In image-guided radiation therapy, extracting features from medical point cloud is the key technique for multimodality registration. This novel framework, denoted Control Point Net (CPN), provides an alternative to th...In image-guided radiation therapy, extracting features from medical point cloud is the key technique for multimodality registration. This novel framework, denoted Control Point Net (CPN), provides an alternative to the common applications of manually designed keypoint descriptors for coarse point cloud registration. The CPN directly consumes a point cloud, divides it into equally spaced 3D voxels and transforms the points within each voxel into a unified feature representation through voxel feature encoding (VFE) layer. Then all volumetric representations are aggregated by Weighted Extraction Layer which selectively extracts features and synthesize into global descriptors and coordinates of control points. Utilizing global descriptors instead of local features allows the available geometrical data to be better exploited to improve the robustness and precision. Specifically, CPN unifies feature extraction and clustering into a single network, omitting time-consuming feature matching procedure. The algorithm is tested on point cloud datasets generated from CT images. Experiments and comparisons with the state-of-the-art descriptors demonstrate that CPN is highly discriminative, efficient, and robust to noise and density changes.展开更多
CRISPR homing gene drives have considerable potential for managing populations of medically and agriculturally significant insects.They operate by Cas9 cleavage followed by homology-directed repair,copying the drive a...CRISPR homing gene drives have considerable potential for managing populations of medically and agriculturally significant insects.They operate by Cas9 cleavage followed by homology-directed repair,copying the drive allele to the wild-type chromosome and thus increasing in frequency and spreading throughout a population.However,resistance alleles formed by end-joining repair pose a significant obstacle.To address this,we create a homing drive targeting the essential hairy gene in Drosophila melanogaster.Nonfunctional resistance alleles are recessive lethal,while drive carriers have a recoded“rescue”version of hairy.The drive inheritance rate is moderate,and multigenerational cage studies show drive spread to 96%–97%of the population.However,the drive does not reach 100%due to the formation of functional resistance alleles despite using four gRNAs.These alleles have a large deletion but likely utilize an alternate start codon.Thus,revised designs targeting more essential regions of a gene may be necessary to avoid such functional resistance.Replacement of the rescue element’s native 3'UTR with a homolog from another species increases drive inheritance by 13%–24%.This was possibly because of reduced homology between the rescue element and surrounding genomic DNA,which could also be an important design consideration for rescue gene drives.展开更多
文摘In image-guided radiation therapy, extracting features from medical point cloud is the key technique for multimodality registration. This novel framework, denoted Control Point Net (CPN), provides an alternative to the common applications of manually designed keypoint descriptors for coarse point cloud registration. The CPN directly consumes a point cloud, divides it into equally spaced 3D voxels and transforms the points within each voxel into a unified feature representation through voxel feature encoding (VFE) layer. Then all volumetric representations are aggregated by Weighted Extraction Layer which selectively extracts features and synthesize into global descriptors and coordinates of control points. Utilizing global descriptors instead of local features allows the available geometrical data to be better exploited to improve the robustness and precision. Specifically, CPN unifies feature extraction and clustering into a single network, omitting time-consuming feature matching procedure. The algorithm is tested on point cloud datasets generated from CT images. Experiments and comparisons with the state-of-the-art descriptors demonstrate that CPN is highly discriminative, efficient, and robust to noise and density changes.
基金supported by laboratory startup funds from Peking University and the Center for Life Sciences,as well as the grants from the National Science Foundation of China(32302455 and 32270672)。
文摘CRISPR homing gene drives have considerable potential for managing populations of medically and agriculturally significant insects.They operate by Cas9 cleavage followed by homology-directed repair,copying the drive allele to the wild-type chromosome and thus increasing in frequency and spreading throughout a population.However,resistance alleles formed by end-joining repair pose a significant obstacle.To address this,we create a homing drive targeting the essential hairy gene in Drosophila melanogaster.Nonfunctional resistance alleles are recessive lethal,while drive carriers have a recoded“rescue”version of hairy.The drive inheritance rate is moderate,and multigenerational cage studies show drive spread to 96%–97%of the population.However,the drive does not reach 100%due to the formation of functional resistance alleles despite using four gRNAs.These alleles have a large deletion but likely utilize an alternate start codon.Thus,revised designs targeting more essential regions of a gene may be necessary to avoid such functional resistance.Replacement of the rescue element’s native 3'UTR with a homolog from another species increases drive inheritance by 13%–24%.This was possibly because of reduced homology between the rescue element and surrounding genomic DNA,which could also be an important design consideration for rescue gene drives.
