Biology subjects

Lee, J.

Publications and source records attributed to Lee, J..

At least 721 records · Page 40Linked to original sources

bioRxiv2021

Arabidopsis ROOT UV-B SENSITIVE 1 and 2 Interact with Aminotransferases to Regulate Vitamin B6 Homeostasis

Pyridoxal-5-phosphate (PLP), the enzymatic cofactor form of Vitamin B6 (vitB6), is a versatile compound that has essential roles in metabolism. Cellular PLP homeostasic regulation is currently not well understood. Here we report that in Arabidopsis, biosynthesized PLP is sequestered by specific aminotransferases (ATs), and that the proteins ROOT UV-B SENSITIVE 1 (RUS1) and RUS2 function with ATs to regulate PLP homeostasis. The stunted growth phenotypes of rus1 and rus2 mutants were previously shown to be rescuable by exogenously supplied vitB6. Specific residue changes near the PLP-binding pocket in ASPARTATE AMINOTRANSFERASE2 (ASP2) also rescued rus1 and rus2 phenotypes. In this study, saturated suppressor screens identified 14 additional suppressor of rus (sor) alleles in four aminotransferase genes (ASP1, ASP2, ASP3, or ALANIN AMINOTRANSFERASE1 (AAT1)), which suppressed the rus phenotypes to varying degrees. Each of the sor mutations altered an amino acid in the PLP-binding pocket of the protein, and sor proteins were found to have reduced levels of PLP conjugation. Genetic data revealed that the availability of PLP normally requires both RUS1 and RUS2, and that increasing the number of sor mutants additively enhanced the suppression of rus phenotypes. Biochemical results showed that RUS1 and RUS2 physically interacted with ATs. Our studies suggest a mechanism in which RUS1, RUS2 and specific ATs work together to regulate PLP homeostasis in Arabidopsis.

genetics↗

bioRxiv2021

Automatic segmentation and cardiac mechanics analysis of evolving zebrafish using deep-learning

ObjectiveIn the study of early cardiac development, it is important to acquire accurate volume changes of the heart chambers. Although advanced imaging techniques, such as light-sheet fluorescent microscopy (LSFM), provide an accurate procedure for analyzing the structure of the heart, rapid and robust segmentation is required to reduce laborious time and accurately quantify developmental cardiac mechanics. MethodsThe traditional biomedical analysis involving segmentation of the intracardiac volume is usually carried out manually, presenting bottlenecks due to enormous data volume at high axial resolution. Our advanced deep-learning techniques provide a robust method to segment the volume within a few minutes. Our U-net based segmentation adopted manually segmented intracardiac volume changes as training data and produced the other LSFM zebrafish cardiac motion images automatically. ResultsThree cardiac cycles from 2 days post fertilization (dpf) to 5 dpf were successfully segmented by our U-net based network providing volume changes over time. In addition to understanding the cardiac function for each of the two chambers, the ventricle and atrium were separated by 3D erode morphology methods. Therefore, cardiac mechanical properties were measured rapidly and demonstrated incremental volume changes of both chambers separately. Interestingly, stroke volume (SV) remains similar in the atrium while that of the ventricle increases SV gradually. ConclusionOur U-net based segmentation provides a delicate method to segment the intricate inner volume of zebrafish heart during development; thus providing an accurate, robust and efficient algorithm to accelerate cardiac research by bypassing the labor-intensive task as well as improving the consistency in the results.

bioengineering↗

bioRxiv2021

Parallel Characterization of cis-Regulatory Elements for Multiple Genes UsingCRISPRpath

Current pooled CRISPR screens for cis-regulatory elements (CREs) can only accommodate one gene based on its expression level. Here, we describe CRISPRpath, a scalable screening strategy for parallelly characterizing CREs of genes linked to the same biological pathway and converging phenotypes. We demonstrate the ability of CRISPRpath for simultaneously identifying functional enhancers of six genes in the 6-thioguanine-induced DNA mismatch repair pathway using both CRISPR interference (CRISPRi) and CRISPR nuclease (CRISPRn) approaches. 60% of the identified enhancers are known promoters with distinct epigenomic features compared to other active promoters, including increased chromatin accessibility and interactivity. Furthermore, by imposing different levels of selection pressure, CRISPRpath can distinguish enhancers exerting strong impact on gene expression from those exerting weak impact. Our results offer a nuanced view of cis-regulation and demonstrate that CRISPRpath can be leveraged for understanding the complex gene regulatory program beyond transcriptional output at scale.

genomics↗

bioRxiv2021

Promoting axon regeneration by enhancing the non-coding function of the injury-responsive coding gene Gpr151

Gene expression profiling in response to nerve injury has been mainly focused on protein functions of coding genes to understand mechanisms of axon regeneration and to identify targets of potential therapeutics for nerve repair. However, the protein functions of several highly injury-induced genes including Gpr151 for regulating the regenerative ability remain unclear. Here we present an alternative approach focused on non-coding functions of the coding genes, which led to the identification of the non-coding function of Gpr151 RNA interacting with RNA- binding proteins such as CSDE1. Gpr151 promotes axon regeneration by the function of its 5- untranslated region (5UTR) and expression of an engineered form of the 5UTR improves regenerative capacity in vitro and in vivo in both sciatic nerve and optic nerve injury models. Our data suggest that searching injury-induced coding genes potentially functioning by their non- coding regions is required for the RNA-based gene therapy for improving axon regeneration.

neuroscience↗

bioRxiv2021

Crowdsourced Identification of Multi-Target Kinase Inhibitors for RET- and TAU- Based Disease: the Multi-Targeting Drug DREAM Challenge

A continuing challenge in modern medicine is the identification of safer and more efficacious drugs. Precision therapeutics, which have one molecular target, have been long promised to be safer and more effective than traditional therapies. This approach has proven to be challenging for multiple reasons including lack of efficacy, rapidly acquired drug resistance, and narrow patient eligibility criteria. An alternative approach is the development of drugs that address the overall disease network by targeting multiple biological targets ( polypharmacology). Rational development of these molecules will require improved methods for predicting single chemical structures that target multiple drug targets. To address this need, we developed the Multi-Targeting Drug DREAM Challenge, in which we challenged participants to predict single chemical entities that target pro-targets but avoid anti-targets for two unrelated diseases: RET-based tumors and a common form of inherited Tauopathy. Here, we report the results of this DREAM Challenge and the development of two neural network-based machine learning approaches that were applied to the challenge of rational polypharmacology. Together, these platforms provide a potentially useful first step towards developing lead therapeutic compounds that address disease complexity through rational polypharmacology. Author SummaryMany modern drugs are developed with the goal of modulating a single cellular pathway or target. However, many drugs are, in fact, dirty; they target multiple cellular pathways or targets. This phenomenon is known as multi-targeting or polypharmacology. While some strive to develop cleaner therapeutics that eliminate secondary targets, recent work has shown that multi-targeting therapeutics have key advantages for a variety of diseases. However, while multi-targeting drugs that affect a precisely-defined profile of targets may be more effective, it is difficult to computationally predict which molecules have desirable target profiles. Here, we report the results of a competitive crowdsourcing project (the Multi-Targeting Drug DREAM Challenge), where we challenged participants to predict chemicals that have desired target profiles for cancer and neurodegenerative disease.

pharmacology and toxicology↗

bioRxiv2021

Meta-analysis of scRNA seq data to define the landscape of human liver immune homeostasis.

The liver is unique in both its ability to maintain immune homeostasis and in its potential for immune tolerance following solid organ transplantation. Single-cell RNA sequencing (scRNA seq) is a powerful approach to generate highly dimensional transcriptome data to understand cellular phenotypes. However, when scRNA data is produced by different groups, with different data models, different standards, and samples processed in different ways, it can be challenging to draw meaningful conclusions from the aggregated data. The goal of this study was to establish a method to combine human liver scRNA seq datasets by 1) characterizing the heterogeneity between studies and 2) using the meta-atlas to define the dominant phenotypes across immune cell subpopulations in healthy human liver. Publicly available scRNA seq data generated from liver samples obtained from a combined total of 17 patients and [~]32,000 cells were analyzed. Liver-specific immune cells (CD45+) were extracted from each dataset, and immune cell subpopulations (myeloid cells, NK and T cells, plasma cells, and B cells) were examined using dimensionality reduction (UMAP), differential gene expression, and ingenuity pathway analysis. All datasets co-clustered, but cell proportions differed between studies. Gene expression correlation demonstrated similarity across all studies, and canonical pathways that differed between datasets were related to cell stress and oxidative phosphorylation rather than immune-related function. Next, a meta-atlas was generated via data integration and compared against PBMC data to define gene signatures for each hepatic immune subpopulation. This analysis defined key features of hepatic immune homeostasis, with decreased expression across immunologic pathways and enhancement of pathways involved with cell death. This method for meta-analysis of scRNA seq data provides a novel approach to broadly define the features of human liver immune homeostasis. Specific pathways and cellular phenotypes described in this human liver immune meta-atlas provide a critical reference point for further study of immune mediated disease processes within the liver.

immunology↗

bioRxiv2021

Memory destabilization during reconsolidation - A consequence of homeostatic plasticity?

Remembering is not a static process: when retrieved, a memory can be destabilized and become prone to modifications. This phenomenon has been demonstrated in a number of brain regions, but the neuronal mechanisms that rule memory destabilization and its boundary conditions remain elusive. Using two distinct computational models that combine Hebbian plasticity and synaptic downscaling, we show that homeostatic plasticity can function as a destabilization mechanism, accounting for behavioral results of protein synthesis inhibition upon reactivation with different reexposure times. Furthermore, by performing systematic reviews, we identify a series of overlapping molecular mechanisms between memory destabilization and synaptic downscaling, although direct experimental links between both phenomena remain scarce. In light of these results, we propose a theoretical framework where memory destabilization can emerge as an epiphenomenon of homeostatic adaptations prompted by memory retrieval.

neuroscience↗

bioRxiv2021

Differences in interactions between transmembrane domains tune the activation of metabotropic glutamate receptors

The metabotropic glutamate receptors (mGluRs) form a family of neuromodulatory G protein-coupled receptors that contain both a seven-helix transmembrane domain (TMD) and a large extracellular ligand-binding domain (LBD) which enables stable dimerization. While numerous studies have revealed variability across subtypes in the initial activation steps at the level of LBD dimers, an understanding of inter-TMD interaction and rearrangement remains limited. Here we use a combination of single molecule fluorescence, molecular dynamics, functional assays, and conformational sensors to reveal that distinct TMD assembly properties drive differences between mGluR subtypes. We uncover a variable region within transmembrane helix 4 (TM4) that contributes to homo- and heterodimerization in a subtype-specific manner and tunes orthosteric, allosteric and basal activation. We also confirm a critical role for a conserved inter-TM6 interface in stabilizing the active state during orthosteric or allosteric activation. Together this study informs a working model of inter-TMD rearrangement that drives mGluR function.

neuroscience↗

bioRxiv2021

Intrinsic bias at non-canonical, β-arrestin-coupled seven transmembrane receptors

G protein-coupled receptors (GPCRs) are typically characterized by their seven transmembrane (7TM) architecture, and interaction with two universal signal-transducers namely, the heterotrimeric G-proteins and {beta}-arrestins ({beta}arrs). Synthetic ligands and receptor mutants have been designed to elicit transducer-coupling preferences and distinct downstream signaling outcomes for many GPCRs. This raises the question if some naturally-occurring 7TMRs may selectively engage one of these two signal-transducers, even in response to their endogenous agonists. Although there are scattered hints in the literature that some 7TMRs lack G-protein coupling but interact with {beta}arrs, an in-depth understanding of their transducer-coupling preference, GRK-engagement, downstream signaling and structural mechanism remains elusive. Here, we use an array of cellular, biochemical and structural approaches to comprehensively characterize two non-canonical 7TMRs namely, the human decoy D6 receptor (D6R) and the human complement C5a receptor (C5aR2), in parallel with their canonical GPCR counterparts, CCR2 and C5aR1, respectively. We discover that D6R and C5aR2 couple exclusively to {beta}arrs, exhibit distinct GRK-preference, and activate non-canonical downstream signaling partners. We also observe that {beta}arrs, in complex with these receptors, adopt distinct conformations compared to their canonical GPCR counterparts despite being activated by a common natural agonist. Our study therefore establishes D6R and C5aR2 as bona-fide arrestin-coupled receptors (ACRs), and provides important insights into their regulation by GRKs and downstream signaling with direct implications for biased agonism.

biochemistry↗

bioRxiv2021

Expression and Manufacturing of Protein Therapeutics in Spirulina

Arthrospira platensis (commonly known as spirulina) is a photosynthetic cyanobacterium1. It is a highly nutritious food that has been consumed for decades in the US, and even longer by indigenous cultures2. Its widespread use as a safe food source and proven scalability have driven frequent attempts to convert it into a biomanufacturing platform. But these were repeatedly frustrated by spirulinas genetic intractability. We report here efficient and versatile genetic engineering methodology for spirulina that allows stable expression of bioactive protein therapeutics at high levels. We further describe large-scale, indoor cultivation and downstream processing methods appropriate for the manufacturing of biopharmaceuticals in spirulina. The potential of the platform is illustrated by pre-clinical development and human testing of an orally delivered antibody therapeutic against campylobacter, a major cause of infant mortality in the developing world and a growing antibiotic resistance threat3,4. This integrated development and manufacturing platform blends the safety of food-based biotechnology with the ease of genetic manipulation, rapid growth rates and high productivity characteristic of microbial platforms. These features combine for exceptionally low-cost production of biopharmaceuticals to address medical needs that are unfeasible with current biotechnology platforms.

molecular biology↗

bioRxiv2021

Human stem cell model of HNF1A deficiency shows uncoupled insulin to C-peptide secretion with accumulation of abnormal insulin granules

Mutations in HNF1A cause Maturity Onset Diabetes of the Young type 3 (MODY3), the most prevalent form of monogenic diabetes. We generated stem cell-derived pancreatic endocrine cells from human embryonic stem cells (hESCs) with induced hypomorphic mutations in HNF1A. Using these cells, we show that HNF1A orchestrates a transcriptional program required for distinct aspects of {beta}-cell fate and function. During islet cell differentiation, HNF1A deficiency biases islet endocrine cells towards an -cell fate associated with PAX4 down-regulation. HNF1A- deficient {beta}-cells display impaired basal and glucose stimulated-insulin secretion in association with a reduction in CACNA1A and intracellular calcium levels, and impaired insulin granule exocytosis in association with SYT13 down-regulation. Knockout of PAX4, CACNA1A and SYT13 reproduce the relevant phenotypes. Reduction of insulin secretion is associated with accumulation of enlarged secretory granules, and altered stoichiometry of secreted insulin to C-peptide. In HNF1A deficient {beta}-cells, glibenclamide, a sulfonylurea drug used in the treatment of MODY3 patients, increases intracellular calcium to levels beyond those achieved by glucose, and restores C-peptide and insulin secretion to a normal stoichiometric ratio. To study HNF1A deficiency in the context of a human disease model, we also generated stem cell-derived pancreatic endocrine cells from two MODY3 patients induced pluripotent stem cells (iPSCs). While insulin secretion defects are constitutive in cells with complete HNF1A loss of function, {beta}-cells heterozygous for hypomorphic HNF1A mutations are initially normal, but lose the ability to secrete insulin and acquire abnormal stoichiometric secretion ratios. Importantly, the defects observed in these stem cell models are also seen in circulating proportions of insulin:C-peptide in nine MODY3 patients. One sentence of summaryDeficiency of the transcription factor HNF1A biases islet endocrine cell fate towards -cells, impairs intracellular calcium homeostasis and insulin exocytosis, alters the stoichiometry of insulin to C-peptide release, and leads to an accumulation of abnormal insulin secretory granules in {beta}-cells.

cell biology↗

bioRxiv2021

4D chromosome reconstruction elucidates the spatial reorganization of the mammalian X-chromosome

Chromosomes are segmented into domains and compartments; yet, how these structures are spatially related in 3D is unclear. Here, by directly integrating Hi-C capture experiments and 3D modeling, we use X-inactivation as a model to examine the time evolution of 3D chromosome architecture during substantial changes in gene expression. First, we show that gene expression A/B compartments are consistent with phase separation in 3D space. Second, we show that residuals of smaller scale structures persist through transitions, despite further large-scale reorganization into the final inactive configuration, comprising two "megadomains". Interestingly, these previously hidden residual structures were not detectable in 2D Hi-C maps or principal component analyses. Third, time-dependent reaction-diffusion simulations reveal how Xist RNA particles diffuse across the 3D X-superstructure as it reorganizes. Our 4DHiC pipeline helps satisfy the growing demand for methodologies that produce 3D chromosome reconstructions directly from 2D datasets, which are consistent with the empirical data.

biophysics↗

bioRxiv2021

IGFBP2 protects against pulmonary fibrosis through inhibiting P21-mediated senescence

Accumulation of senescent cells contributes to age related diseases including idiopathic pulmonary fibrosis (IPF). Insulin-like growth factor binding proteins (IGFBPs) are evolutionarily conserved proteins that play a vital role in many biological processes. Overall, little is known about the functions of IGFBP2 in the epigenetic regulation of cellular senescence and pulmonary fibrosis. Here, we show that Igfbp2 expression was significantly downregulated at both mRNA and protein levels in a low-dose bleomycin-induced pulmonary fibrosis model of aged mice. Using the reduced representation of bisulfite sequencing technique, we demonstrated Igfbp2 downregulation is attributed to DNA methylation of CpG islands in fibrotic lungs of aged mice. Furthermore, Igfbp2 siRNA knockdown increased both P53 and P21 protein levels in mouse lung epithelial cells exposed to hypoxia treatment. Lentiviral mediated expression of Igfb2 decreased P21 protein levels and significantly reduced beta galactosidase activity in mouse lung epithelial cells challenged with a senescent drug (atazanavir) and hypoxia treatments. Using the RT2 Profiler PCR Array, we found that P21, PAI-1, IRF-5 and IRF-7, important regulators of senescence pathway, were significantly downregulated specifically in type-II alveolar epithelial cells (AECs) of aged human-Igfbp2 transgenic mice after bleomycin challenge. Finally, transgenic expression of human-Igfbp2 in type-II AECs from aged bleomycin challenged mice significantly decreased senescent associated secretory phenotype factors and also reduced extracellular matrix markers compared to aged wild-type mice challenged with bleomycin injury. Collectively, these findings reveal that epigenetic repression of Igfbp2 promotes pulmonary fibrosis and that restoring IGFBP2 in fibrotic lungs could prove effective in IPF treatment.

molecular biology↗

bioRxiv2021

Combined Enhanced Biological Phosphorus Removal (EBPR) and Nitrite Accumulation for Treating High-strength Wastewater

The enhanced biological phosphorus removal (EBPR) has been widely applied in treating domestic wastewater, while the performance on high-strength P wastewater is less investigated and the feasibility of coupling with short-cut nitrogen removal process remains unknown. This study first achieved the simultaneous high-efficient P removal and stable nitrite accumulation in one sequencing batch reactor for treating the synthetic digested manure wastewater. The average effluent P could be down to 0.8 {+/-} 1.0 mg P/L and the P removal efficiency was 99.5 {+/-} 0.8%. Candidatus Accumulibacter was the dominant polyphosphate accumulating organism (PAO) with the relative abundance of 14.2-33.1% in the reactor. Examination of the micro-diversity of Candidatus Accumulibacter using 16s rRNA gene-based oligotyping analysis revealed one unique Accumulibacter oligotype that different from the conventional system, which accounted for 64.2-87.9% of the total Accumulibacter abundance. The presence of high-abundant glycogen accumulating organisms (GAO) (15.6-40.3%, Defluviicoccus and Candidatus Competibacter) did not deteriorate the EBPR performance. Moreover, nitrite accumulation happened in the system with the effluent nitrite up to 20.4 {+/-} 6.4 mg N/L and the nitrite accumulation ratio was nearly 100% maintained for 140 days (420 cycles). Nitrosomonas was the dominant ammonia-oxidizing bacteria with relative abundance of 0.3-2.4% while nitrite-oxidizing bacteria were almost undetected (<0.1%). The introduction of extended anaerobic phase and high volatile fatty acid concentrations were proposed to be the potential selector forces to promote partial nitrification. This is the first study that combined EBPR with nitrite-accumulation for digested manure wastewater treatment, and it provided new sights in strategies to combine the EBPR and short-cut nitrogen removal via nitrite to achieve simultaneous nitrogen and phosphorus removal.

bioengineering↗

bioRxiv2021

Dermatan Sulfate Is a Potential Master Regulator of IgH via Interactions with Pre-BCR, GTF2I, and BiP ER Complex in Pre-B Lymphoblasts

Dermatan sulfate (DS) and autoantigen (autoAg) complexes are capable of stimulating autoreactive CD5+ B-1 cells. We examined the activity of DS on CD5+ pre-B lymphoblast NFS-25 cells. CD19, CD5, CD72, PI3K, and Fas possess varying degrees of DS affinity. The three pre-BCR components, Ig heavy chain mu (IgH), VpreB, and lambda 5, display differential DS affinities, with IgH having the strongest affinity. DS attaches to NFS-25 cells, gradually accumulates in the ER, and eventually localizes to the nucleus. DS and IgH co-localize on the cell surface and in the ER. DS associates strongly with 17 ER proteins (e.g., BiP/Grp78, Grp94, Hsp90ab1, Ganab, Vcp, Canx, Kpnb1, Prkcsh, Pdia3), which points to an IgH-associated multiprotein complex in the ER. In addition, DS interacts with nuclear proteins (Ncl, Xrcc6, Prmt5, Eftud2, Supt16h) and Lck. We also discovered that DS binds GTF2I, a required gene transcription factor at the IgH locus. These findings support DS as a potential master regulator of IgH in pre-B cells at protein and gene levels. We propose a (DS*autoAg)-autoBCR dual signal model in which an autoBCR is engaged by both autoAg and DS, and, once internalized, DS recruits a cascade of molecules that may help avert apoptosis and steer autoreactive B cell fate. Through its affinity with autoAgs and its control of IgH, DS emerges as a potential key player in the development of autoreactive B cells and autoimmunity.

immunology↗

bioRxiv2021

Molecular programs of fibrotic change in aging human lung

Aging is associated with both overt and subclinical lung fibrosis, which increases risk for mortality from viruses and other respiratory pathogens. The molecular programs that induce fibrosis in the aging lung are not well understood. To overcome this knowledge gap, we undertook multimodal profiling of distal lung samples from healthy human donors across the lifespan. Telomere shortening, a cause of cell senescence and fibrosis, was progressive with age in a sample of 86 lungs and was associated with foci of DNA damage. Bulk RNA sequencing confirmed activation of cellular senescence and pro-fibrotic pathways as well as genes necessary for collagen processing with increasing age. These findings were validated in independent datasets for lung and sun-exposed skin, but not other organs including heart, liver and kidney. Cell type deconvolution analysis revealed a progressive loss of lung epithelial cells and an increasing proportion of fibroblasts. Consistent with the observed pro-fibrotic transcriptional profile, second harmonic imaging demonstrated increased density of interstitial collagen in aged human lungs. Furthermore, regions of parenchymal fibrosis were associated with decreased alveolar expansion and surfactant secretion. These findings reveal the transcriptional and structural features of fibrosis and associated physiologic impairments in normal lung aging.

genomics↗

bioRxiv2021

The GA4GH Variation Representation Specification (VRS): a Computational Framework for the Precise Representation and Federated Identification of Molecular Variation

Maximizing the personal, public, research, and clinical value of genomic information will require that clinicians, researchers, and testing laboratories exchange genetic variation data reliably. Developed by a partnership among national information resource providers, public initiatives, and diagnostic testing laboratories under the auspices of the Global Alliance for Genomics and Health (GA4GH), the Variation Representation Specification (VRS, pronounced "verse") is an extensible framework for the semantically precise and computable representation of variation that complements contemporary human-readable and flat file standards for variation representation. VRS objects are designed to be semantically precise representations of variation, and leverage this design to enable unique, federated identification of molecular variation. We describe the components of this framework, including the terminology and information model, schema, data sharing conventions, and a reference implementation, each of which is intended to be broadly useful and freely available for community use. The specification, documentation, examples, and community links are available at https://vrs.ga4gh.org/.

bioinformatics↗

bioRxiv2021

Trans-channel fluorescence learning improves high-content screening for Alzheimer's disease therapeutics

In microscopy-based drug screens, fluorescent markers carry critical information on how compounds affect different biological processes. However, practical considerations may hinder the use of certain fluorescent markers. Here, we present a deep learning method for overcoming this limitation. We accurately generated predicted fluorescent signals from other related markers and validated this new machine learning (ML) method on two biologically distinct datasets. We used the ML method to improve the selection of biologically active compounds for Alzheimers disease (AD) from high-content high-throughput screening (HCS). The ML method identified novel compounds that effectively blocked tau aggregation, which would have been missed by traditional screening approaches unguided by ML. The method improved triaging efficiency of compound rankings over conventional rankings by raw image channels. We reproduced this ML pipeline on a biologically independent cancer-based dataset, demonstrating its generalizability. The approach is disease-agnostic and applicable across diverse fluorescence microscopy datasets.

pharmacology and toxicology↗