater in sushi bars(50%)than in the other two channels(16.67%),suggesting that stronger monitoring and enforcement measures are necessary for the aquatic food catering industry.展开更多
The dried shellfish products with rich nutrients and low-calorie content are favorite food in China,especially in coastal areas.However,the species of dried shellfish products in the market are usually unknown,as the ...The dried shellfish products with rich nutrients and low-calorie content are favorite food in China,especially in coastal areas.However,the species of dried shellfish products in the market are usually unknown,as the taxonomic features were removed during the production process.This study described the application of DNA barcoding technique to the identification of 100 dried shellfish(scallop,squid,octopus and cuttlefish)products in markets.Samples were authenticated by comparing mitochondrial cytochrome oxidase subunit I(COI)gene and 16S ribosomal RNA(16S rRNA)gene sequences with public reference taxonomic databases.The results showed that all the 100 products can be identified at species level.Sixty four scallop adductor products were processed using the bay scallop,Argopecten irradians,and one was from Portuguese oyster,Crassostrea angulata.All the 27 squid,2 cuttlefish and 6 octopus products were produced by the Jumbo flying squid,Dosidicus gigas.The neighbour-joining tree is in agreement with the results of DNA barcoding analysis.The 64 scallop samples formed one A.irradians cluster,leaving Sca65 clustered with the reference oyster sequence C.angulata(MH997922).All the 35 cephalopod(squid,octopus and cuttlefish)samples formed a D.gigas cluster.This investigation revealed a low variety of dried shellfish products sold on the market,and highlighted the high rate of mislabeling and species substitution.Our present work provides a practical method for tracing and authenticating shellfish products.展开更多
The development of image classification is one of the most important research topics in remote sensing. The prediction accuracy depends not only on the appropriate choice of the machine learning method but also on the...The development of image classification is one of the most important research topics in remote sensing. The prediction accuracy depends not only on the appropriate choice of the machine learning method but also on the quality of the training datasets. However, real-world data is not perfect and often suffers from noise. This paper gives an overview of noise filtering methods. Firstly, the types of noise and the consequences of class noise on machine learning are presented. Secondly, class noise handling methods at both the data level and the algorithm level are introduced. Then ensemble-based class noise handling methods including class noise removal, correction, and noise robust ensemble learners are presented. Finally, a summary of existing data-cleaning techniques is given.展开更多
Large-scale datasets are driving the rapid developments of deep convolutional neural networks for visual sentiment analysis.However,the annotation of large-scale datasets is expensive and time consuming.Instead,it ise...Large-scale datasets are driving the rapid developments of deep convolutional neural networks for visual sentiment analysis.However,the annotation of large-scale datasets is expensive and time consuming.Instead,it iseasy to obtain weakly labeled web images from the Internet.However,noisy labels st.ill lead to seriously degraded performance when we use images directly from the web for training networks.To address this drawback,we propose an end-to-end weakly supervised learning network,which is robust to mislabeled web images.Specifically,the proposed attention module automatically eliminates the distraction of those samples with incorrect labels bv reducing their attention scores in the training process.On the other hand,the special-class activation map module is designed to stimulate the network by focusing on the significant regions from the samples with correct labels in a weakly supervised learning approach.Besides the process of feature learning,applying regularization to the classifier is considered to minimize the distance of those samples within the same class and maximize the distance between different class centroids.Quantitative and qualitative evaluations on well-and mislabeled web image datasets demonstrate that the proposed algorithm outperforms the related methods.展开更多
基金provided by the Ministry of Science,Technology and Innovation Malaysia(IRPA 01-02-02-0015PR0003/03-02,02-01-02-SF0403)Universiti Kebangsaan Malaysia(UKM-AP-BPB-13-2009,GUP-2013-039)
文摘Acacia hybrids offer a great potential for paper industry in Southeast Asia due to their fast growth and ability to grow on abandoned or marginal lands. Breeding Acacia hybrids with desirable traits can be achieved through marker assisted selection(MAS) breeding. To develop a MAS program requires development of linkage maps and QTL analysis. Two mapping populations were developed through interspecific hybridization for linkage mapping and QTL analysis. All seeds per pod were cultured initially to improve hybrid yield as quality and density of linkage mapping is affected by the size of the mapping population. Progenies from two mapping populations were field planted for phenotypic and genotypic evaluation at three locations in Malaysia,(1) Forest Research Institute Malaysia field station at Segamat, Johor,(2) Borneo Tree Seeds and Seedlings Supplies Sdn, Bhd.(BTS) field trial site at Bintulu, Sarawak, and(3) Asiaprima RCF field trial site at Lancang, Pahang. During field planting, mislabeling was reported at Segamat, Johor, and a similar problem was suspected for Bintulu, Sarawak. Early screening with two isozymes effectively selected hybrid progenies, and these hybrids were subsequently further confirmed by using species-specific SNPs. During field planting, clonal mislabeling was reported and later confirmed by using a small set of STMS markers. A large set of SNPs were also used to screen all ramets in both populations. A total of 65.36% mislabeled ramets were encountered in the wood density population and 60.34% in the fibre length mapping population. No interpopulation pollen contamination was detected because all ramets found their match within the same population in question.However, mislabeling was detected among ramets of the same population. Mislabeled individuals were identified and grouped as they originated from 93 pods for wood density and 53 pods for fibre length mapping populations.On average 2 meiotically unique seeds per pod(179 seeds/93 pods) for wood density and 3 meiotically unique seeds per p
基金the National Key Research and Development Program of China(No.2019YFD0901000)the Natural Science Foundation of Shandong Pro-vince,China(No.ZR2020MC194).
文摘There is an increasing demand for salmonid authentication due to the globalization of the salmonid trade.DNA barcoding and mini-DNA barcoding are widely used for identifying fish species based on a fragment of the mitochondrial cytochrome c oxidase subunit I(COI)sequence.In this study,rainbow trout(Oncorhynchus mykiss),steelhead trout(O.mykiss),and Atlantic salmon(Salmo salar)collected from two salmonid aquaculture bases in China were authenticated by DNA barcoding(about 650 bp)and mini-DNA barcoding(127 bp)to evaluate the accuracy of the two methods in the identification of different salmonid species.The results revealed that both methods could effectively distinguish O.mykiss and S.salar with 100%accuracy.However,the two methods failed to separate rainbow trout(O.mykiss)and steelhead trout(O.mykiss),which are the same species but cultured in different water environments.Moreover,salmonid samples from three main distribution channels in the Qingdao area(traditional supermarkets,online supermarkets,and sushi bars)were identified by the two methods.Substitution of S.salar with O.mykiss was discovered,and the 27.78%overall substitution rate of salmonids in the Qingdao area was higher than those in other regions reported in previous studies.In addition,the mislabeling rates of salmonids from traditional supermarkets,online supermarkets,and sushi bars were compared in this study.The mislabeling rate was significantly greater in sushi bars(50%)than in the other two channels(16.67%),suggesting that stronger monitoring and enforcement measures are necessary for the aquatic food catering industry.
基金This work was supported by research grants from the Fundamental Research Funds for the Central Universities(No.201762014)the National Natural Science Foundation of China(No.31772414)the National Natural Science Foundation of Qingdao City(No.20-3-4-16-nsh).
文摘The dried shellfish products with rich nutrients and low-calorie content are favorite food in China,especially in coastal areas.However,the species of dried shellfish products in the market are usually unknown,as the taxonomic features were removed during the production process.This study described the application of DNA barcoding technique to the identification of 100 dried shellfish(scallop,squid,octopus and cuttlefish)products in markets.Samples were authenticated by comparing mitochondrial cytochrome oxidase subunit I(COI)gene and 16S ribosomal RNA(16S rRNA)gene sequences with public reference taxonomic databases.The results showed that all the 100 products can be identified at species level.Sixty four scallop adductor products were processed using the bay scallop,Argopecten irradians,and one was from Portuguese oyster,Crassostrea angulata.All the 27 squid,2 cuttlefish and 6 octopus products were produced by the Jumbo flying squid,Dosidicus gigas.The neighbour-joining tree is in agreement with the results of DNA barcoding analysis.The 64 scallop samples formed one A.irradians cluster,leaving Sca65 clustered with the reference oyster sequence C.angulata(MH997922).All the 35 cephalopod(squid,octopus and cuttlefish)samples formed a D.gigas cluster.This investigation revealed a low variety of dried shellfish products sold on the market,and highlighted the high rate of mislabeling and species substitution.Our present work provides a practical method for tracing and authenticating shellfish products.
基金supported by the National Natural Science Foundation of China (62201438,61772397,12005169)the Basic Research Program of Natural Sciences of Shaanxi Province (2021JC-23)+2 种基金Yulin Science and Technology Bureau Science and Technology Development Special Project (CXY-2020-094)Shaanxi Forestry Science and Technology Innovation Key Project (SXLK2022-02-8)the Project of Shaanxi F ederation of Social Sciences (2022HZ1759)。
文摘The development of image classification is one of the most important research topics in remote sensing. The prediction accuracy depends not only on the appropriate choice of the machine learning method but also on the quality of the training datasets. However, real-world data is not perfect and often suffers from noise. This paper gives an overview of noise filtering methods. Firstly, the types of noise and the consequences of class noise on machine learning are presented. Secondly, class noise handling methods at both the data level and the algorithm level are introduced. Then ensemble-based class noise handling methods including class noise removal, correction, and noise robust ensemble learners are presented. Finally, a summary of existing data-cleaning techniques is given.
基金Project supported by the Key Project of the National Natural Science Foundation of China(No.U1836220)the National Nat-ural Science Foundation of China(No.61672267)+1 种基金the Qing Lan Talent Program of Jiangsu Province,China,the Jiangsu Key Laboratory of Security Technology for Industrial Cyberspace,China,the Finnish Cultural Foundation,the Jiangsu Specially-Appointed Professor Program,China(No.3051107219003)the liangsu Joint Research Project of Sino-Foreign Cooperative Education Platform,China,and the Talent Startup Project of Nanjing Institute of Technology,China(No.YKJ201982)。
文摘Large-scale datasets are driving the rapid developments of deep convolutional neural networks for visual sentiment analysis.However,the annotation of large-scale datasets is expensive and time consuming.Instead,it iseasy to obtain weakly labeled web images from the Internet.However,noisy labels st.ill lead to seriously degraded performance when we use images directly from the web for training networks.To address this drawback,we propose an end-to-end weakly supervised learning network,which is robust to mislabeled web images.Specifically,the proposed attention module automatically eliminates the distraction of those samples with incorrect labels bv reducing their attention scores in the training process.On the other hand,the special-class activation map module is designed to stimulate the network by focusing on the significant regions from the samples with correct labels in a weakly supervised learning approach.Besides the process of feature learning,applying regularization to the classifier is considered to minimize the distance of those samples within the same class and maximize the distance between different class centroids.Quantitative and qualitative evaluations on well-and mislabeled web image datasets demonstrate that the proposed algorithm outperforms the related methods.
基金University of Chinese Academy of Sciences(Y95401TXX2)Beijing Natural Science Foundation(Z190004)Key Program of Joint Funds of the National Natural Science Foundation of China(U19B2040)。
