Jinzhong Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Jinzhong Xu has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Andrew Klapper
University of Kentucky
1 shared publication - Baokun Shi
Beijing University of Technology, China
1 shared publication - Bin Wu
Beijing University of Technology
1 shared publication Edgar Enochs
Department of Mathematics, University of Kentucky, Lexington, Kentucky 40506
1 shared publicationGikin Tan
The State Key Laboratory for Complex Severe and Rare Diseases; Beijing Key Laboratory of Intelligent Organ Biofabrication and Regenerative Repair; SXMU-Tsinghua Collaborative Innovation Center for Frontier Medicine; School of Basic Medical Sciences, Tsinghua Medicine, Tsinghua University, Beijing, 100084, China
1 shared publication- Hui Gao
The State Key Laboratory for Complex Severe and Rare Diseases; Beijing Key Laboratory of Intelligent Organ Biofabrication and Regenerative Repair; SXMU-Tsinghua Collaborative Innovation Center for Frontier Medicine; School of Basic Medical Sciences, Tsinghua Medicine, Tsinghua University, Beijing, 100084, China
1 shared publication
Latest Publications
- Open Access
- 17 July 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (29) , eaee0143
Spaceflight multi-omics reveals vulnerabilities of human germ cell development
Understanding the impact of spaceflight on human reproduction is critical for interplanetary exploration, yet technical barriers have limited direct studies of germ cell biology in orbit. Here, we utilized an automated bioreactor that supported long-term differentiation of human embryonic stem cells into human induced primordial germ cells (hiPGCs), human induced ovarian follicles (hiOFs), and human induced spermatogonial stem cells (hiSSCs) aboard spacecraft. Integrated real-time imaging, programmable medium perfusion, and in situ preservation enabled time-resolved multi-omics analysis. During missions on China’s Tianzhou-1 and Tianzhou-6 spacecraft, spaceflight reduced hiPGC specification efficiency by approximately 50% and suppressed hiSSC proliferation by 26%. Transcriptome-translatome coordination revealed cell-type-specific dysregulation of extracellular matrix organization, microtubule dynamics, and lipid metabolism. Whole-exome sequencing and DNA methylome analysis demonstrated preserved genomic integrity despite these functional perturbations. These findings provide direct evidence that spaceflight perturbs human germ cell development and establish a scalable framework for monitoring cellular adaptation during deep-space missions.
- 17 July 2026
- Published by American Association for the Advancement of Science (AAAS) in Science Advances
- Vol. 12 (29) , eaee0143
Open AccessSpaceflight multi-omics reveals vulnerabilities of human germ cell development
Understanding the impact of spaceflight on human reproduction is critical for interplanetary exploration, yet technical barriers have limited direct studies of germ cell biology in orbit. Here, we utilized an automated bioreactor that supported long-term differentiation of human embryonic stem cells into human induced primordial germ cells (hiPGCs), human induced ovarian follicles (hiOFs), and human induced spermatogonial stem cells (hiSSCs) aboard spacecraft. Integrated real-time imaging, programmable medium perfusion, and in situ preservation enabled time-resolved multi-omics analysis. During missions on China’s Tianzhou-1 and Tianzhou-6 spacecraft, spaceflight reduced hiPGC specification efficiency by approximately 50% and suppressed hiSSC proliferation by 26%. Transcriptome-translatome coordination revealed cell-type-specific dysregulation of extracellular matrix organization, microtubule dynamics, and lipid metabolism. Whole-exome sequencing and DNA methylome analysis demonstrated preserved genomic integrity despite these functional perturbations. These findings provide direct evidence that spaceflight perturbs human germ cell development and establish a scalable framework for monitoring cellular adaptation during deep-space missions.
- Open Access
- 3 November 2022
- Published by Association for Computing Machinery (ACM)
- p. 33-40
Distributed Learning based on Asynchronized Discriminator GAN for remote sensing image segmentation
Remote sensing images are usually distributed in different departments and contain private information, so they normally cannot be available publicly. However, it is a trend to jointly use remote sensing images from different departments, because it normally enables the model to capture more information and remote sensing image analysis based on deep learning generally requires lots of training data. To address the above problem, in this paper, we apply a distributed asynchronized discriminator GAN framework (DGAN) to jointly learn remote sensing images from different client nodes. The DGAN is composed of multiple distributed discriminators and a central generator, and only the synthetic remote sensing images generated by the DGAN are used to train a semantic segmentation model. Based on DGAN, we establish an experimental platform composed of multiple different hosts, which adopts socket and multi-process technology to realize asynchronous communication between hosts, and visualize the training and testing process. During DGAN training, instead of original remote sensing images or convolutional network model information, only synthetic images, losses and labeled images are exchanged between nodes. Therefore, the DGAN well protects the privacy and security of the original remote sensing images. We verify the performance of the DGAN on three remote sensing image datasets (City-OSM, WHU and Kaggle Ship). In the experiments, we take different distributions of remote sensing images in client nodes into consideration. The experiments show that the DGAN has a great capacity for distributed remote sensing image learning without sharing the original remote sensing images or the convolutional network model. Moreover, compared with a centralized GAN trained on all remote sensing images collected from all client nodes, the DGAN can achieve almost the same performance in semantic segmentation tasks for remote sensing images.
- 3 November 2022
- Published by Association for Computing Machinery (ACM)
- p. 33-40
Open AccessDistributed Learning based on Asynchronized Discriminator GAN for remote sensing image segmentation
Remote sensing images are usually distributed in different departments and contain private information, so they normally cannot be available publicly. However, it is a trend to jointly use remote sensing images from different departments, because it normally enables the model to capture more information and remote sensing image analysis based on deep learning generally requires lots of training data. To address the above problem, in this paper, we apply a distributed asynchronized discriminator GAN framework (DGAN) to jointly learn remote sensing images from different client nodes. The DGAN is composed of multiple distributed discriminators and a central generator, and only the synthetic remote sensing images generated by the DGAN are used to train a semantic segmentation model. Based on DGAN, we establish an experimental platform composed of multiple different hosts, which adopts socket and multi-process technology to realize asynchronous communication between hosts, and visualize the training and testing process. During DGAN training, instead of original remote sensing images or convolutional network model information, only synthetic images, losses and labeled images are exchanged between nodes. Therefore, the DGAN well protects the privacy and security of the original remote sensing images. We verify the performance of the DGAN on three remote sensing image datasets (City-OSM, WHU and Kaggle Ship). In the experiments, we take different distributions of remote sensing images in client nodes into consideration. The experiments show that the DGAN has a great capacity for distributed remote sensing image learning without sharing the original remote sensing images or the convolutional network model. Moreover, compared with a centralized GAN trained on all remote sensing images collected from all client nodes, the DGAN can achieve almost the same performance in semantic segmentation tasks for remote sensing images.
- Open Access
- 29 November 2019
- Published by Springer Nature in Journal of Food Science and Technology
- Vol. 57 (4) , 1216-1232
A comprehensive analysis of 13C isotope ratios data of authentic honey types produced in China using the EA-IRMS and LC-IRMS
In the current study, we have comprehensively analyzed different kinds of pure honey which was produced in various areas in China according to δ13C-EA -IRMS (AOAC method 998.12) and δ13C-LC-IRMS (proposed by the Intertek laboratory in Europe) methods. As for the δ13C-EA -IRMS method, the study was confirmed that the C4 sugar of all authentic honey samples was qualified. Further inter-laboratory comparison experiments using the δ13C-LC-IRMS method found that all authentic honey samples had Δδ13C (‰) values within the naturally occurring range of ± 1‰ for Δδ13C (‰) fru-glu. However, about 70% samples had Δδ13C (‰) values outside the range of ± 2.1‰ for Δδ13C (‰) max., indicating that a large proportion of pure honey in China can’t pass the δ13C-LC-IRMS test, although these honeys were extracted from unadulterated sources. Based on the present findings, we consider that the δ13C-LC-IRMS method is not appropriate to reliably detect adulterated honeys with C3 sugars in China.
- 29 November 2019
- Published by Springer Nature in Journal of Food Science and Technology
- Vol. 57 (4) , 1216-1232
Open AccessA comprehensive analysis of 13C isotope ratios data of authentic honey types produced in China using the EA-IRMS and LC-IRMS
In the current study, we have comprehensively analyzed different kinds of pure honey which was produced in various areas in China according to δ13C-EA -IRMS (AOAC method 998.12) and δ13C-LC-IRMS (proposed by the Intertek laboratory in Europe) methods. As for the δ13C-EA -IRMS method, the study was confirmed that the C4 sugar of all authentic honey samples was qualified. Further inter-laboratory comparison experiments using the δ13C-LC-IRMS method found that all authentic honey samples had Δδ13C (‰) values within the naturally occurring range of ± 1‰ for Δδ13C (‰) fru-glu. However, about 70% samples had Δδ13C (‰) values outside the range of ± 2.1‰ for Δδ13C (‰) max., indicating that a large proportion of pure honey in China can’t pass the δ13C-LC-IRMS test, although these honeys were extracted from unadulterated sources. Based on the present findings, we consider that the δ13C-LC-IRMS method is not appropriate to reliably detect adulterated honeys with C3 sugars in China.
- 1 January 1999
- Published by Springer Nature
- p. 379-392
Feedback with Carry Shift Registers over Z / (N)
- 1 January 1999
- Published by Springer Nature
- p. 379-392
Feedback with Carry Shift Registers over Z / (N)
- Open Access
- 1 April 1997
- Published by American Mathematical Society (AMS) in Proceedings of the American Mathematical Society
- Vol. 125 (4) , 951-960
On invariants dual to the Bass numbers
Let R R be a commutative Noetherian ring, and let M M be an R R -module. In earlier papers by Bass (1963) and Roberts (1980) the Bass numbers μ i ( p , M ) \mu _i(p,M) were defined for all primes p p and all integers i ≥ 0 i\ge 0 by use of the minimal injective resolution of M M . It is well known that μ i ( p , M ) = dim k ( p ) Ext R p i ( k ( p ) , M p ) \mu _i(p,M)=\dim _{k(p)}\operatorname {Ext} _{R_p}^i(k(p),M_p) . On the other hand, if M M is finitely generated, the Betti numbers β i ( p , M ) \beta _i(p,M) are defined by the minimal free resolution of M p M_p over the local ring R p R_p . In an earlier paper of the second author (1995), using the flat covers of modules, the invariants π i ( p , M ) \pi _i(p,M) were defined by the minimal flat resolution of M M over Gorenstein rings. The invariants π i ( p , M ) \pi _i(p,M) were shown to be somehow dual to the Bass numbers. In this paper, we use homologies to compute these invariants and show that \[ π i ( p , M ) = dim k ( p ) Tor i R p ( k ( p ) , Hom R ( R p , M ) ) \pi _i(p,M)= \dim _{k(p)}\operatorname {Tor}_i^{R_p}(k(p), \operatorname {Hom}_R(R_p,M)) \] for any cotorsion module M M . Comparing this with the computation of the Bass numbers, we see that Ext \operatorname {Ext} is replaced by Tor \operatorname {Tor} and the localization M p M_p is replaced by Hom R ( R p , M ) \operatorname {Hom}_R(R_p,M) (which was called the colocalization of M M at the prime ideal p p by Melkersson and Schenzel).
- 1 April 1997
- Published by American Mathematical Society (AMS) in Proceedings of the American Mathematical Society
- Vol. 125 (4) , 951-960
Open AccessOn invariants dual to the Bass numbers
Let R R be a commutative Noetherian ring, and let M M be an R R -module. In earlier papers by Bass (1963) and Roberts (1980) the Bass numbers μ i ( p , M ) \mu _i(p,M) were defined for all primes p p and all integers i ≥ 0 i\ge 0 by use of the minimal injective resolution of M M . It is well known that μ i ( p , M ) = dim k ( p ) Ext R p i ( k ( p ) , M p ) \mu _i(p,M)=\dim _{k(p)}\operatorname {Ext} _{R_p}^i(k(p),M_p) . On the other hand, if M M is finitely generated, the Betti numbers β i ( p , M ) \beta _i(p,M) are defined by the minimal free resolution of M p M_p over the local ring R p R_p . In an earlier paper of the second author (1995), using the flat covers of modules, the invariants π i ( p , M ) \pi _i(p,M) were defined by the minimal flat resolution of M M over Gorenstein rings. The invariants π i ( p , M ) \pi _i(p,M) were shown to be somehow dual to the Bass numbers. In this paper, we use homologies to compute these invariants and show that \[ π i ( p , M ) = dim k ( p ) Tor i R p ( k ( p ) , Hom R ( R p , M ) ) \pi _i(p,M)= \dim _{k(p)}\operatorname {Tor}_i^{R_p}(k(p), \operatorname {Hom}_R(R_p,M)) \] for any cotorsion module M M . Comparing this with the computation of the Bass numbers, we see that Ext \operatorname {Ext} is replaced by Tor \operatorname {Tor} and the localization M p M_p is replaced by Hom R ( R p , M ) \operatorname {Hom}_R(R_p,M) (which was called the colocalization of M M at the prime ideal p p by Melkersson and Schenzel).