Ying Liu
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
Research Subjects
Ying Liu has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
- Mandeep S. Singh
Wilmer Eye Institute, Johns Hopkins Hospital, Baltimore, Maryland
2 shared publications - Minda M. McNally
Wilmer Eye Institute Johns Hopkins University School of Medicine Baltimore Maryland USA
2 shared publications Alexis Biggs-Garcia
Wilmer Eye Institute, Johns Hopkins University School of Medicine, Baltimore, MD, USA
1 shared publication- Annabelle Pan
Johns Hopkins University School of Medicine, Baltimore, MD, USA
1 shared publication - Bani Antonio-Aguirre
Wilmer Eye Institute, Johns Hopkins Hospital, Baltimore, MD, USA
1 shared publication Baohan Pan
From the Department of Neurology, Johns Hopkins University, Baltimore, MD
1 shared publication
Latest Publications
- Open Access
- 3 December 2025
- Published by Wiley in Clinical and Translational Science
- Vol. 18 (12) , e70420
Application of Machine Learning to Discriminate Photoreceptor Cell Species in Xenotransplanted Chimeric Retinas
Photoreceptor transplantation is being studied to restore visual function in retinal diseases causing blindness, including age‐related macular degeneration, hereditary eye diseases, and traumatic retinopathy. Preclinical studies often involve delivering exogenous human photoreceptor cells into animal models' retinas. A key readout in such experiments is distinguishing donor cell integration from artificial labeling secondary to material transfer of cytosolic or nuclear labels. Recognizing donor (human) versus animal photoreceptor nuclei is key, but purely immunohistology discrimination is challenging due to antigenic species overlap or intercellular antigen transfer. To address this, we sought to develop and validate a computational technique to discriminate between photoreceptor cells of different animal species based on machine learning of nuclear morphology. We aim to evaluate the feasibility of computer‐assisted nuclear detection combined with random forest classification to automate species differentiation in DAPI‐stained photoreceptors after xenotransplantation into mouse and pig retinas. Our models were trained on single‐species samples and validated with mixed‐species samples. We then transplanted human embryonic stem cell‐derived retinal organoid cells into rodent and pig retinal degeneration models. The random forest model accurately determined cell identity post‐xenotransplantation, validated by histological assessment using an antihuman nuclear antibody. Our results support the potential efficacy of employing machine learning image analysis and classification techniques that may promote experimental rigor, minimize observer bias, and enable high throughput semiautomated workflows for transplantation outcomes analysis. The methodological framework reported here may enable a more nuanced and precise analysis of the behavior of transplanted photoreceptors for the purposes of human retinal regeneration.
- 3 December 2025
- Published by Wiley in Clinical and Translational Science
- Vol. 18 (12) , e70420
Open AccessApplication of Machine Learning to Discriminate Photoreceptor Cell Species in Xenotransplanted Chimeric Retinas
Photoreceptor transplantation is being studied to restore visual function in retinal diseases causing blindness, including age‐related macular degeneration, hereditary eye diseases, and traumatic retinopathy. Preclinical studies often involve delivering exogenous human photoreceptor cells into animal models' retinas. A key readout in such experiments is distinguishing donor cell integration from artificial labeling secondary to material transfer of cytosolic or nuclear labels. Recognizing donor (human) versus animal photoreceptor nuclei is key, but purely immunohistology discrimination is challenging due to antigenic species overlap or intercellular antigen transfer. To address this, we sought to develop and validate a computational technique to discriminate between photoreceptor cells of different animal species based on machine learning of nuclear morphology. We aim to evaluate the feasibility of computer‐assisted nuclear detection combined with random forest classification to automate species differentiation in DAPI‐stained photoreceptors after xenotransplantation into mouse and pig retinas. Our models were trained on single‐species samples and validated with mixed‐species samples. We then transplanted human embryonic stem cell‐derived retinal organoid cells into rodent and pig retinal degeneration models. The random forest model accurately determined cell identity post‐xenotransplantation, validated by histological assessment using an antihuman nuclear antibody. Our results support the potential efficacy of employing machine learning image analysis and classification techniques that may promote experimental rigor, minimize observer bias, and enable high throughput semiautomated workflows for transplantation outcomes analysis. The methodological framework reported here may enable a more nuanced and precise analysis of the behavior of transplanted photoreceptors for the purposes of human retinal regeneration.
- Open Access
- 1 September 2021
- Published by Elsevier in Experimental Eye Research
- Vol. 210, 108715
Characterization and allogeneic transplantation of a novel transgenic cone-rich donor mouse line
- 1 September 2021
- Published by Elsevier in Experimental Eye Research
- Vol. 210, 108715
Open AccessCharacterization and allogeneic transplantation of a novel transgenic cone-rich donor mouse line
- 12 December 2019
- Published by Springer Nature in Theoretical and Applied Genetics
- Vol. 133 (2) , 479-490
Fine mapping and candidate gene analysis of a seed glucosinolate content QTL, qGSL-C2, in rapeseed (Brassica napus L.)
Key message QTL mapping and candidate gene analysis indicate that allelic variations in BnaC2.MYB28 resulted from homeologous exchange and determine difference in seed glucosinolate content. Abstract A low seed glucosinolate content has long been an important breeding objective in rapeseed improvement. However, the molecular mechanisms underlying seed GSL content variations remain to be elucidated in allotetraploid Brassica napus. Here, we developed a double haploid population from a cross between two B. napus accessions that possess relatively low, but significantly different seed GSL contents and identified a major QTL, qGSL-C2, on chromosome C02 that explains 30.88–72.87% of the phenotypic variation observed in five environments. Using near-isogenic lines, we further delimited qGSL-C2 to a physical region of 49 kb on the B. rapa chromosome A02 which is highly homologous to the target C02 interval. Among five candidate genes, BnaC2.MYB28, a homologue of the Arabidopsis MYB28 encoding a putative R2R3-MYB-type transcription factor functioning in aliphatic methionine-derived GSL synthesis, was most likely to be the target gene underlying the QTL. Sequence analysis revealed multiple insertion/deletion and SNP variations in the genomic region between the alleles of the NILs. Furthermore, the allelic variations in BnaC2.MYB28 in the natural B. napus population were significantly associated with seed GSL content. Remarkably, the phylogenetic analysis and sequence comparison suggested that while the BnaC2.MYB28 allele from the parental line G120 was inherited from B. oleracea BolC2.MYB28, its counterpart from the other parent, 9172, most likely evolved from B. rapa BraA2.MYB28 via possible homeologous exchange. Our study promotes greater understanding of the molecular regulation of seed GSL content and provides useful molecular markers for seed GSL improvement in B. napus.
- 12 December 2019
- Published by Springer Nature in Theoretical and Applied Genetics
- Vol. 133 (2) , 479-490
Fine mapping and candidate gene analysis of a seed glucosinolate content QTL, qGSL-C2, in rapeseed (Brassica napus L.)
Key message QTL mapping and candidate gene analysis indicate that allelic variations in BnaC2.MYB28 resulted from homeologous exchange and determine difference in seed glucosinolate content. Abstract A low seed glucosinolate content has long been an important breeding objective in rapeseed improvement. However, the molecular mechanisms underlying seed GSL content variations remain to be elucidated in allotetraploid Brassica napus. Here, we developed a double haploid population from a cross between two B. napus accessions that possess relatively low, but significantly different seed GSL contents and identified a major QTL, qGSL-C2, on chromosome C02 that explains 30.88–72.87% of the phenotypic variation observed in five environments. Using near-isogenic lines, we further delimited qGSL-C2 to a physical region of 49 kb on the B. rapa chromosome A02 which is highly homologous to the target C02 interval. Among five candidate genes, BnaC2.MYB28, a homologue of the Arabidopsis MYB28 encoding a putative R2R3-MYB-type transcription factor functioning in aliphatic methionine-derived GSL synthesis, was most likely to be the target gene underlying the QTL. Sequence analysis revealed multiple insertion/deletion and SNP variations in the genomic region between the alleles of the NILs. Furthermore, the allelic variations in BnaC2.MYB28 in the natural B. napus population were significantly associated with seed GSL content. Remarkably, the phylogenetic analysis and sequence comparison suggested that while the BnaC2.MYB28 allele from the parental line G120 was inherited from B. oleracea BolC2.MYB28, its counterpart from the other parent, 9172, most likely evolved from B. rapa BraA2.MYB28 via possible homeologous exchange. Our study promotes greater understanding of the molecular regulation of seed GSL content and provides useful molecular markers for seed GSL improvement in B. napus.
- 12 August 2016
- Published by Springer Nature in AAPS PharmSciTech
- Vol. 18 (4) , 1374-1381
Folate-Functionalized Lipid Nanoemulsion to Deliver Chemo-Radiotherapeutics Together for the Effective Treatment of Nasopharyngeal Carcinoma
Aim of the investigation was to develop folate-functionalized lipid nanoemulsion (LNE) comprising chemo-radiotherapeutics for targeted delivery to nasopharyngeal carcinoma (NPC). Soy lecithin nanoemulsion of doxorubicin (Dox) and yittrium-90 (90Y) was prepared by nanoprecipitation using ultrasonic homogenization technique followed by folic acid conjugation. Nanoemulsion (Dox-LNE) was characterized as positively charged (zeta potential), spherical shape (transmission electron microscopy) nano-droplets of uniform size distribution (polydispersity index). No significant variation in parameters such as particle size, zeta potential, and polydispersity index was observed when the stability of Dox-LNE was assessed during long-term storage at room temperature and at 8000 rpm, 121°C temperature, and 5000 time dilution in water. In vitro release of Dox from Dox-LNE was observed to be controlled for at least 48 h. Folate decoration over Dox-LNE surface (FD-Dox-LNE) and incorporation of 90Y in FD-Dox-LNE (FD-Dox + 90Y-LNE) changed droplet size up to 50 nm; however, surface charge of Dox-LNE did not change significantly. FD-Dox + 90Y-LNE inhibited growth of cancerous cell line like CNE1 (folate receptor rich) in vitro and alleviated tumor volume in NPC-induced nude mice significantly as compared to Dox + 90Y-LNE. Massive necrosis and hemorrhage of CNE1 cells were observed by FD-Dox + 90Y-LNE (89.9%); however, inhibition of growth of nasal epithelial cells (RPMI 2650; folate deficient) by FD-Dox + 90Y-LNE and Dox + 90Y-LNE was observed to be 21.5 and 43.65%, respectively. The investigation highlights the vast utility of folate-decorated lipid emulsion in delivering chemo-radiotherapeutics to the specific NPC site. FD-Dox + 90Y-LNE might offer a cost-effective, safe, efficacious, and clinically pertinent option to the available therapeutics.
- 12 August 2016
- Published by Springer Nature in AAPS PharmSciTech
- Vol. 18 (4) , 1374-1381
Folate-Functionalized Lipid Nanoemulsion to Deliver Chemo-Radiotherapeutics Together for the Effective Treatment of Nasopharyngeal Carcinoma
Aim of the investigation was to develop folate-functionalized lipid nanoemulsion (LNE) comprising chemo-radiotherapeutics for targeted delivery to nasopharyngeal carcinoma (NPC). Soy lecithin nanoemulsion of doxorubicin (Dox) and yittrium-90 (90Y) was prepared by nanoprecipitation using ultrasonic homogenization technique followed by folic acid conjugation. Nanoemulsion (Dox-LNE) was characterized as positively charged (zeta potential), spherical shape (transmission electron microscopy) nano-droplets of uniform size distribution (polydispersity index). No significant variation in parameters such as particle size, zeta potential, and polydispersity index was observed when the stability of Dox-LNE was assessed during long-term storage at room temperature and at 8000 rpm, 121°C temperature, and 5000 time dilution in water. In vitro release of Dox from Dox-LNE was observed to be controlled for at least 48 h. Folate decoration over Dox-LNE surface (FD-Dox-LNE) and incorporation of 90Y in FD-Dox-LNE (FD-Dox + 90Y-LNE) changed droplet size up to 50 nm; however, surface charge of Dox-LNE did not change significantly. FD-Dox + 90Y-LNE inhibited growth of cancerous cell line like CNE1 (folate receptor rich) in vitro and alleviated tumor volume in NPC-induced nude mice significantly as compared to Dox + 90Y-LNE. Massive necrosis and hemorrhage of CNE1 cells were observed by FD-Dox + 90Y-LNE (89.9%); however, inhibition of growth of nasal epithelial cells (RPMI 2650; folate deficient) by FD-Dox + 90Y-LNE and Dox + 90Y-LNE was observed to be 21.5 and 43.65%, respectively. The investigation highlights the vast utility of folate-decorated lipid emulsion in delivering chemo-radiotherapeutics to the specific NPC site. FD-Dox + 90Y-LNE might offer a cost-effective, safe, efficacious, and clinically pertinent option to the available therapeutics.
- 21 April 2015
- Published by Wolters Kluwer Health in Neurology
- Vol. 84 (16) , 1652-1659
Factors influencing sweat gland innervation in diabetes
Background: Using a stereologic approach, the density of nerve fibers innervating sweat gland (SG) fragments in patients with diabetes mellitus (DM) and healthy controls using protein gene product (PGP), tyrosine hydroxylase (TH), and vasoactive intestinal peptide (VIP) was measured to determine which marker best detected differences between the groups. Factors associated with SG nerve fiber (SGNF) innervation were assessed and the change in SG innervation over a 1-year time period was determined. Methods: Ninety-two control subjects and 2 groups of subjects with DM totaling 97 were assessed in this cross-sectional study. Intraepidermal nerve fiber density and SG innervation were determined from leg skin biopsies that were immunohistochemically stained for ubiquitin hydrolase, VIP, and TH. Factors associated with SG innervation were assessed and 15 subjects were longitudinally followed for 1 year. Results: SGNF innervation was reduced in subjects with DM compared with controls. Lower SG innervation values were associated with increasing glycated hemoglobin A1c, body mass index (BMI), men compared with women, and tobacco use, but not diabetes type or age. Sex, A1c, and BMI remained significant in multivariate modeling. SG innervation measured by VIP+ fibers is a more sensitive marker for neuropathy than either PGP or TH. Fifteen subjects with DM followed for 1 year showed a significant decrease in SGNF innervation but not intraepidermal nerve fiber density. Conclusions: Stereologic measurement of SG innervation is feasible to assess postganglionic autonomic nerve fiber densities. SG innervation was reduced in subjects with DM compared with control subjects and was associated with sex, A1c, and BMI in multivariate modeling. VIP+ SGNF is more severely reduced in DM than TH+ or PGP9.5+-based assessments. Progression of diabetic polyneuropathy was detected by SGNF over a 1-year time period.
- 21 April 2015
- Published by Wolters Kluwer Health in Neurology
- Vol. 84 (16) , 1652-1659
Factors influencing sweat gland innervation in diabetes
Background: Using a stereologic approach, the density of nerve fibers innervating sweat gland (SG) fragments in patients with diabetes mellitus (DM) and healthy controls using protein gene product (PGP), tyrosine hydroxylase (TH), and vasoactive intestinal peptide (VIP) was measured to determine which marker best detected differences between the groups. Factors associated with SG nerve fiber (SGNF) innervation were assessed and the change in SG innervation over a 1-year time period was determined. Methods: Ninety-two control subjects and 2 groups of subjects with DM totaling 97 were assessed in this cross-sectional study. Intraepidermal nerve fiber density and SG innervation were determined from leg skin biopsies that were immunohistochemically stained for ubiquitin hydrolase, VIP, and TH. Factors associated with SG innervation were assessed and 15 subjects were longitudinally followed for 1 year. Results: SGNF innervation was reduced in subjects with DM compared with controls. Lower SG innervation values were associated with increasing glycated hemoglobin A1c, body mass index (BMI), men compared with women, and tobacco use, but not diabetes type or age. Sex, A1c, and BMI remained significant in multivariate modeling. SG innervation measured by VIP+ fibers is a more sensitive marker for neuropathy than either PGP or TH. Fifteen subjects with DM followed for 1 year showed a significant decrease in SGNF innervation but not intraepidermal nerve fiber density. Conclusions: Stereologic measurement of SG innervation is feasible to assess postganglionic autonomic nerve fiber densities. SG innervation was reduced in subjects with DM compared with control subjects and was associated with sex, A1c, and BMI in multivariate modeling. VIP+ SGNF is more severely reduced in DM than TH+ or PGP9.5+-based assessments. Progression of diabetic polyneuropathy was detected by SGNF over a 1-year time period.
- Open Access
- 1 March 2013
- Published by SAGE Publications in Stem Cells and Development
- Vol. 22 (5) , 689-694
C-MYC Controlled TIP110 Protein Expression Regulates OCT4 mRNA Splicing in Human Embryonic Stem Cells
We reported previously that HIV-1 the Tat-interacting protein of 110 kDa (TIP110; P110(NRB)/SART3/p110) is important in regulation of hematopoiesis, and in maintaining pluripotent factor (NANOG, OCT4, and SOX2) expression in and pluripotency of human embryonic stem cells (hESCs). Here we show that TIP110 expression in hESCs line H9 and embryonal carcinoma cell line NT-2 is regulated by C-MYC expression in ESCs through an E box present in the TIP110 promoter region. Through up- and down- modulation of expression, TIP110 induces OCT4 exon 1a inclusion and exon 1b skipping in our OCT4 minigene model. Thus, TIP110 expression in ESCs regulates alternative splicing of OCT4, an event required for pluripotency of ESCs.
- 1 March 2013
- Published by SAGE Publications in Stem Cells and Development
- Vol. 22 (5) , 689-694
Open AccessC-MYC Controlled TIP110 Protein Expression Regulates OCT4 mRNA Splicing in Human Embryonic Stem Cells
We reported previously that HIV-1 the Tat-interacting protein of 110 kDa (TIP110; P110(NRB)/SART3/p110) is important in regulation of hematopoiesis, and in maintaining pluripotent factor (NANOG, OCT4, and SOX2) expression in and pluripotency of human embryonic stem cells (hESCs). Here we show that TIP110 expression in hESCs line H9 and embryonal carcinoma cell line NT-2 is regulated by C-MYC expression in ESCs through an E box present in the TIP110 promoter region. Through up- and down- modulation of expression, TIP110 induces OCT4 exon 1a inclusion and exon 1b skipping in our OCT4 minigene model. Thus, TIP110 expression in ESCs regulates alternative splicing of OCT4, an event required for pluripotency of ESCs.
- 2 May 2012
- Published by Springer Nature in Inflammation
- Vol. 35 (4) , 1558-1566
HSF1 Is a Transcriptional Activator of IL-10 Gene Expression in RAW264.7 Macrophages
The heat shock transcription factor (HSF) is an important transactivator of the heat shock genes. Recent studies have shown that HSF1 acts as a repressor of non-heat shock genes to protect against endotoxemia. In this study, we found that heat shock treatment and HSF1 over-expression augmented the induction of interleukin (IL)-10 mRNA. Computational analysis of the mouse IL-10 promoter region showed that three potential heat shock elements (HSEs) were located at mouse IL-10 gene promoter, among which only the −387/−360 probe formed a complex with HSF1. The lack of binding of the other two HSEs to HSF1 suggested the critical role of the flanking sequences in the binding specificity of HSE to HSF1. Moreover, we showed that HSF1 overexpression transactivated mouse IL-10 gene promoter and this transcriptional activation was inhibited by the mutation of HSE in the −387/−360 region of IL-10 gene promoter using luciferase reporter assay. These findings indicate that HSF1 is a transcriptional activator of anti-inflammatory mediator IL-10 gene in RAW264.7 macrophages.
- 2 May 2012
- Published by Springer Nature in Inflammation
- Vol. 35 (4) , 1558-1566
HSF1 Is a Transcriptional Activator of IL-10 Gene Expression in RAW264.7 Macrophages
The heat shock transcription factor (HSF) is an important transactivator of the heat shock genes. Recent studies have shown that HSF1 acts as a repressor of non-heat shock genes to protect against endotoxemia. In this study, we found that heat shock treatment and HSF1 over-expression augmented the induction of interleukin (IL)-10 mRNA. Computational analysis of the mouse IL-10 promoter region showed that three potential heat shock elements (HSEs) were located at mouse IL-10 gene promoter, among which only the −387/−360 probe formed a complex with HSF1. The lack of binding of the other two HSEs to HSF1 suggested the critical role of the flanking sequences in the binding specificity of HSE to HSF1. Moreover, we showed that HSF1 overexpression transactivated mouse IL-10 gene promoter and this transcriptional activation was inhibited by the mutation of HSE in the −387/−360 region of IL-10 gene promoter using luciferase reporter assay. These findings indicate that HSF1 is a transcriptional activator of anti-inflammatory mediator IL-10 gene in RAW264.7 macrophages.