Polyploidization is a major driver of speciation and its importance to plant evolution has been well recognized.Bamboos comprise one diploid herbaceous and three polyploid woody lineages,and are members of the only ma...Polyploidization is a major driver of speciation and its importance to plant evolution has been well recognized.Bamboos comprise one diploid herbaceous and three polyploid woody lineages,and are members of the only major subfamily in grasses that diversified in forests,with the woody members having a tree-like lignified culm.In this study,we generated four draft genome assemblies of major bamboo lineages with three different ploidy levels(diploid,tetraploid,and hexaploid).We also constructed a high-density genetic linkage map for a hexaploid species of bamboo,and used a linkage-map-based strategy for genome assembly and identification of subgenomes in polyploids.Further phylogenomic analyses using a large dataset of syntenic genes with expected copies based on ploidy levels revealed that woody bamboos originated subsequent to the divergence of the herbaceous bamboo lineage,and experienced complex reticulate evolution through three independent allopolyploid events involving four extinct diploid ancestors.A shared but distinct subgenome was identified in all polyploid forms,and the progenitor of this subgenome could have been critical in ancient polyploidizations and the origin of woody bamboos.Important genetic clues to the unique flowering behavior and woody trait in bamboos were also found.Taken together,our study provides significant insights into ancient reticulate evolution at the subgenome level in the absence of extant donor species,and offers a potential model scenario for broad-scale study of angiosperm origination by allopolyploidization.展开更多
Image-based breast tumor classification is an active and challenging problem.In this paper,a robust breast tumor classification framework is presented based on deep feature representation learning and exploiting avail...Image-based breast tumor classification is an active and challenging problem.In this paper,a robust breast tumor classification framework is presented based on deep feature representation learning and exploiting available information in existing samples.Feature representation learning of mammograms is fulfilled by a modified nonnegative matrix factorization model called LPML-LRNMF,which is motivated by hierarchical learning and layer-wise pre-training(LP)strategy in deep learning.Low-rank(LR)constraint is integrated into the feature representation learning model by considering the intrinsic characteristics of mammograms.Moreover,the proposed LPML-LRNMF model is optimized via alternating direction method of multipliers and the corresponding convergence is analyzed.For completing classification,an inverse projection sparse representation model is introduced to exploit information embedded in existing samples,especially in test ones.Experiments on the public dataset and actual clinical dataset show that the classification accuracy,specificity and sensitivity achieve the clinical acceptance level.展开更多
基金funding from Strategic Priority Program of the Chinese Academy of Sciences(CAS)(XDB31000000 to D.-Z.L)the National Natural Science Foundation of China(grants 31430011 to D.-Z.L.and 31670227 to Z.-H.G.)+3 种基金Leading Talents Program of Yunnan Province(2017HA014 to D.-Z.L.)CAS Youth Innovation Promotion Association(2015321 to P.-F.M.)a grant from Germplasm Bank of Wild Species(Y77P4412Z1 to Z.-H.G.)CAS Pioneer Hundred Talents Program(292015312D11035 to J.-Y.H.).
文摘Polyploidization is a major driver of speciation and its importance to plant evolution has been well recognized.Bamboos comprise one diploid herbaceous and three polyploid woody lineages,and are members of the only major subfamily in grasses that diversified in forests,with the woody members having a tree-like lignified culm.In this study,we generated four draft genome assemblies of major bamboo lineages with three different ploidy levels(diploid,tetraploid,and hexaploid).We also constructed a high-density genetic linkage map for a hexaploid species of bamboo,and used a linkage-map-based strategy for genome assembly and identification of subgenomes in polyploids.Further phylogenomic analyses using a large dataset of syntenic genes with expected copies based on ploidy levels revealed that woody bamboos originated subsequent to the divergence of the herbaceous bamboo lineage,and experienced complex reticulate evolution through three independent allopolyploid events involving four extinct diploid ancestors.A shared but distinct subgenome was identified in all polyploid forms,and the progenitor of this subgenome could have been critical in ancient polyploidizations and the origin of woody bamboos.Important genetic clues to the unique flowering behavior and woody trait in bamboos were also found.Taken together,our study provides significant insights into ancient reticulate evolution at the subgenome level in the absence of extant donor species,and offers a potential model scenario for broad-scale study of angiosperm origination by allopolyploidization.
基金supported by the Fundamental Research Funds for the Central Universities of China(No.30920130121013)the National Natural Science Foundation of China(No.31300111)
基金This work was supported in part by the National Natural Science Foundation of China(No.11701144)National Science Foundation of US(No.DMS1719932)+1 种基金Natural Science Foundation of Henan Province(No.162300410061)Project of Emerging Interdisciplinary(No.xxjc20170003).
文摘Image-based breast tumor classification is an active and challenging problem.In this paper,a robust breast tumor classification framework is presented based on deep feature representation learning and exploiting available information in existing samples.Feature representation learning of mammograms is fulfilled by a modified nonnegative matrix factorization model called LPML-LRNMF,which is motivated by hierarchical learning and layer-wise pre-training(LP)strategy in deep learning.Low-rank(LR)constraint is integrated into the feature representation learning model by considering the intrinsic characteristics of mammograms.Moreover,the proposed LPML-LRNMF model is optimized via alternating direction method of multipliers and the corresponding convergence is analyzed.For completing classification,an inverse projection sparse representation model is introduced to exploit information embedded in existing samples,especially in test ones.Experiments on the public dataset and actual clinical dataset show that the classification accuracy,specificity and sensitivity achieve the clinical acceptance level.
