Biology subjects

Zheng, G.

Publications and source records attributed to Zheng, G..

At least 37 records · Page 2Linked to original sources

bioRxiv2025

Integrating multi-modal transcriptomics identifies cellular subtypes with distinct roles in PDAC progression

BackgroundPancreatic ductal adenocarcinoma (PDAC) features a complex tumor microenvironment (TME) that significantly influences patient outcomes. Understanding the TMEs cellular composition and interactions is crucial for identifying therapeutic targets to improve treatment. MethodsWe performed an integrative analysis combining 88 single-cell RNA sequencing (scRNA-seq) samples with 187,520 cells, 20 Visium spatial transcriptomics (ST-seq) samples with 67,933 spots, and 1383 bulk RNA-seq samples and 2 Xenium high-resolution ST-seq samples with 307,679 cells, to delineate and characterize distinct subpopulations of fibroblasts, macrophages, T/NK cells, and B/plasma cells within the PDAC microenvironment. Correlations among major cell types across 12 PDAC datasets were assessed through gene set variation analysis (GSVA). Kaplan-Meier survival analysis was used to evaluate the prognostic significance of cell subtypes. Furthermore, multiplex immunohistochemistry (mIHC) and the Xenium platform was utilized to validate cellular interactions in human PDAC tissues at both protein and RNA levels. ResultsSix fibroblast subtypes and eight macrophage/monocyte subtypes were identified. POSTN high fibroblasts and SPP1 high macrophages highly infiltrated tumor tissues and were associated with poor prognosis. Most immune cell subtypes mediate adverse prognoses, except for CCL4 high CD8+ T EFF cells and IGHG1 high plasma cells, which are linked to favorable outcomes. ST-seq revealed spatial colocalization of POSTN high Fibro and SPP1 high Macro cells, as well as CCL4 high CD8+ T EFF and IGHG1 high plasma cells. These findings were corroborated by mIHC and validated using Xenium spatial transcriptomics with single-cell resolution, which confirmed the expression and spatial proximity of these markers at both the gene and protein levels. ConclusionThis integrative analysis of the PDAC TME underscores the prognostic importance of POSTN high fibroblasts and SPP1 high macrophages, while also highlighting the protective roles of CCL4 high CD8+ T EFF cells and IGHG1 high plasma cells. These insights could improve PDAC treatment strategies.

cancer biology↗

bioRxiv2025

Computational Fluid Dynamics (CFD) Modeling of Ilamycin E Production in Streptomyces atratus SCSIO ZH16 Submerged Fermentation

A computational fluid dynamics (CFD) model was developed and validated against experiments for a laboratory-scale 5-L bioreactor. Numerical simulation was performed to describe the bioreaction of Streptomyces atratus SCSIO ZH16 fermentation for ilamycin E production with the dynamic changes in viscosity of the fermentation broth due to biomass growth and decay. This model can account for the two-way coupling between the fermentation environment and medium, which allowed for the elucidation of the impact of the flow field in the bioreactor on bacterial growth and production, as well as the influence of viscosity changes on the flow field. This work represents the first integration of Streptomyces fermentation with CFD, enabling the simulation of flow field and mass transfer under varying stirring speed, aeration rate, and viscosity during Streptomyces fermentation. An optimum range of fermentation broth viscosity (10-30 mPa s) was identified for ilamycin E production by S. atratus SCSIO ZH16 fermentation. Furthermore, the addition of sorbitol to optimize the viscosity of the fermentation broth in the later stages of fermentation. The enhanced mass transfer efficiency strengthens the respiration, energy supply, and carbon source consumption in Streptomyces, thereby increasing the ilamycin E production. This research offers a practical strategy for the process intensification and industrial scale-up for such bioreactors. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=78 SRC="FIGDIR/small/630827v1_ufig1.gif" ALT="Figure 1"> View larger version (27K): org.highwire.dtl.DTLVardef@3d463corg.highwire.dtl.DTLVardef@ce3859org.highwire.dtl.DTLVardef@d55c7forg.highwire.dtl.DTLVardef@1abaee5_HPS_FORMAT_FIGEXP M_FIG C_FIG

bioengineering↗

bioRxiv2024

Targeted inhibition of Bcl-xL following radiation reduces tumourigenesis in preclinical models of H3K27M-altered diffuse midline glioma

BackgroundDiffuse midline gliomas (DMGs) with histone H3K27M mutations represent a devastating paediatric brain cancer characterized by abysmal prognosis and limited treatment options. The only approved treatment is radiotherapy (RT), but most of the tumours relapse with fatal consequences. In this study, we sought to investigate whether irradiation leads to senescence induction and explore the efficacy of senolytics against DMG. MethodsWe have characterised the senescent phenotype of five genetically heterogeneous H3K27M-altered human DMG cell lines, combining cellular and/or molecular approaches. The sensitivity of senescent cells to Bcl-xL inhibition has been demonstrated in dose/response curves in vitro and in a PDX model of DMG. ResultsHere, we show that ionizing radiation induces senescence and SASP responses in both TP53 mutant and wild-type H3K27M-altered human DMG cell lines. We identify Navitoclax as a potent senolytic agent that selectively targets senescent DMG cells into apoptosis by inhibiting Bcl-xL. Related compounds, such as a proteolysis-targeting chimera (PROTAC)-mediated Bcl-xL degradation and a galacto-conjugated form of Navitoclax also show an effective senolytic activity in senescent cancer cells. Finally, we show that a combination therapy of irradiation and Navitoclax results in reduced tumor burden and increased mouse survival in an orthotopic xenograft DMG model. ConclusionThese results offer a rationale for further clinical development of senolytic therapies as part of multimodal treatment approaches for DMG patients/ Key PointsO_LIIonising irradiation induces senescence in human DMG cells independently of the p53 status. C_LIO_LIBcl-xL inhibition results in apoptosis of human DMG senescent cells in synergy with irradiation. C_LIO_LICombination of irradiation and BcL-xL inhibition reduces tumourigenesis in a PDX model of DMG. C_LI Importance of the StudyH3K27M-altered DMG are devastating paediatric tumours with an abysmal prognosis. The only approved treatment is radiotherapy but this is palliative and tumours almost always relapse with fatal consequences for the patients. In this study, we show that radiotherapy results in senescence induction in five genomically heterogeneous human DMG cell lines. We identify that drugs targeting the anti-apoptotic protein Bcl-xL show a strong senolytic activity in conjunction with radiotherapy both in vitro in DMG cells and in vivo in a PDX model of H3K27M-altered DMG. Treatment with Bcl-xL inhibitor Navitoclax, or related compounds targeting Bcl-xL protein degradation or containing a galactose conjugated form of Navitoclax results in DMG cancer cell apoptosis. As several of these inhibitors are currently being tested in ongoing clinical trials against other diseases, our data support the use of Bcl-xL inhibition mediated senolytics as an adjuvant therapy to radiotherapy to potentially improve outcomes in this challenging disease setting.

cancer biology↗

bioRxiv2024

DNA StairLoop: Achieving High Error-correcting and Parallel-processing Capabilities in DNA-based Data Storage

DNA-based data storage is a promising solution to the challenges of large-scale data storage. However, the low throughput of the mainstream inkjet-based DNA synthesis method has hindered its widespread adoption. In contrast, high-throughput electrochemical synthesis provides higher throughput but with more nucleotide insertion, deletion, and substitution errors. Here, we propose an innovative coding scheme with high error correction capabilities, called DNA StairLoop. This coding scheme features a staircase interleaver and allows for component codes such as the convolutional code and the ow-density Parity-check code, allowing for flexible adaptation of the coding scheme. Both the row and the column decoders are soft input and soft output, enabling further improvement in data recovery accuracy through iterative decoding. The staircase interleaver facilitates extensive parallel decoding capabilities while effectively preserving parallelism across a multitude of nodes. In the in vitro experiment, DNA StairLoop successfully recovered the raw information in a staircase block with a nucleotide error rate of more than 8%. The simulations revealed that the DNA StairLoop can correct 10% nucleotide errors. Moreover, in parallel computing processing, the decoding time of our code continues to decrease dramatically.

synthetic biology↗

bioRxiv2024

BEDMS: A metadata standardizer for genomic regionattributes

High-throughput sequencing technologies have generated vast omics data annotating genomic regions. A challenge arises in integrating this data because the associated metadata does not follow a uniform schema. This hinders data management, discovery, interoperability, and reusability. Existing tools that address metadata standardization issues are generally limited in scope and targeted toward specific data sets or types and are not generally applicable to custom schemas. To improve standardization of genomic interval metadata, we have developed BEDMS. We developed and evaluated several model architectures and trained models that achieved high performance on held-out training data. With a trained model, BEDMS provides users with predicted standardized metadata attributes that follow a standardized schema. Furthermore, BEDMS provides the ability to train custom models. To demonstrate, we trained BEDMS on three different schemas, allowing users to choose which schema to standardize into. We also deployed BEDMS on PEPhub, which provides a graphical user interface to allow users to standardize metadata without requiring any local training or software at all. In conclusion, BEDMS offers a practical one-stop solution for metadata management and standardization for genomic interval data.

genomics↗

bioRxiv2024

Proteolysis targeting chimera extracellular vesicles for therapeutic development treating triple negative breast cancer

Proteolysis targeting chimeras (PROTACs) are an emerging targeted cancer therapy approach, but wide-spread clinical use of PROTAC is limited due to poor cell targeting and penetration, and instability in vivo. To overcome such issues and enhance the in vivo efficacy of PROTAC drugs, microfluidic droplet-based electroporation ({micro}DES) was developed as a novel extracellular vesicle (EVs) transfection system, which enables the high-efficient PROTAC loading and effective delivery in vivo. Our previously developed YX968 PROTAC drug had shown the selectively degradation of HDAC3 and 8, which effectively suppresses the growth of breast tumor cell lines, including MDA-MB-231 triple negative breast cancer (TNBC) line, via dual degradation without provoking a global histone hyperacetylation. In this study, we demonstrated that {micro}DES-based PROTAC loading in EVs significantly enhanced therapeutic function of PROTAC drug in vivo in the TNBC breast tumor mouse model. NSG mice with pre-established MDA-MB-231 tumors and treated with intraperitoneal injection of EVs for tumor inhibition study, which showed significantly higher HDAC 3 and 8 degradation efficiency and tumor inhibition than PROTAC only group. The liver, spleen, kidney, lung, heart, and brain were collected for safety testing, which exhibited improved toxicity. The EV delivery of PROTAC drug enhances drug stability and bioavailability in vivo, transportability, and drug targeting ability, which fills an important gap in current development of PROTAC therapeutic functionality in vivo and clinical translation. This novel EV-based drug transfection and delivery strategy could be applicable to various therapeutics for enhancing in vivo delivery, efficacy, and safety.

bioengineering↗

bioRxiv2024

Wnt signaling dosage controlled by a Ctnnb1 enhancer balances homeostasis and tumorigenesis of intestinal epithelia

The {beta}-catenin-dependent canonical Wnt signaling is pivotal in organ development, tissue homeostasis, and cancer. Here we identified an upstream enhancer of Ctnnb1 - the coding gene for {beta}-catenin, named ieCtnnb1 (intestinal enhancer of Ctnnb1), which is crucial for intestinal homeostasis. ieCtnnb1 is predominantly active in the base of small intestinal crypts and throughout the epithelia of large intestine. Knockout of ieCtnnb1 led to a reduction in Ctnnb1 transcription, compromising the canonical Wnt signaling in intestinal crypts. Single-cell sequencing revealed that ieCtnnb1 knockout altered epithelial compositions and potentially compromised functions of small intestinal crypts. While deletion of ieCtnnb1 hampered epithelial turnovers in physiologic conditions, it prevented occurrence and progression of Wnt/{beta}-catenin-driven colorectal cancers. Human ieCTNNB1 drove reporter gene expression in a pattern highly similar to mouse ieCtnnb1. ieCTNNB1 contains a single-nucleotide polymorphism associated with CTNNB1 expression levels in human gastrointestinal epithelia. The enhancer activity of ieCTNNB1 in colorectal cancer tissues was stronger than that in adjacent normal tissues. HNF4 and phosphorylated CREB1 were identified as key trans-factors binding to ieCTNNB1 and regulating CTNNB1 transcription. Together, these findings unveil an enhancer-dependent mechanism controlling the dosage of Wnt signaling and homeostasis in intestinal epithelia.

cell biology↗

bioRxiv2024

A step-emulsion based digital-RPA for Pathogenic bacteria detection

Foodborne bacteria are major causes that affect human health. Development of new method that could achieve Rapid, sensitive and quantitative detection of pathogen is urgently needed. In this research, a step emulsion microfluidic,combined with droplet-based digital-RPA, was developed to detect Vibrio parahaemolyticus, a major seafood-borne pathogenic bacteria. Specific and rapid detection of Vibrio parahaemolyticus in 30 min has been achieved by this new device, with a detection limit of 10 CFU/L, about 10-times lower than classical tube-based RPA. This device was demonstrated to be a promising tool for detection of pathogenic bacteria.

microbiology↗

bioRxiv2024

EAP: a versatile cloud-based platform for comprehensive and interactive analysis of large-scale ChIP/ATAC-seq data sets

Epigenome profiling techniques such as ChIP-seq and ATAC-seq have revolutionized our understanding of gene expression regulation in tissue development and disease progression. However, increasing amount of ChIP/ATAC-seq data poses challenges in computational resources, and the absence of systematic tools for epigenomic analysis underscores the necessity for an efficient analysis platform. To address these issues, we developed the Epigenomic Analysis Platform (EAP, https://www.biosino.org/epigenetics), a scalable cloud-based tool that efficiently analyzes large-scale ChIP/ATAC-seq data sets. EAP employs advanced computational algorithms to derive biologically meaningful insights from heterogeneous datasets and automatically generates publication-ready figures and tabular results, enabling comprehensive epigenomic analysis and data mining in areas like cancer subtyping and therapeutic target discovery.

bioinformatics↗

bioRxiv2024

PROTAC-mediated dual degradation of BCL-xL and BCL-2 is a highly effective therapeutic strategy in small-cell lung cancer

BCL-xL and BCL-2 are validated therapeutic targets in small-cell lung cancer (SCLC). Targeting these proteins with navitoclax (formerly ABT263, a dual BCL-xL/2 inhibitor) induces dose-limiting thrombocytopenia through on-target BCL-xL inhibition in platelets. Therefore, platelet toxicity poses a barrier in advancing the clinical translation of navitoclax. We have developed a strategy to selectively target BCL-xL in tumors, while sparing platelets, by utilizing proteolysis-targeting chimeras (PROTACs) that hijack the cellular ubiquitin proteasome system for target ubiquitination and subsequent degradation. In our previous study, the first-in-class BCL-xL PROTAC, called DT2216, was shown to have synergistic antitumor activities when combined with venetoclax (formerly ABT199, BCL-2-selective inhibitor) in a BCL-xL/2 co-dependent SCLC cell line, NCI-H146 (hereafter referred to as H146), in vitro and in a xenograft model. Guided by these findings, we evaluated our newly developed BCL-xL/2 dual degrader, called 753b, in three BCL-xL/2 co-dependent SCLC cell lines and the H146 xenograft models. 753b was found to degrade both BCL-xL and BCL-2 in these cell lines. Importantly, it was considerably more potent than DT2216, navitoclax, or DT2216+venetoclax to reduce the viability of BCL-xL/2 co-dependent SCLC cell lines in cell culture. In vivo, 5 mg/kg weekly dosing of 753b leads to significant tumor growth delay similar to the DT2216+venetoclax combination in H146 xenografts by degrading both BCL-xL and BCL-2. Additionally, 753b administration at 5 mg/kg every four days induced tumor regressions. 753b at this dosage was well tolerated in mice without induction of severe thrombocytopenia as seen with navitoclax nor induced significant changes in mouse body weights. These results suggest that the BCL-xL/2 dual degrader could be an effective and safe therapeutic for a subset of SCLC patients warranting clinical trials in future.

cancer biology↗

bioRxiv2024

Multifunctional biomimetic porphyrin-lipid nanoparticles - novel nanoscale theranostics for atherosclerotic cardiovascular disease

BackgroundHigh-density lipoprotein (HDL) nanoagents have unrealized potential for atherosclerosis theranostics. Porphyrin-lipid HDL mimetic nanoparticles (Por-HDL-NPs) incorporate porphyrin-lipid which permits near infrared fluorescence imaging and positron emission tomography (PET) through chelation of Copper-64 (64Cu). The outer shell contains apolipoprotein A-I mimetic peptide R4F that interacts with scavenger receptor SR-BI, enabling macrophage targeting and therapeutic effects. We leveraged the theranostic properties of Por-HDL-NPs for testing in atherosclerosis. Methods and ResultsIn vitro, Por-HDL-NPs were internalised by immortalised bone marrow-derived macrophages (iBMDMs), visualised via fluorescence microscopy and flow cytometry. Por-HDL-NPs increased cholesterol efflux from [3H]-cholesterol-loaded iBMDMs, (49%, P<0.05), compared to reconstituted HDL. Incubation of iBMDMs with Por-HDL-NPs reduced mRNA levels of inflammatory mediators Il-1{beta} (88%), Il-18 (54%) and Ccl5 (75%), and protein secretion of IL-1{beta} (69%) and CCL5 (82%), P<0.05. Por-HDL-NPs suppressed inflammasome components Nlrp3 (69%) and Asc (36%), P<0.05. Studies using siRNA deletion of SR-B1 and methyl-{beta}-cyclodextrin, revealed the anti-inflammatory properties of Por-HDL-NPs were independent of SR-B1 and cholesterol efflux. However, Por-HDL-NPs suppressed activation of inflammatory transcription factor NF-{kappa}B (53%, P<0.05). In Apoe-/- mice, PET imaging showed 64Cu-Por-HDL-NPs localised in hearts and detected increases in plaque over time with high-cholesterol diet. Por-HDL-NP fluorescence was visualised in aortic sinus plaques, co-localised with CD68+ macrophages, and by fluorescence IVIS imaging in aortic arch plaque. Por-HDL-NP-treated mice had smaller early-stage (22%) and unstable plaques (52%) and fewer circulating monocytes (32%) than control PBS-treated mice, P<0.05 for all. ConclusionsPor-HDL-NPs have theranostic properties, exhibiting both multi-modal imaging capabilities for identifying plaque and athero-protective therapeutic effects. Clinical PerspectiveO_ST_ABSWhat is new?C_ST_ABSO_LIPorphyrin high-density lipoprotein (HDL) mimetic nanoparticles (Por-HDL-NPs) have theranostic application in atherosclerosis. C_LIO_LIPor-HDL-NPs are internalized by macrophages in vitro and plaque macrophages in vivo, enabling the visualization of atherosclerosis by both positron emission tomography and multiple fluorescence imaging modalities. C_LIO_LIPor-HDL-NPs exhibit atheroprotective effects and suppress inflammation, promote cholesterol efflux, reduce atherosclerotic plaque development and lower the number of circulating monocytes. C_LI What are the clinical implications?O_LIThe PET imaging and plaque targeting capabilities of Por-HDL-NPs have implications for improved non-invasive tracking of human atherosclerosis development. C_LIO_LIThe excellent fluorescence imaging and plaque targeting properties of Por-HDL-NPs have clinical significance for improved detection of early-stage plaque using intravascular imaging strategies. C_LIO_LIPor-HDL-NPs provide therapeutic capabilities that target plaque directly, independent of lipid-lowering, suggestive of their potential to provide benefit on top of current lipid-lowering strategies. C_LI

cell biology↗

bioRxiv2024

MobiChIP: a compatible library construction method of single-cell ChIP-seq based droplets

In order to illustrate the epigenetic heterogeneity, versatile tools of single-cell ChIP-seq (scChIP-seq) are necessary to meet the convenience and accuracy. Here, we develop MobiChIP, a compatible ChIP-seq library construction method based current sequencing platform with single cell level. As a novel capture strategy, MobiChIP is efficient to capture the fragments from tagmented nuclei of numerous species and execute the mixing of samples from different tissues or species. Especially, this strategy enables the flexible sequencing manipulation and sufficient nucleosome amplification without customized sequencing primers. MobiChIP reveals the landscape of chromatin regulation regions with active(H3K27ac) and repressive(H3K27me3) histone modification markers in peripheral blood mononuclear cells (PBMCs), and accurately unveiled the epigenetic repression of hox gene cluster in PBMCs than ATAC-seq. Meanwhile, we complete the bioinformatics pipeline to integrates the scChIP-seq data and scRNA-seq to illustrate the cellular epigenetic and genetic heterogeneity. One-Sentence SummaryA high-throughput single-cell ChIP-seq based droplet reveals the integration of scRNA-seq data and scChIP-seq data.

molecular biology↗

bioRxiv2023

EryDB: a transcriptomic profile database for erythropoiesis and erythroid-related diseases

Erythropoiesis is a finely regulated and complex process that involves multiple transformations from hematopoietic stem cells to mature red blood cells at hematopoietic sites from the embryonic to the adult stages. Investigations into its molecular mechanisms have generated a wealth of expression data, including bulk and single-cell RNA sequencing data. A comprehensively integrated and well-curated erythropoiesis-specific database will greatly facilitate the mining of gene expression data and enable large-scale research of erythropoiesis and erythroid related diseases. Here, we present EryDB, an open-access and comprehensive database dedicated to the collection, integration, analysis, and visualization of transcriptomic data for erythropoiesis and erythroid-related diseases. Currently, the database includes expertly curated quality-assured data of 3,803 samples and 1,187,119 single cells derived from 107 public studies of three species (Homo sapiens, Mus musculus, and Danio rerio), nine tissue types, and five diseases. EryDB provides users with the ability to not only browse the molecular features of erythropoiesis between tissues and species, but also perform computational analyses of single-cell and bulk RNA sequencing data, thus serving as a convenient platform for customized queries and analyses. EryDB v1.0 is freely accessible at https://ngdc.cncb.ac.cn/EryDB/home.

bioinformatics↗

bioRxiv2023

Genome assembly and genomic architecture of a prominent cold-resistant rapeseed germplasm

Graphical AbstractThe genomic landscape of cold-tolerant winter rapeseed (Brassica napus, L) has been poorly characterized. We assembled a high-quality reference genome of a prominent cold-tolerant winter rapeseed cultivar, NTS57, and performed phylogenetic and pan-genomic analyses by integrating other reported Brassica napus accessions. The transcriptome analysis revealed that microtubule-associated biological pathways were much more active in NTS57 under cold stress than in the cold-sensitive variety. Whole genome methylation data analysis revealed that DNA demethylation on protein-coding genes and repetitive elements, especially at CHH sites, is essential for cold response in winter rapeseed.

genomics↗

bioRxiv2023

Pindel-TD: a tandem duplication detector based on a pattern growth approach

Tandem duplication (TD) is a major type of structural variation (SV), and plays an important role in novel gene formation and human diseases. However, TDs are often missed or incorrectly classified as insertions by most of modern SV detection methods due to the lacking of specialized operation on TD related mutational signals. Herein, we developed a TD detection module of Pindel referred as Pindel-TD based on a TD specific pattern growth approach. Pindel-TD detects TDs with a wide size range at single nucleotide resolution. Using simulation and real read data of HG002, we demonstrate that Pindel-TD outperformed other leading methods in terms of precision, recall, F1-score and robustness. Further applying Pindel-TD on data generated from K562 cancer cell line, we identified a TD located at the seventh exon of SAGE1, explaining its high expression. Pindel-TD is available at https://github.com/xjtu-omics/pindel and free for non-commercial use.

bioinformatics↗

bioRxiv2023

Methods for evaluating unsupervised vector representations of genomic regions

Representation learning models have become a mainstay of modern genomics. These models are trained to yield vector representations, or embeddings, of various biological entities, such as cells, genes, individuals, or genomic regions. Recent applications of unsupervised embedding approaches have been shown to learn relationships among genomic regions that define functional elements in a genome. Unsupervised representation learning of genomic regions is free of the supervision from curated metadata and can condense rich biological knowledge from publicly available data to region embeddings. However, there exists no method for evaluating the quality of these embeddings in the absence of metadata, making it difficult to assess the reliability of analyses based on the embeddings, and to tune model training to yield optimal results. To bridge this gap, we propose four evaluation metrics: the cluster tendency score (CTS), the reconstruction score (RCS), the genome distance scaling score (GDSS), and the neighborhood preserving score (NPS). The CTS and RCS statistically quantify how well region embeddings can be clustered and how well the embeddings preserve information in training data. The GDSS and NPS exploit the biological tendency of regions close in genomic space to have similar biological functions; they measure how much such information is captured by individual region embeddings in a set. We demonstrate the utility of these statistical and biological scores for evaluating unsupervised genomic region embeddings and provide guidelines for learning reliable embeddings. AvailabilityCode is available at https://github.com/databio/geniml

bioinformatics↗

bioRxiv2023

Joint representation learning for retrieval and annotation of genomic interval sets

MotivationAs available genomic interval data increases in scale, we require fast systems to search it. A common approach is simple string matching to compare a search term to metadata, but this is limited by incomplete or inaccurate annotations. An alternative is to compare data directly through genomic region overlap analysis, but these approaches lead to challenges like sparsity, high dimensionality, and computational expense. We require novel methods to quickly and flexibly query large, messy genomic interval databases. ResultsHere, we develop a genomic interval search system using representation learning. We train numerical embeddings for a collection of region sets simultaneously with their metadata labels, capturing similarity between region sets and their metadata in a low-dimensional space. Using these learned co-embeddings, we develop a system that solves three related information retrieval tasks using embedding distance computations: retrieving region sets related to a user query string; suggesting new labels for database region sets; and retrieving database region sets similar to a query region set. We evaluate these use cases and show that jointly learned representations of region sets and metadata are a promising approach for fast, flexible, and accurate genomic region information retrieval.

bioinformatics↗

bioRxiv2023

Unleashing the Power of NR4A1 Degradation as a Novel Strategy for Cancer Immunotherapy

An effective cancer therapy requires both killing cancer cells and targeting tumor-promoting pathways or cell populations within the tumor microenvironment (TME). We purposely search for molecules that are critical for multiple tumor-promoting cell types and identified nuclear receptor subfamily 4 group A member 1 (NR4A1) as one such molecule. NR4A1 has been shown to promote the aggressiveness of cancer cells and maintain the immune suppressive TME. Using genetic and pharmacological approaches, we establish NR4A1 as a valid therapeutic target for cancer therapy. Importantly, we have developed the first-of-its kind proteolysis-targeting chimera (PROTAC, named NR-V04) against NR4A1. NR-V04 effectively degrades NR4A1 within hours of treatment in vitro and sustains for at least 4 days in vivo, exhibiting long-lasting NR4A1-degradation in tumors and an excellent safety profile. NR-V04 leads to robust tumor inhibition and sometimes eradication of established melanoma tumors. At the mechanistic level, we have identified an unexpected novel mechanism via significant induction of tumor-infiltrating (TI) B cells as well as an inhibition of monocytic myeloid derived suppressor cells (m-MDSC), two clinically relevant immune cell populations in human melanomas. Overall, NR-V04-mediated NR4A1 degradation holds promise for enhancing anti- cancer immune responses and offers a new avenue for treating various types of cancer.

cancer biology↗