Yuheng (Rene) Cai

All Publications

• Intelligent Beam Optimization for Light-Sheet Fluorescence Microscopy through Deep Learning. Intelligent computing (Washington, D.C.) Li, C., Rai, M. R., Cai, Y., Ghashghaei, H. T., Greenbaum, A. 2024; 3

  Abstract

  Light-sheet fluorescence microscopy (LSFM) provides the benefit of optical sectioning coupled with rapid acquisition times, enabling high-resolution 3-dimensional imaging of large tissue-cleared samples. Inherent to LSFM, the quality of the imaging heavily relies on the characteristics of the illumination beam, which only illuminates a thin section of the sample. Therefore, substantial efforts are dedicated to identifying slender, nondiffracting beam profiles that yield uniform and high-contrast images. An ongoing debate concerns the identification of optimal illumination beams for different samples: Gaussian, Bessel, Airy patterns, and/or others. However, comparisons among different beam profiles are challenging as their optimization objectives are often different. Given that our large imaging datasets (approximately 0.5 TB of images per sample) are already analyzed using deep learning models, we envisioned a different approach to the problem by designing an illumination beam tailored to boost the performance of the deep learning model. We hypothesized that integrating the physical LSFM illumination model (after passing it through a variable phase mask) into the training of a cell detection network would achieve this goal. Here, we report that joint optimization continuously updates the phase mask and results in improved image quality for better cell detection. The efficacy of our method is demonstrated through both simulations and experiments that reveal substantial enhancements in imaging quality compared to the traditional Gaussian light sheet. We discuss how designing microscopy systems through a computational approach provides novel insights for advancing optical design that relies on deep learning models for the analysis of imaging datasets.

  View details for DOI 10.34133/icomputing.0095

  View details for PubMedID 39099879

  View details for PubMedCentralID PMC11298055

• COMBINe enables automated detection and classification of neurons and astrocytes in tissue-cleared mouse brains. Cell reports methods Cai, Y., Zhang, X., Li, C., Ghashghaei, H. T., Greenbaum, A. 2023; 3 (4): 100454

  Abstract

  Tissue clearing renders entire organs transparent to accelerate whole-tissue imaging; for example, with light-sheet fluorescence microscopy. Yet, challenges remain in analyzing the large resulting 3D datasets that consist of terabytes of images and information on millions of labeled cells. Previous work has established pipelines for automated analysis of tissue-cleared mouse brains, but the focus there was on single-color channels and/or detection of nuclear localized signals in relatively low-resolution images. Here, we present an automated workflow (COMBINe, Cell detectiOn in Mouse BraIN) to map sparsely labeled neurons and astrocytes in genetically distinct mouse forebrains using mosaic analysis with double markers (MADM). COMBINe blends modules from multiple pipelines with RetinaNet at its core. We quantitatively analyzed the regional and subregional effects of MADM-based deletion of the epidermal growth factor receptor (EGFR) on neuronal and astrocyte populations in the mouse forebrain.

  View details for DOI 10.1016/j.crmeth.2023.100454

  View details for PubMedID 37159668

  View details for PubMedCentralID PMC10163164

• Tissue clearing and three-dimensional imaging of the whole cochlea and vestibular system from multiple large-animal models. STAR protocols Moatti, A., Cai, Y., Li, C., Popowski, K. D., Cheng, K., Ligler, F. S., Greenbaum, A. 2023; 4 (2): 102220

  Abstract

  The inner ear of humans and large animals is embedded in a thick and dense bone that makes dissection challenging. Here, we present a protocol that enables three-dimensional (3D) characterization of intact inner ears from large-animal models. We describe steps for decalcifying bone, using solvents to remove color and lipids, and imaging tissues in 3D using confocal and light sheet microscopy. We then detail a pipeline to count hair cells in antibody-stained and 3D imaged cochleae using open-source software. For complete details on the use and execution of this protocol, please refer to (Moatti et al., 2022).1.

  View details for DOI 10.1016/j.xpro.2023.102220

  View details for PubMedID 37060559

  View details for PubMedCentralID PMC10140170

• Bulk and mosaic deletions of Egfr reveal regionally defined gliogenesis in the developing mouse forebrain. iScience Zhang, X., Xiao, G., Johnson, C., Cai, Y., Horowitz, Z. K., Mennicke, C., Coffey, R., Haider, M., Threadgill, D., Eliscu, R., Oldham, M. C., Greenbaum, A., Ghashghaei, H. T. 2023; 26 (3): 106242

  Abstract

  The epidermal growth factor receptor (EGFR) plays a role in cell proliferation and differentiation during healthy development and tumor growth; however, its requirement for brain development remains unclear. Here we used a conditional mouse allele for Egfr to examine its contributions to perinatal forebrain development at the tissue level. Subtractive bulk ventral and dorsal forebrain deletions of Egfr uncovered significant and permanent decreases in oligodendrogenesis and myelination in the cortex and corpus callosum. Additionally, an increase in astrogenesis or reactive astrocytes in effected regions was evident in response to cortical scarring. Sparse deletion using mosaic analysis with double markers (MADM) surprisingly revealed a regional requirement for EGFR in rostrodorsal, but not ventrocaudal glial lineages including both astrocytes and oligodendrocytes. The EGFR-independent ventral glial progenitors may compensate for the missing EGFR-dependent dorsal glia in the bulk Egfr-deleted forebrain, potentially exposing a regenerative population of gliogenic progenitors in the mouse forebrain.

  View details for DOI 10.1016/j.isci.2023.106242

  View details for PubMedID 36915679

  View details for PubMedCentralID PMC10006693

• Maternal organophosphate flame retardant exposure alters the developing mesencephalic dopamine system in fetal rat. Toxicological sciences : an official journal of the Society of Toxicology Newell, A. J., Kapps, V. A., Cai, Y., Rai, M. R., St Armour, G., Horman, B. M., Rock, K. D., Witchey, S. K., Greenbaum, A., Patisaul, H. B. 2023; 191 (2): 357-373

  Abstract

  Organophosphate flame retardants (OPFRs) have become the predominant substitution for legacy brominated flame retardants but there is concern about their potential developmental neurotoxicity (DNT). OPFRs readily dissociate from the fireproofed substrate to the environment, and they (or their metabolites) have been detected in diverse matrices including air, water, soil, and biota, including human urine and breastmilk. Given this ubiquitous contamination, it becomes increasingly important to understand the potential effects of OPFRs on the developing nervous system. We have previously shown that maternal exposure to OPFRs results in neuroendocrine disruption, alterations to developmental metabolism of serotonin (5-HT) and axonal extension in male fetal rats, and potentiates adult anxiety-like behaviors. The development of the serotonin and dopamine systems occur in parallel and interact, therefore, we first sought to enhance our prior 5-HT work by first examining the ascending 5-HT system on embryonic day 14 using whole mount clearing of fetal heads and 3-dimensional (3D) brain imaging. We also investigated the effects of maternal OPFR exposure on the development of the mesocortical dopamine system in the same animals through 2-dimensional and 3D analysis following immunohistochemistry for tyrosine hydroxylase (TH). Maternal OPFR exposure induced morphological changes to the putative ventral tegmental area and substantia nigra in both sexes and reduced the overall volume of this structure in males, whereas 5-HT nuclei were unchanged. Additionally, dopaminergic axogenesis was disrupted in OPFR exposed animals, as the dorsoventral spread of ventral telencephalic TH afferents were greater at embryonic day 14, while sparing 5-HT fibers. These results indicate maternal exposure to OPFRs alters the development trajectory of the embryonic dopaminergic system and adds to growing evidence of OPFR DNT.

  View details for DOI 10.1093/toxsci/kfac137

  View details for PubMedID 36562574

  View details for PubMedCentralID PMC9936211

• Inhalable dry powder mRNA vaccines based on extracellular vesicles. Matter Popowski, K. D., Moatti, A., Scull, G., Silkstone, D., Lutz, H., López de Juan Abad, B., George, A., Belcher, E., Zhu, D., Mei, X., Cheng, X., Cislo, M., Ghodsi, A., Cai, Y., Huang, K., Li, J., Brown, A. C., Greenbaum, A., Dinh, P. C., Cheng, K. 2022; 5 (9): 2960-2974

  Abstract

  Respiratory diseases are a global burden, with millions of deaths attributed to pulmonary illnesses and dysfunctions. Therapeutics have been developed, but they present major limitations regarding pulmonary bioavailability and product stability. To circumvent such limitations, we developed room-temperature-stable inhalable lung-derived extracellular vesicles or exosomes (Lung-Exos) as mRNA and protein drug carriers. Compared with standard synthetic nanoparticle liposomes (Lipos), Lung-Exos exhibited superior distribution to the bronchioles and parenchyma and are deliverable to the lungs of rodents and nonhuman primates (NHPs) by dry powder inhalation. In a vaccine application, severe acute respiratory coronavirus 2 (SARS-CoV-2) spike (S) protein encoding mRNA-loaded Lung-Exos (S-Exos) elicited greater immunoglobulin G (IgG) and secretory IgA (SIgA) responses than its loaded liposome (S-Lipo) counterpart. Importantly, S-Exos remained functional at room-temperature storage for one month. Our results suggest that extracellular vesicles can serve as an inhaled mRNA drug-delivery system that is superior to synthetic liposomes.

  View details for DOI 10.1016/j.matt.2022.06.012

  View details for PubMedID 35847197

  View details for PubMedCentralID PMC9272513

• Ontogeny of cellular organization and LGR5 expression in porcine cochlea revealed using tissue clearing and 3D imaging. iScience Moatti, A., Li, C., Sivadanam, S., Cai, Y., Ranta, J., Piedrahita, J. A., Cheng, A. G., Ligler, F. S., Greenbaum, A. 2022; 25 (8): 104695

  Abstract

  Over 11% of the world's population experience hearing loss. Although there are promising studies to restore hearing in rodent models, the size, ontogeny, genetics, and frequency range of hearing of most rodents' cochlea do not match that of humans. The porcine cochlea can bridge this gap as it shares many anatomical, physiological, and genetic similarities with its human counterpart. Here, we provide a detailed methodology to process and image the porcine cochlea in 3D using tissue clearing and light-sheet microscopy. The resulting3D images can be employed to compare cochleae across different ages and conditions, investigate the ontogeny of cochlear cytoarchitecture, and produce quantitative expression maps of LGR5, a marker of cochlear progenitors in mice. These data reveal that hair cell organization, inner ear morphology, cellular cartography in the organ of Corti, and spatiotemporal expression of LGR5 are dynamic over developmental stages in a pattern not previously documented.

  View details for DOI 10.1016/j.isci.2022.104695

  View details for PubMedID 35865132

• Automated Annotation of Untargeted All-Ion Fragmentation LC-MS Metabolomics Data with MetaboAnnotatoR. Analytical chemistry Graça, G., Cai, Y., Lau, C. E., Vorkas, P. A., Lewis, M. R., Want, E. J., Herrington, D., Ebbels, T. M. 2022; 94 (8): 3446-3455

  Abstract

  Untargeted metabolomics and lipidomics LC-MS experiments produce complex datasets, usually containing tens of thousands of features from thousands of metabolites whose annotation requires additional MS/MS experiments and expert knowledge. All-ion fragmentation (AIF) LC-MS/MS acquisition provides fragmentation data at no additional experimental time cost. However, analysis of such datasets requires reconstruction of parent-fragment relationships and annotation of the resulting pseudo-MS/MS spectra. Here, we propose a novel approach for automated annotation of isotopologues, adducts, and in-source fragments from AIF LC-MS datasets by combining correlation-based parent-fragment linking with molecular fragment matching. Our workflow focuses on a subset of features rather than trying to annotate the full dataset, saving time and simplifying the process. We demonstrate the workflow in three human serum datasets containing 599 features manually annotated by experts. Precision and recall values of 82-92% and 82-85%, respectively, were obtained for features found in the highest-rank scores (1-5). These results equal or outperform those obtained using MS-DIAL software, the current state of the art for AIF data annotation. Further validation for other biological matrices and different instrument types showed variable precision (60-89%) and recall (10-88%) particularly for datasets dominated by nonlipid metabolites. The workflow is freely available as an open-source R package, MetaboAnnotatoR, together with the fragment libraries from Github (https://github.com/gggraca/MetaboAnnotatoR).

  View details for DOI 10.1021/acs.analchem.1c03032

  View details for PubMedID 35180347

  View details for PubMedCentralID PMC8892435

• Detection and classification of neurons and glial cells in the MADM mouse brain using RetinaNet. PloS one Cai, Y., Zhang, X., Kovalsky, S. Z., Ghashghaei, H. T., Greenbaum, A. 2021; 16 (9): e0257426

  Abstract

  The ability to automatically detect and classify populations of cells in tissue sections is paramount in a wide variety of applications ranging from developmental biology to pathology. Although deep learning algorithms are widely applied to microscopy data, they typically focus on segmentation which requires extensive training and labor-intensive annotation. Here, we utilized object detection networks (neural networks) to detect and classify targets in complex microscopy images, while simplifying data annotation. To this end, we used a RetinaNet model to classify genetically labeled neurons and glia in the brains of Mosaic Analysis with Double Markers (MADM) mice. Our initial RetinaNet-based model achieved an average precision of 0.90 across six classes of cells differentiated by MADM reporter expression and their phenotype (neuron or glia). However, we found that a single RetinaNet model often failed when encountering dense and saturated glial clusters, which show high variability in their shape and fluorophore densities compared to neurons. To overcome this, we introduced a second RetinaNet model dedicated to the detection of glia clusters. Merging the predictions of the two computational models significantly improved the automated cell counting of glial clusters. The proposed cell detection workflow will be instrumental in quantitative analysis of the spatial organization of cellular populations, which is applicable not only to preparations in neuroscience studies, but also to any tissue preparation containing labeled populations of cells.

  View details for DOI 10.1371/journal.pone.0257426

  View details for PubMedID 34559842

  View details for PubMedCentralID PMC8462685

• Three-dimensional imaging of intact porcine cochlea using tissue clearing and custom-built light-sheet microscopy. Biomedical optics express Moatti, A., Cai, Y., Li, C., Sattler, T., Edwards, L., Piedrahita, J., Ligler, F. S., Greenbaum, A. 2020; 11 (11): 6181-6196

  Abstract

  Hearing loss is a prevalent disorder that affects people of all ages. On top of the existing hearing aids and cochlear implants, there is a growing effort to regenerate functional tissues and restore hearing. However, studying and evaluating these regenerative medicine approaches in a big animal model (e.g. pigs) whose anatomy, physiology, and organ size are similar to a human is challenging. In big animal models, the cochlea is bulky, intricate, and veiled in a dense and craggy otic capsule. These facts complicate 3D microscopic analysis that is vital in the cochlea, where structure-function relation is time and again manifested. To allow 3D imaging of an intact cochlea of newborn and juvenile pigs with a volume up to ∼ 250 mm3, we adapted the BoneClear tissue clearing technique, which renders the bone transparent. The transparent cochleae were then imaged with cellular resolution and in a timely fashion, which prevented bubble formation and tissue degradation, using an adaptive custom-built light-sheet fluorescence microscope. The adaptive light-sheet microscope compensated for deflections of the illumination beam by changing the angles of the beam and translating the detection objective while acquiring images. Using this combination of techniques, macroscopic and microscopic properties of the cochlea were extracted, including the density of hair cells, frequency maps, and lower frequency limits. Consequently, the proposed platform could support the growing effort to regenerate cochlear tissues and assist with basic research to advance cures for hearing impairments.

  View details for DOI 10.1364/BOE.402991

  View details for PubMedID 33282483

  View details for PubMedCentralID PMC7687970