Natural products(NPs)have long held a significant position in various fields such as medicine,food,agriculture,and materials.The chemical space covered by NPs is extensive but often underexplored.Therefore,high-throug...Natural products(NPs)have long held a significant position in various fields such as medicine,food,agriculture,and materials.The chemical space covered by NPs is extensive but often underexplored.Therefore,high-throughput and efficient methodologies for the annotation and discovery of NPs are desired to address the complexity and diversity of NP-based systems.Mass spectrometry(MS)has emerged as a powerful platform for the annotation and discovery of NPs.MS databases provide vital support for the structural characterization of NPs by integrating extensive mass spectral data and sample information.Additionally,the released annotation methodologies,based on a variety of informatics tools,continuously improve the ability to annotate the structure and properties of compounds.This review examines the current mainstream databases and annotation methodologies,focusing on their advantages and limitations.Prospects for future technological advancements are then discussed in terms of novel applications and research objectives.Through a systematic overview,this review aims to provide valuable insights and a reference for MS-based NPs annotation,thereby promoting the discovery of novel natural entities.展开更多
Glaucoma is an eye disease characterized by pathologically elevated intraocular pressure,optic nerve atrophy,and visual field defects,which can lead to irreversible vision loss.In recent years,the rapid development of...Glaucoma is an eye disease characterized by pathologically elevated intraocular pressure,optic nerve atrophy,and visual field defects,which can lead to irreversible vision loss.In recent years,the rapid development of artificial intelligence(AI)technology has provided new approaches for the early diagnosis and management of glaucoma.By classifying and annotating glaucoma-related images,AI models can learn and recognize the specific pathological features of glaucoma,thereby achieving automated imaging analysis and classification.Research on glaucoma imaging classification and annotation mainly involves color fundus photography(CFP),optical coherence tomography(OCT),anterior segment optical coherence tomography(AS-OCT),and ultrasound biomicroscopy(UBM)images.CFP is primarily used for the annotation of the optic cup and disc,while OCT is used for measuring and annotating the thickness of the retinal nerve fiber layer,and AS-OCT and UBM focus on the annotation of the anterior chamber angle structure and the measurement of anterior segment structural parameters.To standardize the classification and annotation of glaucoma images,enhance the quality and consistency of annotated data,and promote the clinical application of intelligent ophthalmology,this guideline has been developed.This guideline systematically elaborates on the principles,methods,processes,and quality control requirements for the classification and annotation of glaucoma images,providing standardized guidance for the classification and annotation of glaucoma images.展开更多
The success of deep transfer learning in fault diagnosis is attributed to the collection of high-quality labeled data from the source domain.However,in engineering scenarios,achieving such high-quality label annotatio...The success of deep transfer learning in fault diagnosis is attributed to the collection of high-quality labeled data from the source domain.However,in engineering scenarios,achieving such high-quality label annotation is difficult and expensive.The incorrect label annotation produces two negative effects:1)the complex decision boundary of diagnosis models lowers the generalization performance on the target domain,and2)the distribution of target domain samples becomes misaligned with the false-labeled samples.To overcome these negative effects,this article proposes a solution called the label recovery and trajectory designable network(LRTDN).LRTDN consists of three parts.First,a residual network with dual classifiers is to learn features from cross-domain samples.Second,an annotation check module is constructed to generate a label anomaly indicator that could modify the abnormal labels of false-labeled samples in the source domain.With the training of relabeled samples,the complexity of diagnosis model is reduced via semi-supervised learning.Third,the adaptation trajectories are designed for sample distributions across domains.This ensures that the target domain samples are only adapted with the pure-labeled samples.The LRTDN is verified by two case studies,in which the diagnosis knowledge of bearings is transferred across different working conditions as well as different yet related machines.The results show that LRTDN offers a high diagnosis accuracy even in the presence of incorrect annotation.展开更多
The application of whole genome sequencing is expanding in clinical diagnostics across various genetic disorders, and the significance of non-coding variants in penetrant diseases is increasingly being demonstrated. T...The application of whole genome sequencing is expanding in clinical diagnostics across various genetic disorders, and the significance of non-coding variants in penetrant diseases is increasingly being demonstrated. Therefore, it is urgent to improve the diagnostic yield by exploring the pathogenic mechanisms of variants in non-coding regions. However, the interpretation of non-coding variants remains a significant challenge, due to the complex functional regulatory mechanisms of non-coding regions and the current limitations of available databases and tools. Hence, we develop the non-coding variant annotation database (NCAD, http://www.ncawdb.net/), encompassing comprehensive insights into 665,679,194 variants, regulatory elements, and element interaction details. Integrating data from 96 sources, spanning both GRCh37 and GRCh38 versions, NCAD v1.0 provides vital information to support the genetic diagnosis of non-coding variants, including allele frequencies of 12 diverse populations, with a particular focus on the population frequency information for 230,235,698 variants in 20,964 Chinese individuals. Moreover, it offers prediction scores for variant functionality, five categories of regulatory elements, and four types of non-coding RNAs. With its rich data and comprehensive coverage, NCAD serves as a valuable platform, empowering researchers and clinicians with profound insights into non-coding regulatory mechanisms while facilitating the interpretation of non-coding variants.展开更多
Brassica oleracea has been developed into many important crops,including cabbage,kale,cauliflower,broccoli and so on.The genome and gene annotation of cabbage(cultivar JZS),a representative morphotype of B.oleracea,ha...Brassica oleracea has been developed into many important crops,including cabbage,kale,cauliflower,broccoli and so on.The genome and gene annotation of cabbage(cultivar JZS),a representative morphotype of B.oleracea,has been widely used as a common reference in biological research.Although its genome assembly has been updated twice,the current gene annotation still lacks information on untranslated regions(UTRs)and alternative splicing(AS).Here,we constructed a high-quality gene annotation(JZSv3)using a full-length transcriptome acquired by nanopore sequencing,yielding a total of 59452 genes and 75684 transcripts.Additionally,we re-analyzed the previously reported transcriptome data related to the development of different tissues and cold response using JZSv3 as a reference,and found that 3843 out of 11908 differentially expressed genes(DEGs)underwent AS during the development of different tissues and 309 out of 903 cold-related genes underwent AS in response to cold stress.Meanwhile,we also identified many AS genes,including BolLHCB5 and BolHSP70,that displayed distinct expression patterns within variant transcripts of the same gene,highlighting the importance of JZSv3 as a pivotal reference for AS analysis.Overall,JZSv3 provides a valuable resource for exploring gene function,especially for obtaining a deeper understanding of AS regulation mechanisms.展开更多
Deep learning has been widely applied in surrogate modeling for airfoil flow field prediction.The success of deep learning relies heavily on large-scale,high-quality labeled samples.However,acquiring labeled samples w...Deep learning has been widely applied in surrogate modeling for airfoil flow field prediction.The success of deep learning relies heavily on large-scale,high-quality labeled samples.However,acquiring labeled samples with complete annotations is prohibitively expensive,and the available annotations in practical engineering are often sparse due to limited observation.To leverage samples with sparse annotations,this paper proposes an uncertainty-based active transfer learning method.The most valuable positions in the flow field are selected based on uncertainty for annotation,effectively improving prediction accuracy and reducing annotation costs.Our method involves a novel active annotation based on synchronous quantile regression,which can mitigate the computational cost of query annotation.Besides,a novel quantile levels-based consistency regularization is proposed to constrain the remaining unlabeled regions and further improve the model performance.Experiments show that our method can significantly reduce prediction errors with only 1%extra annotations,and is a promising tool for achieving rapid and accurate flow field prediction.展开更多
●AIM:To investigate a pioneering framework for the segmentation of meibomian glands(MGs),using limited annotations to reduce the workload on ophthalmologists and enhance the efficiency of clinical diagnosis.●METHODS...●AIM:To investigate a pioneering framework for the segmentation of meibomian glands(MGs),using limited annotations to reduce the workload on ophthalmologists and enhance the efficiency of clinical diagnosis.●METHODS:Totally 203 infrared meibomian images from 138 patients with dry eye disease,accompanied by corresponding annotations,were gathered for the study.A rectified scribble-supervised gland segmentation(RSSGS)model,incorporating temporal ensemble prediction,uncertainty estimation,and a transformation equivariance constraint,was introduced to address constraints imposed by limited supervision information inherent in scribble annotations.The viability and efficacy of the proposed model were assessed based on accuracy,intersection over union(IoU),and dice coefficient.●RESULTS:Using manual labels as the gold standard,RSSGS demonstrated outcomes with an accuracy of 93.54%,a dice coefficient of 78.02%,and an IoU of 64.18%.Notably,these performance metrics exceed the current weakly supervised state-of-the-art methods by 0.76%,2.06%,and 2.69%,respectively.Furthermore,despite achieving a substantial 80%reduction in annotation costs,it only lags behind fully annotated methods by 0.72%,1.51%,and 2.04%.●CONCLUSION:An innovative automatic segmentation model is developed for MGs in infrared eyelid images,using scribble annotation for training.This model maintains an exceptionally high level of segmentation accuracy while substantially reducing training costs.It holds substantial utility for calculating clinical parameters,thereby greatly enhancing the diagnostic efficiency of ophthalmologists in evaluating meibomian gland dysfunction.展开更多
Funding agencies play a pivotal role in bolstering research endeavors by allocating financial resources for data collection and analysis.However,the lack of detailed information regarding the methods employed for data...Funding agencies play a pivotal role in bolstering research endeavors by allocating financial resources for data collection and analysis.However,the lack of detailed information regarding the methods employed for data gathering and analysis can obstruct the replication and utilization of the results,ultimately affecting the study’s transparency and integrity.The task of manually annotating extensive datasets demands considerable labor and financial investment,especially when it entails engaging specialized individuals.In our crowd counting study,we employed the web-based annotation tool SuperAnnotate to streamline the human annotation process for a dataset comprising 3,000 images.By integrating automated annotation tools,we realized substantial time efficiencies,as demonstrated by the remarkable achievement of 858,958 annotations.This underscores the significant contribution of such technologies to the efficiency of the annotation process.展开更多
Dealing with issues such as too simple image features and word noise inference in product image sentence anmotation, a product image sentence annotation model focusing on image feature learning and key words summariza...Dealing with issues such as too simple image features and word noise inference in product image sentence anmotation, a product image sentence annotation model focusing on image feature learning and key words summarization is described. Three kernel descriptors such as gradient, shape, and color are extracted, respectively. Feature late-fusion is executed in turn by the multiple kernel learning model to obtain more discriminant image features. Absolute rank and relative rank of the tag-rank model are used to boost the key words＇ weights. A new word integration algorithm named word sequence blocks building （WSBB） is designed to create N-gram word sequences. Sentences are generated according to the N-gram word sequences and predefined templates. Experimental results show that both the BLEU-1 scores and BLEU-2 scores of the sentences are superior to those of the state-of-art baselines.展开更多
It is very important in the field of bioinformatics to apply computer to perform the function annotation for new sequenced bio-sequences. Based on GO database and BLAST program, a novel method for the function annotat...It is very important in the field of bioinformatics to apply computer to perform the function annotation for new sequenced bio-sequences. Based on GO database and BLAST program, a novel method for the function annotation of new biological sequences is presented by using the variable-precision rough set theory. The proposed method is applied to the real data in GO database to examine its effectiveness. Numerical results show that the proposed method has better precision, recall-rate and harmonic mean value compared with existing methods.展开更多
基金supported by the National Natural Science Foundation of China(Nos.82274064,82374026,and 82204591)。
文摘Natural products(NPs)have long held a significant position in various fields such as medicine,food,agriculture,and materials.The chemical space covered by NPs is extensive but often underexplored.Therefore,high-throughput and efficient methodologies for the annotation and discovery of NPs are desired to address the complexity and diversity of NP-based systems.Mass spectrometry(MS)has emerged as a powerful platform for the annotation and discovery of NPs.MS databases provide vital support for the structural characterization of NPs by integrating extensive mass spectral data and sample information.Additionally,the released annotation methodologies,based on a variety of informatics tools,continuously improve the ability to annotate the structure and properties of compounds.This review examines the current mainstream databases and annotation methodologies,focusing on their advantages and limitations.Prospects for future technological advancements are then discussed in terms of novel applications and research objectives.Through a systematic overview,this review aims to provide valuable insights and a reference for MS-based NPs annotation,thereby promoting the discovery of novel natural entities.
基金Supported by Guangdong Basic and Applied Basic Research Foundation(No.2025A1515011627)San Ming Project of Medicine in Shenzhen(No.SZSM202311012).
文摘Glaucoma is an eye disease characterized by pathologically elevated intraocular pressure,optic nerve atrophy,and visual field defects,which can lead to irreversible vision loss.In recent years,the rapid development of artificial intelligence(AI)technology has provided new approaches for the early diagnosis and management of glaucoma.By classifying and annotating glaucoma-related images,AI models can learn and recognize the specific pathological features of glaucoma,thereby achieving automated imaging analysis and classification.Research on glaucoma imaging classification and annotation mainly involves color fundus photography(CFP),optical coherence tomography(OCT),anterior segment optical coherence tomography(AS-OCT),and ultrasound biomicroscopy(UBM)images.CFP is primarily used for the annotation of the optic cup and disc,while OCT is used for measuring and annotating the thickness of the retinal nerve fiber layer,and AS-OCT and UBM focus on the annotation of the anterior chamber angle structure and the measurement of anterior segment structural parameters.To standardize the classification and annotation of glaucoma images,enhance the quality and consistency of annotated data,and promote the clinical application of intelligent ophthalmology,this guideline has been developed.This guideline systematically elaborates on the principles,methods,processes,and quality control requirements for the classification and annotation of glaucoma images,providing standardized guidance for the classification and annotation of glaucoma images.
基金the National Key R&D Program of China(2022YFB3402100)the National Science Fund for Distinguished Young Scholars of China(52025056)+4 种基金the National Natural Science Foundation of China(52305129)the China Postdoctoral Science Foundation(2023M732789)the China Postdoctoral Innovative Talents Support Program(BX20230290)the Open Foundation of Hunan Provincial Key Laboratory of Health Maintenance for Mechanical Equipment(2022JXKF JJ01)the Fundamental Research Funds for Central Universities。
文摘The success of deep transfer learning in fault diagnosis is attributed to the collection of high-quality labeled data from the source domain.However,in engineering scenarios,achieving such high-quality label annotation is difficult and expensive.The incorrect label annotation produces two negative effects:1)the complex decision boundary of diagnosis models lowers the generalization performance on the target domain,and2)the distribution of target domain samples becomes misaligned with the false-labeled samples.To overcome these negative effects,this article proposes a solution called the label recovery and trajectory designable network(LRTDN).LRTDN consists of three parts.First,a residual network with dual classifiers is to learn features from cross-domain samples.Second,an annotation check module is constructed to generate a label anomaly indicator that could modify the abnormal labels of false-labeled samples in the source domain.With the training of relabeled samples,the complexity of diagnosis model is reduced via semi-supervised learning.Third,the adaptation trajectories are designed for sample distributions across domains.This ensures that the target domain samples are only adapted with the pure-labeled samples.The LRTDN is verified by two case studies,in which the diagnosis knowledge of bearings is transferred across different working conditions as well as different yet related machines.The results show that LRTDN offers a high diagnosis accuracy even in the presence of incorrect annotation.
基金supported by the National Natural Science Foundation of China(82171836)the 1·3·5 project for disciplines of excellence,West China Hospital,Sichuan University(ZYJC20002).
文摘The application of whole genome sequencing is expanding in clinical diagnostics across various genetic disorders, and the significance of non-coding variants in penetrant diseases is increasingly being demonstrated. Therefore, it is urgent to improve the diagnostic yield by exploring the pathogenic mechanisms of variants in non-coding regions. However, the interpretation of non-coding variants remains a significant challenge, due to the complex functional regulatory mechanisms of non-coding regions and the current limitations of available databases and tools. Hence, we develop the non-coding variant annotation database (NCAD, http://www.ncawdb.net/), encompassing comprehensive insights into 665,679,194 variants, regulatory elements, and element interaction details. Integrating data from 96 sources, spanning both GRCh37 and GRCh38 versions, NCAD v1.0 provides vital information to support the genetic diagnosis of non-coding variants, including allele frequencies of 12 diverse populations, with a particular focus on the population frequency information for 230,235,698 variants in 20,964 Chinese individuals. Moreover, it offers prediction scores for variant functionality, five categories of regulatory elements, and four types of non-coding RNAs. With its rich data and comprehensive coverage, NCAD serves as a valuable platform, empowering researchers and clinicians with profound insights into non-coding regulatory mechanisms while facilitating the interpretation of non-coding variants.
基金supported by the National Natural Science Foundation of China (Grant Nos.31972411,31722048,and 31630068)the Central Public-interest Scientific Institution Basal Research Fund (Grant No.Y2022PT23)+1 种基金the Innovation Program of the Chinese Academy of Agricultural Sciences,and the Key Laboratory of Biology and Genetic Improvement of Horticultural Crops,Ministry of Agriculture and Rural Affairs,P.R.Chinasupported by NIFA,the Department of Agriculture,via UC-Berkeley,USA。
文摘Brassica oleracea has been developed into many important crops,including cabbage,kale,cauliflower,broccoli and so on.The genome and gene annotation of cabbage(cultivar JZS),a representative morphotype of B.oleracea,has been widely used as a common reference in biological research.Although its genome assembly has been updated twice,the current gene annotation still lacks information on untranslated regions(UTRs)and alternative splicing(AS).Here,we constructed a high-quality gene annotation(JZSv3)using a full-length transcriptome acquired by nanopore sequencing,yielding a total of 59452 genes and 75684 transcripts.Additionally,we re-analyzed the previously reported transcriptome data related to the development of different tissues and cold response using JZSv3 as a reference,and found that 3843 out of 11908 differentially expressed genes(DEGs)underwent AS during the development of different tissues and 309 out of 903 cold-related genes underwent AS in response to cold stress.Meanwhile,we also identified many AS genes,including BolLHCB5 and BolHSP70,that displayed distinct expression patterns within variant transcripts of the same gene,highlighting the importance of JZSv3 as a pivotal reference for AS analysis.Overall,JZSv3 provides a valuable resource for exploring gene function,especially for obtaining a deeper understanding of AS regulation mechanisms.
基金supported by the National Natural Science Foundation of China(No.92371206).
文摘Deep learning has been widely applied in surrogate modeling for airfoil flow field prediction.The success of deep learning relies heavily on large-scale,high-quality labeled samples.However,acquiring labeled samples with complete annotations is prohibitively expensive,and the available annotations in practical engineering are often sparse due to limited observation.To leverage samples with sparse annotations,this paper proposes an uncertainty-based active transfer learning method.The most valuable positions in the flow field are selected based on uncertainty for annotation,effectively improving prediction accuracy and reducing annotation costs.Our method involves a novel active annotation based on synchronous quantile regression,which can mitigate the computational cost of query annotation.Besides,a novel quantile levels-based consistency regularization is proposed to constrain the remaining unlabeled regions and further improve the model performance.Experiments show that our method can significantly reduce prediction errors with only 1%extra annotations,and is a promising tool for achieving rapid and accurate flow field prediction.
基金Supported by Natural Science Foundation of Fujian Province(No.2020J011084)Fujian Province Technology and Economy Integration Service Platform(No.2023XRH001)Fuzhou-Xiamen-Quanzhou National Independent Innovation Demonstration Zone Collaborative Innovation Platform(No.2022FX5)。
文摘●AIM:To investigate a pioneering framework for the segmentation of meibomian glands(MGs),using limited annotations to reduce the workload on ophthalmologists and enhance the efficiency of clinical diagnosis.●METHODS:Totally 203 infrared meibomian images from 138 patients with dry eye disease,accompanied by corresponding annotations,were gathered for the study.A rectified scribble-supervised gland segmentation(RSSGS)model,incorporating temporal ensemble prediction,uncertainty estimation,and a transformation equivariance constraint,was introduced to address constraints imposed by limited supervision information inherent in scribble annotations.The viability and efficacy of the proposed model were assessed based on accuracy,intersection over union(IoU),and dice coefficient.●RESULTS:Using manual labels as the gold standard,RSSGS demonstrated outcomes with an accuracy of 93.54%,a dice coefficient of 78.02%,and an IoU of 64.18%.Notably,these performance metrics exceed the current weakly supervised state-of-the-art methods by 0.76%,2.06%,and 2.69%,respectively.Furthermore,despite achieving a substantial 80%reduction in annotation costs,it only lags behind fully annotated methods by 0.72%,1.51%,and 2.04%.●CONCLUSION:An innovative automatic segmentation model is developed for MGs in infrared eyelid images,using scribble annotation for training.This model maintains an exceptionally high level of segmentation accuracy while substantially reducing training costs.It holds substantial utility for calculating clinical parameters,thereby greatly enhancing the diagnostic efficiency of ophthalmologists in evaluating meibomian gland dysfunction.
基金The authors acknowledge that this research work was supported through project number UCS&T/R&D-09/19–20/17533 titled“An Intelligent Computational Model for Crowd Demonstration and Risk Analysis during Spiritual Events in Haridwar’by the Uttarakhand Council for Science and Technology(UCOST),India.”。
文摘Funding agencies play a pivotal role in bolstering research endeavors by allocating financial resources for data collection and analysis.However,the lack of detailed information regarding the methods employed for data gathering and analysis can obstruct the replication and utilization of the results,ultimately affecting the study’s transparency and integrity.The task of manually annotating extensive datasets demands considerable labor and financial investment,especially when it entails engaging specialized individuals.In our crowd counting study,we employed the web-based annotation tool SuperAnnotate to streamline the human annotation process for a dataset comprising 3,000 images.By integrating automated annotation tools,we realized substantial time efficiencies,as demonstrated by the remarkable achievement of 858,958 annotations.This underscores the significant contribution of such technologies to the efficiency of the annotation process.
基金The National Natural Science Foundation of China(No.61133012)the Humanity and Social Science Foundation of the Ministry of Education(No.12YJCZH274)+1 种基金the Humanity and Social Science Foundation of Jiangxi Province(No.XW1502,TQ1503)the Science and Technology Project of Jiangxi Science and Technology Department(No.20121BBG70050,20142BBG70011)
文摘Dealing with issues such as too simple image features and word noise inference in product image sentence anmotation, a product image sentence annotation model focusing on image feature learning and key words summarization is described. Three kernel descriptors such as gradient, shape, and color are extracted, respectively. Feature late-fusion is executed in turn by the multiple kernel learning model to obtain more discriminant image features. Absolute rank and relative rank of the tag-rank model are used to boost the key words＇ weights. A new word integration algorithm named word sequence blocks building （WSBB） is designed to create N-gram word sequences. Sentences are generated according to the N-gram word sequences and predefined templates. Experimental results show that both the BLEU-1 scores and BLEU-2 scores of the sentences are superior to those of the state-of-art baselines.
基金the support of the National Natural Science Foundation of China under Grant No.60673023,60433020,10501017,3040016the European Commission for TH/Asia Link/010 under Grant No.111084.
文摘It is very important in the field of bioinformatics to apply computer to perform the function annotation for new sequenced bio-sequences. Based on GO database and BLAST program, a novel method for the function annotation of new biological sequences is presented by using the variable-precision rough set theory. The proposed method is applied to the real data in GO database to examine its effectiveness. Numerical results show that the proposed method has better precision, recall-rate and harmonic mean value compared with existing methods.
