Biology subjects

Li, R.

Publications and source records attributed to Li, R..

At least 217 records · Page 12Linked to original sources

bioRxiv2025

Enhanced Detection of RNA Modifications in Escherichia coli Utilizing Nanopore RNA004 Technology

RNA modifications are critical regulators of diverse cellular processes, yet their roles in prokaryotic mRNAs remain poorly understood. Recent advances in Oxford Nanopore sequencing--especially the RNA004 kit--have enabled higher yields, reduced signal-to-noise ratio, and improved read accuracy, making them promising tools for investigating bacterial epitranscriptomes. Here, we presented a comprehensive walkthrough for Escherichia coli RNA modification analysis based on RNA004. Using both native (WT) and in vitro-transcribed (IVT) RNA samples, we first evaluated the Dorado modification detection models ({Psi}, mA, mC, and A-to-I). While each model successfully identified known rRNA modification sites, it also generated many false positives, emphasizing the need for careful data interpretation. To address these limitations, we introduced nanoSundial (https://github.com/lrslab/nanoSundial), a new comparative method that leveraged raw current features from WT and IVT samples to detect multiple types of RNA modifications in prokaryotes. We optimized nanoSundial on well-studied rRNA sites and validated its effectiveness with tRNA modifications. Through technical and biological replicate analyses, nanoSundial demonstrated reproducibility exceeding 95% in tRNA, rRNA, and ncRNA regions, albeit with lower reproducibility ([~]61%) in mRNA. We further found enrichment of mRNA modifications at the start or end of coding sequences. In total, 190 stably modified CDS regions were identified in E. coli, many of which cluster near the end of highly expressed transcriptional units (TUs) in each operon. Overall, this study highlighted the strengths and limitations of current nanopore-based modification detection methods on bacterial RNA, introduced a robust new comparative tool, and elucidated previously uncharacterized mRNA modification landscapes. Our findings open new avenues for understanding the functional impacts of bacterial RNA modifications and advancing epitranscriptomic research in prokaryotes.

bioinformatics↗

bioRxiv2025

Engineered spermidine-secreting Saccharomyces boulardii ameliorate colitis and colon cancer in mice

Experimental studies suggest that the probiotic yeast Saccharomyces boulardii can mitigate the symptoms of inflammatory bowel disease. However, these results are equivocal and S boulardii probiotic therapy has not gained widespread acceptance in clinical practice. To assess whether the therapeutic properties of S boulardii might be improved upon, we engineered S boulardii to overproduce and secrete spermidine, a pro-regenerative natural metabolite. We employed CRISPR gene deletion and transposon-mediated gene integration to manipulate expression of key enzymes in the polyamine synthetic and transport pathways. We tested the engineered yeast by oral gavage of mice treated with azoxymethane and dextran sulfate sodium to induce chronic colitis and colon cancer. We demonstrate that oral delivery of spermidine-secreting S boulardii in mice populates the gastrointestinal tract with viable spermidine-secreting S boulardii cells and raises free spermidine levels in the gastrointestinal tract. Strikingly, spermidine-secreting S boulardii strains were significantly more effective than wild-type S boulardii in reducing dextran sulfate sodium-induced colitis as well as colitis-associated colon cancer in mice. These results suggest that in situ spermidine secretion by engineered synthetic biotic yeast strains may be an effective and low-cost therapy to mitigate inflammatory bowel disease and colon cancer.

cancer biology↗

bioRxiv2025

An improved FLARE system for recording and manipulating neuronal activity

Recording and manipulating neuronal ensembles that underlie cognition and behavior in vivo is challenging. FLARE is a light- and calcium-gated transcriptional reporting system for labeling activated neurons on the order of minutes. However, FLARE is limited by its sensitivity to prolonged neuronal activities. Here, we present an improved version of FLARE, termed cytoFLARE. cytoFLARE incorporates cytosolic expression of the transcription factor and a more sensitive pair of calcium sensing domains. We showed that cytoFLARE provides more calcium- and light-dependent signals in HEK293T cells and higher signal-to-background ratios in neuronal cultures. We further established cytoFLARE transgenic Drosophila models and applied cytoFLARE to label activated neurons upon sensory or optogenetic stimulation within a defined time window. Notably, through cytoFLARE-driven expression of an optogenetic actuator, we successfully reactivated neurons involved in the larval nociceptive system. Our findings demonstrate the first characterization and application of time-gated calcium integrators in Drosophila.

bioengineering↗

bioRxiv2025

The Dark Side of Photosymbiosis: Elaborate Repeats with TEs and Unknown ORFs in Photosymbiotic Bivalves Made Exceptionally Big Metazoan Mitogenomes

Limited truly "complete" mitogenomes have been published for marine invertebrates, with many being partial and focused on protein-coding genes. Dispelling the traditional myth of "metazoan mitogenomes being highly conserved in size and gene order", long-read technologies have revealed novel structures and complexities in animal mitogenomes. Our investigation of PacBio-assembled mitogenomes of several marine bivalve species (family Cardiidae), revealed a photosymbiotic bivalve, Fragum sueziense, possesses one of the largest circular metazoan mitogenome (92,770 bp). Additionally, mitogenomes of photosymbiotic bivalves from the genera Fragum and Tridacna ranged from 22 to 92.8 kb, surpassing the more typical sizes in non-photosymbiotic cardiids ([~]14 to 19 kb). Those expansions in mitogenomes are attributed to elaborate, species-specific repetitive sequences in the major non-coding region (NCR) which manifest an inability to assemble "complete" mitogenomes from Illumina short reads (at 150 bp) alone; divergent nature of those NCRs also hinder interspecies alignment. However, we annotated supernumerary tRNAs, transposable element fragments and open reading frames in NCRs despite their hitherto unknown functions. We postulate that NCR inflation in these photosymbiotic species may be associated with elevated reactive oxygen species in their mantle and altered immune states due to host-symbiont interactions involving photosymbionts. More complete mitogenomes are needed to uncover novel genetic elements and their functions otherwise undocumented to science. SIGNIFICANCE STATEMENTMany published animal mitogenomes does not truly span the entirety of the major non-coding region (NCR), often due to the inability of short-read sequencing to confidently cover highly repetitive sequences. In this study, long-read sequencing enabled us to access "more complete" mitogenomes for a comparative analysis focused on the nature of the NCRs between photosymbiotic and non-photosymbiotic bivalves, and the elements within them. We not only uncovered one of the largest circular metazoan mitogenomes (Fragum sueziense), but also found ample genetic elements within the conventionally "non-coding" regions, including extensive elaborate repeat patterns, transposable elements and unknown open reading frames. To our knowledge, this is the first time extensive transposable elements have been reported in animal mitogenomes. Our study revealed extreme mitogenome expansions and complexities as potential costs to bivalve-algal photosymbiosis, and provide insights into metazoan symbiosis evolution.

genomics↗

bioRxiv2025

Intracellular tension relaxation engineered through D-enantiomeric hydrogel maneuvers neurogenesis and immunomodulation to facilitate spinal cord repair

Microenvironmental mechanics regulate morphogenesis and post-injury inflammation, however, the fragile mechanical strength and oxidative physiological environment hinder precise and consistent mechanical management after spinal cord injury (SCI). Here, we engineered self-assembling hydrogels of enantiomeric peptides with neural tissue- matching mechanical properties to persistently manipulate mechanosensing and mechanotransduction through stereo conformational recognition and consequent protein affinity difference. While hindering proliferation and morphogenesis in non-neural cells, D-hydrogel-induced intracellular tension relaxation triggered neurogenesis and ECM remolding in astrocytes, while simultaneously suppressing pro-inflammation and promoting pro-regeneration in microglia, which together enable neuroprotection from degeneration and enhance functional recovery in severe SCI rat models. These effects are mediated through neurogenic morphology changes resulting from cytoskeletal tension relaxation, leading to the opening of mechanosensitive ion channels in the cellular membrane, chromatin unfolding, and YAP nuclear translocation. This exclusive D- hydrogel-dependent neurogenesis, triggered by intracellular tension relaxation, revealed a neural-specific response to mechanical cues and provided a targeted tissue repair strategy for nerve injury. TeaserIntracellular tension relaxion activates morphogenesis specifically in neural cells through reversing neurogenic cellular morphology.

neuroscience↗

bioRxiv2024

GP-ML-DC: An Ensemble Machine Learning-Based Genomic Prediction Approach with Automated Two-Phase Dimensionality Reduction via Divide-and-Conquer Techniques

Traditional machine learning (ML) and deep learning (DL) methods for genome prediction often face challenges due to the imbalance between the limited number of samples (n) and the large number of single nucleotide polymorphisms (SNPs) (p), where n is much smaller than p. To address this, we propose GP-ML-DC, an innovative genome predictor that combines traditional ML and DL models with a unique two-phase, parameter-free dimensionality reduction technique. Initially, GP-ML-DC reduces feature dimensionality by characterizing genes as features. Building on big data methodologies, it employs a divide-and-conquer approach to segment gene regions into multiple haplotypes, further decreasing dimensionality. Each haplotype segment is processed by a sub-task based on traditional ML, followed by integration via a neural network that synthesizes the results of all sub-tasks. Our experiments, conducted on four cattle milk-related traits using ten-fold cross-validation and independent testing, show that GP-ML-DC significantly surpasses current state-of-the-art genome predictors in prediction performance.

bioinformatics↗

bioRxiv2024

Methyl-CODEC enables simultaneous methylation and duplex sequencing

DNA mutations and methylation often contribute to disease development in a synergistic manner. While duplex sequencing is the most accurate method for detecting DNA mutations, it typically lacks the ability to simultaneously assess methylation or requires many reads. Here, we developed Methyl-CODEC to enable simultaneous methylation sequencing and duplex sequencing using single read pairs. To achieve this, Methyl-CODEC links an enzymatically deaminated sense strand to the reverse complement of the antisense strand, which is protected from conversion by using conversion-resistant dCTPs in the strand linking step of CODEC. Methyl-CODEC shows high concordance with standard enzymatic or bisulfite based whole genome methylation sequencing, while also uniquely preserving the original DNA sequence. This improves genetic sequencing accuracy, enables better alignment for next-generation sequencing (NGS), and distinguishes C>T mutations from unmethylated Cs. It also identifies rare mutations including those producing methylated Cs, which are enriched in CpG contexts. Methyl-CODEC opens new horizons for enhanced detection of biomarkers in cancer and molecular medicine.

genomics↗

bioRxiv2024

Parallel evolution of LRP2 gene mediating telescope-eye and celestial-eye in goldfish

After the intensive artificial selection, the development of celestial-eye in goldfish involves the protuberating and turning upwards of eyeballs, and also degeneration of the retinal. Thus, the celestial-eye goldfish provides an excellent model for both evolutionary and human ocular disease studies. Here, two mapping populations with segregating eye phenotypes in the offspring goldfish were constructed. Though whole genome sequencing using individual samples from the parents and pooled samples from the offspring, and RNA-seq for eyeball samples from pure goldfish lines, a premature stop codon in Exon 38 of LRP2 gene was identified as the top candidate mutation that is responsible for celestial-eye in goldfish. Fatty acid metabolisms and epidermal cells, especially keratocytes related functions were inhibited in the eyeballs of celestial-eye, while inflammatory reactions and extracellular matrix secreting were stimulated. These suggest the dysfunction of cornea in the celestial-eye, and same for retinal, which could be the results of the truncated LRP2 protein. Besides, evidence was provided that not all the goldfish lines share the same causal mutation for celestial-eye, while the same gene LRP2 is in charge of the similar phenotypes (celestial-eye and telescope-eye) in goldfish but no shared mutation. Therefore, those mutations and the associated phenotypes exhibit parallel evolutions in molecular level under artificial selections. Overall, the candidate mutation for celestial-eye in goldfish was identified by this study, and further analyses provide insights into the developmental and evolutionary processes of morphological changes in the eyes of goldfish. Author SummaryAs the first domesticated ornamental fish, goldfish is now worldwide spread and has undergone intensively artificial selections for morphological variations. Among them, celestial-eye is one of the most unique traits for goldfish, which consists of enlarged and protuberant eyes while the eyeballs turning upwards and thus suitable for top viewing in pots. In this study, we have identified that a single nucleotide mutation resulting in disrupted function of LRP2 gene, is responsible for the celestial-eye phenotype. Together with our transcriptome analysis, the genetic and cellular mechanisms for celestial-eye in goldfish were reported for the first time, which provides fundamental knowledge for further studies about the development of eyes in fish and also for ophthalmic diseases in humans. In the aspect of evolution, our study and previous studies reported different truncated LRP2 proteins in charge of celestial-eye and telescope-eye, respectively, which are excellent materials to understand the mechanisms of parallel evolution in molecular level, i.e., independent artificial selections resulting in mutations of the same gene.

genetics↗

bioRxiv2024

Epigenetic silencing and blockade of latency reversal by an HIV-1 encoded antisense transcript

The mechanisms that regulate human immunodeficiency virus 1 (HIV-1) latency are not fully elucidated. We reported that an HIV-1 antisense transcript (AST) induces epigenetic modifications at the HIV-1 promoter causing a closed chromatin state that suppresses viral transcription. Here we show that ectopic expression of AST in CD4+ T-cells from people with HIV-1 under antiretroviral therapy blocks latency reversal in response to pharmacologic and T-cell receptor stimulation, enforcing transcriptional silencing. We define structural domains and sequence motifs of AST contributing to its latency-promoting functions. Finally, we report an unbiased proteomics screen of AST interactors that revealed an array of previously known and potential new HIV-1-suppressive factors. Our studies identify AST as a first-in-class biological molecule capable of enforcing HIV-1 latency and with actionable curative potential.

molecular biology↗

bioRxiv2024

In vivo detection of HIV-1 antisense transcripts in untreated and ART-treated individuals

Natural antisense transcripts are expressed in eukaryotes, prokaryotes, and viruses and can possess regulatory functions at the transcriptional and/or post-transcriptional levels. In vitro studies have shown that HIV-1 antisense transcripts (AST) promote viral latency through epigenetic silencing of the proviral 5' long terminal repeat (LTR). However, expression of HIV-1 AST in vivo have not been convincingly demonstrated. Here, we used single RNA template amplification, detection, and sequencing to demonstrate expression of AST in unstimulated PBMC collected from people with HIV-1 (PWH). We found that AST had high genetic diversity that matched proviruses in cells from blood and lymph nodes. We measured a median of 26 copies of AST per 100 infected cells in PWH on ART and a median of 2 copies per 100 infected cells in PWH not on ART. The expression of HIV-1 AST in vivo is consistent with a potential regulatory role in regulation of HIV-1 expression.

microbiology↗

bioRxiv2024

HIFalpha isoform specific activities drive cell-type specificity of VHL-associated oncogenesis

Cancers arising from dysregulation of generally operative signaling pathways are often tissue specific, but the mechanisms underlying this paradox are poorly understood. Based on striking cell-type specificity, we postulated that these mechanisms must operate early in cancer development and set out to study them in a model of von Hippel Lindau (VHL) disease. Biallelic mutation of the VHL ubiquitin ligase leads to constitutive activation of hypoxia inducible factors HIF1A and HIF2A and is generally a truncal event in clear cell renal carcinoma. We used an oncogenic tagging strategy in which VHL-mutant cells are marked by tdTomato, enabling their observation, retrieval, and analysis early after VHL-inactivation. Our findings reveal markedly different consequences of HIF1A and HIF2A activation, but that both contribute to renal cell-type specific consequences of VHL-inactivation in the kidney. Early involvement of HIF2A in promoting proliferation within proximal tubular epithelium supports therapeutic targeting of HIF2A early in VHL disease.

cancer biology↗

bioRxiv2024

Evidence for Stepwise Disruption of E. coli RNA Polymerase-{lambda}R Promoter Contacts and Bubble Collapse in Transcription Initiation

In transcription initiation by E. coli RNA polymerase (RNAP), translocation of the RNA-DNA hybrid disrupts RNAP-promoter and {sigma}70-core RNAP contacts, releasing {sigma}70 and allowing RNAP to escape. Previously, to investigate whether RNAP-promoter contacts break step-by-step as the hybrid lengthens or concertedly at escape, we determined rate constants and activation energies of nucleotide-incorporation steps involving translocation at the {lambda}PR promoter. Trends in these quantities with hybrid length were inconsistent with concerted models and provided evidence for stepwise disruption of RNAP-promoter contacts and collapse of the upstream bubble (-1 to -11). Here we report kinetic m-values quantifying urea and glycine betaine (GB) effects on rate constants of individual nucleotide-incorporation steps and compare with m-values predicted from structural and model compound information. GB is predicted to favor and urea disfavor binding the initiating NTPs and trigger-helix formation in 2-mer (pppApU) synthesis, largely because of burial of phosphate oxygens in NTP binding and amide oxygens in trigger-helix formation. Consistent with these predictions, GB accelerates and urea retards steps synthesizing 3-mer and 4-mer. However, both solutes retard mid-initiation steps (5-mer to 9-mer synthesis) where -10 contacts are proposed to break, exposing DNA phosphates and allowing the upstream initiation bubble to collapse. Strikingly, urea greatly accelerates while GB retards the 11-mer synthesis step, where -35-contacts are proposed to break, {sigma}70 is released and RNAP escapes. Urea and GB kinetic m-values and activation energies of these steps are inconsistent with concerted models and support a stepwise model of contact disruption and bubble collapse in initiation.

biochemistry↗

bioRxiv2024

Single-nucleus multiple-organ chromatin accessibility mapping in the rat

The chromatin accessibility landscape is the basis of cell-specific gene expression. We generated a multiorgan, single-nucleus chromatin accessibility landscape from the model organism Rattus norvegicus. For this single-cell atlas, we constructed 25 libraries via snATAC-seq from nine organs in the rat, with a total of over 110,000 cells. Cell classification integrating gene activity scores with known marker genes identified 77 cell types, which were strongly correlated with those in published mouse single-cell transcriptome atlases. We further investigated the enrichment of cell type- and organ-specific transcription factors (TFs), the dynamics of T-cell developmental trajectories across organs, and the conservation and specificity of gene expression patterns across species. These findings provide a foundation for further investigations of the cell composition and gene regulatory networks throughout the rat body. HighlightsO_LIGeneration of a single-cell atlas of chromatin accessibility in nine organs of the rat C_LIO_LICharacterization of cell type- and organ-specific transcription factors (TFs) C_LIO_LIDynamics of chromatin accessibility in developing T cells revealed by cross-organ analysis C_LIO_LIConservation and specificity of gene expression patterns among humans, mice, and rats revealed by cross-species analysis C_LI

genomics↗

bioRxiv2024

TiFM2.0 - Versatile mechanical measurement and actuation in live embryos

During development, spatial-temporally patterned tissue-level stresses and mechanical properties create diverse tissue shapes. To understand the mechanics of small-scale embryonic tissues, precisely controlled sensors and actuators are needed. Previously, we reported a control-based approach named tissue force microscopy (TiFM1.0), which combines dynamic positioning and imaging of an inserted cantilever probe to directly measure and impose forces in early avian embryos. Here we present an upgraded system (TiFM2.0) that utilises interferometer positioning to minimise probe holder footprint, enhancing accessibility and imaging signal. This new design enables a double-probe configuration for bidirectional stretching, compression and stress propagation experiments. As proof-of-concept, we showcase a variety of examples of TiFM2.0 applications in chicken and zebrafish embryos, including the characterization of mechanical heterogeneities important for the morphogenesis of the chicken posterior body axis. We also present simplified designs and protocols for the replication of TiFM systems with minimal custom engineering for developmental biology labs.

developmental biology↗

bioRxiv2024

Comparative RT-qPCR and qPCR reveals early infection, low-titer infection, and relative cell activity of the HLB bacterium, Candidatus Liberibacter asiaticus

Candidatus Liberibacter asiaticus (Las) is one of the causal agents of citrus huanglongbing (HLB) epidemics worldwide. Due to its fastidious nature, intracellular and systemic infection, detecting Las at early and/or low-titer infection, as well as differentiating between live or dead cells in the host psyllids and citrus plants is critical for effective HLB management. To achieve both sensitive Las detection and differentiation, we employed one-step reverse transcription-quantitative PCR (RT-qPCR) using total nucleic acids as template. This method allows use of both Las 16S rRNA and rDNA as template in the same reaction and increases detection sensitivity by up to 1000-fold in comparison with quantitative PCR (qPCR). The increased sensitivity significantly reduces false negative detection and detects the otherwise undetectable low-titer Las infections. Furthermore, the greater the abundance of 16S rRNA present in the samples, the bigger the Ct gap obtained between RT-qPCR and qPCR results. The numerical Ct gap can be used to deduce relative Las cellular activity and indirectly infer whether cells are alive or dead. In addition, this comparative detection method also can be used to select inoculum and monitor relative cell activity during in vitro Las culture and evaluate the effectiveness of antimicrobial treatments against Las bacteria.

molecular biology↗

bioRxiv2024

Target-specific design of drug-like PPI inhibitors via hot-spot-guided generative deep learning

Protein-protein interactions (PPIs) are vital therapeutic targets. However, the large and flat PPI interfaces pose challenges for the development of small-molecule inhibitors. Traditional computer-aided drug design approaches typically rely on pre-existing libraries or expert knowledge, limiting the exploration of novel chemical spaces needed for effective PPI inhibition. To overcome these limitations, we introduce Hot2Mol, a deep learning framework for the de novo design of drug-like, target-specific PPI inhibitors. Hot2Mol generates small molecules by mimicking the pharmacophoric features of hot-spot residues, enabling precise targeting of PPI interfaces without the need for bioactive ligands. The framework integrates three key components: a conditional transformer for pharmacophore-guided, drug-likeness-constrained molecular generation; an E(n)-equivariant graph neural network for accurate alignment with PPI hot-spot pharmacophores; and a variational autoencoder for generating novel and diverse molecular structures. Experimental evaluations demonstrate that Hot2Mol outperforms baseline models across multiple metrics, including docking affinities, drug-likenesses, synthetic accessibility, validity, uniqueness, and novelty. Furthermore, molecular dynamics simulations confirm the good binding stability of the generated molecules. Case studies underscore Hot2Mols ability to design high-affinity and selective PPI inhibitors, demonstrating its potential to accelerate rational PPI drug discovery.

bioinformatics↗

bioRxiv2024

Systemic inflammation and lymphocyte activation precede rheumatoid arthritis

Some autoimmune diseases, including rheumatoid arthritis (RA), are preceded by a critical subclinical phase of disease activity. Proactive clinical management is hampered by a lack of biological understanding of this subclinical at-risk state and the changes underlying disease development. In a cross-sectional and longitudinal multi-omics study of peripheral immunity in the autoantibody-positive at-risk for RA period, we identified systemic inflammation, proinflammatory-skewed B cells, expanded Tfh17-like cells, epigenetic bias in naive T cells, TNF+IL1B+ monocytes resembling a synovial macrophage population, and CD4 T cell transcriptional features resembling those suppressed by abatacept (CTLA4-Ig) in RA patients. Our findings characterize pathogenesis prior to clinical diagnosis and suggest the at-risk state exhibits substantial immune alterations that could potentially be targeted for early intervention to delay or prevent autoimmunity. We provide a suite of tools at https://apps.allenimmunology.org/aifi/insights/ra-progression/ to facilitate exploration and enhance accessibility of this extensive dataset. One Sentence SummaryACPA+ at-risk individuals show RA-like inflammation and multi-compartment immune dysregulation during transition to clinically active RA

immunology↗

bioRxiv2024

PhenoScreen: A Dual-Space Contrastive Learning Framework-based Phenotypic Screening Method by Linking Chemical Perturbations to Cellular Morphology

Phenotypic drug discovery (PDD) screens compounds in cellular models that represent disease-relevant phenotypes, offering a compelling alternative to traditional target-based approaches. Unlike conventional methods, where compounds act on a single predefined target, PDD identifies compounds capable of exerting therapeutic effects through multiple targets and mechanisms. This makes PDD particularly valuable for discovering first-in-class drugs, especially for diseases with poorly understood molecular mechanisms or those lacking validated therapeutic targets. By enabling broader exploration of biological systems and uncovering multi-target drugs (polypharmacology), PDD provides a powerful strategy for tackling complex diseases. In this study, we introduce PhenoScreen, a deep learning framework designed to advance PDD by utilizing large-scale compound-phenotype association data. Through contrastive learning, PhenoScreen connects chemical space with cellular morphological profiles, allowing for accurate prediction of compound-induced phenotypic changes. PhenoScreen can also accurately identfiy lead compounds which could induce user defined phenotypic shift but more novel scaffolds using different levels of phenotypic information reflected by diverse compounds. The model was validated across multiple screening tasks and successfully predicted active compounds inducing user-specified phenotypes with varying inhibitory effects in the osteosarcoma phenotypic model. Further, other than showing effectiveness to osteosarcoma, our experiments also showed that PhenoScreen demonstrated strong generalization to rhabdomyosarcoma, and the active compound we screened had an IC50 of up to 1.842 M, suggesting its ability to capture key phenotypic features shared across cancer cells. These results underscore PhenoScreens potential to accelerate drug discovery by identifying novel therapeutic pathways and increasing the diversity of viable drug candidates. PhenoScreen is accessible online via our groups web server for compound virtual screening at https://bailab.siais.shanghaitech.edu.cn/services/PhenoScreen/, and the source codes are available at https://github.com/Shihang-Wang-58/PhenoScreen.

bioinformatics↗