Abundance Of Cell Types Research Articles

P53 Guselkumab binding to CD64+ IL-23–producing myeloid cells enhances potency for neutralizing IL-23 signalling

Abstract IL-23 is implicated in the pathogenesis of psoriasis (PsO), and myeloid cells expressing FcγRI (CD64) are the primary source of IL-23 in lesional PsO skin tissue. The incidence and prevalence of psoriatic arthritis increases with PsO severity, and joint disease activity is positively correlated with frequency of peripheral CD64+ monocytes. We evaluated: CD64 and IL-23 expression in PsO patient skin biopsies; binding of anti-IL-23p19 subunit IgG1 monoclonal antibodies (mAbs) guselkumab [(GUS) fully human, native Fc-region] and risankizumab [(RZB) humanized, mutated Fc-region] to CD64; and functional consequences of CD64 binding by anti-IL-23p19 subunit mAbs in in vitro assays. CD64, IL-23p19 and IL-23p40 subunits mRNA transcripts were analysed from bulk and single-cell RNAseq datasets. Binding of mAbs to IFNγ-primed human monocytes and IL-23–secreting inflammatory monocytes, and capture of endogenously secreted IL-23, were evaluated by flow cytometry. Internalization of IL-23 and GUS/RZB within CD64+ macrophages was assessed with live-cell confocal imaging. GUS and RZB potency for inhibiting IL-23 was determined in a co-culture of THP-1 cells (CD64+ monocyte cell line activated to produce IL-23) and an IL-23 reporter cell line (measuring biologically active IL-23). IL-23p19 mRNA transcript expression in the co-culture was measured by qPCR. Expression of CD64, IL-23p19, and IL-23p40 mRNA transcripts was increased in lesional vs. non-lesional PsO skin, as were the abundance of myeloid cell types co-expressing CD64/IL-23p19 mRNA transcripts. In in vitro assays, GUS (not RZB) demonstrated Fc-mediated binding to CD64 on IFNγ-primed monocytes. Moreover, CD64-bound GUS simultaneously captured IL-23 secreted from the same cells. GUS (not RZB) bound to the CD64+ macrophages and mediated IL-23 internalization to low pH intracellular compartments. GUS and RZB equipotently inhibited signalling by IL-23 present in THP-1-conditioned medium. However, in a co-culture of IL-23–producing THP-1 cells with an IL-23–responsive reporter cell line, GUS demonstrated enhanced potency vs. RZB for inhibiting IL-23 signalling. GUS did not alter IL-23p19 mRNA transcript expression in the co-culture. These findings were consistent with previous observations of CD64+ myeloid cells as a key source of IL-23 in lesional PsO skin tissue. GUS binding to CD64 on IL-23-producing cells likely contributed to the enhanced potency of GUS vs. RZB in inhibiting IL-23 signalling in the co-culture assay. These in-vitro data support a hypothesis for optimal localization of GUS in inflamed tissues, where CD64+ IL-23-producing myeloid cells are increased and in proximity to IL-23-responsive lymphoid cells, enhancing GUS neutralization of IL-23 at its source of production.

Chrysalis: decoding tissue compartments in spatial transcriptomics with archetypal analysis

Dissecting tissue compartments in spatial transcriptomics (ST) remains challenging due to limited spatial resolution and dependence on single-cell reference data. We present Chrysalis, a computational method that rapidly uncovers tissue compartments through spatially variable gene (SVG) detection and archetypal analysis without requiring external reference data. Additionally, it offers a unique visualisation approach for swift tissue characterisation and provides access to the underlying gene expression signatures, enabling the identification of spatially and functionally distinct cellular niches. Chrysalis was evaluated through various benchmarks and validated against deconvolution, independently obtained cell type abundance data, and histopathological annotations, demonstrating superior performance compared to other algorithms on both in silico and real-world test examples. Furthermore, we showcased its versatility across different technologies, such as Visium, Visium HD, Slide-seq, and Stereo-seq.

EPCO-15. INTRA-TUMORAL HETEROGENEITY OF DNA METHYLATION PROFILING AND CELLULAR COMPOSITION IN GLIOBLASTOMA

Abstract BACKGROUND DNA methylation profiling has become an important diagnostic and exploratory tool in neuro-oncology. Intra-tumoral molecular heterogeneity is a crucial factor in the treatment resistance of IDH-wildtype glioblastoma (GBM). However, the intra-tumoral heterogeneity of methylation profiles in GBM remains poorly elucidated. This study aimed to investigate the intra-tumoral heterogeneity of DNA methylation subclasses and deconvolve methylation profiles to infer the cellular composition of the tumor. METHODS We conducted genome-wide DNA methylation analysis and applied the DKFZ/Heidelberg CNS tumor classifier to FFPE tissue samples from a multi-sampling cohort with multiple samples per patient and a single-sampling cohort with one sample per patient. Only samples classified as GBM were included. We conducted two different methylation-based deconvolution analyses. Deconvolution 1 inferred the fractions of tumor subtypes (RTK_I, RTK_II, MES_TYP, and MES_ATYP) and cell types in the microenvironment. Deconvolution 2 estimated the abundance of malignant cell states (stem-like and differentiated cell components) and cell types in the microenvironment. RESULTS The multi-sampling cohort included 32 patients with 113 samples, and the single-sampling cohort included 87 patients with 87 samples. In the multi-sampling cohort, 8 patients (25%) exhibited intra-patient heterogeneity in methylation subclasses across their samples. Deconvolution 1 indicated that each tumor subclass component was heterogeneously distributed within samples, and the abundance of the myeloid cell component correlated with the MES_TYP component across samples. Deconvolution 2 similarly revealed heterogeneity in the immune component abundance across samples within patients. Integrating the results of Deconvolution 1 and Deconvolution 2 using the entire cohort revealed significant correlations between RTK_I and stem-like, RTK_II and differentiated, and MES_TYP and immune components. These findings were also replicated in TCGA dataset. CONCLUSIONS Our findings underscore the importance of considering intra-tumoral heterogeneity in methylation profiles and the biological characteristics of each methylation subclass in developing novel GBM therapies.

NIMG-27. RADIOMETHYLOMICS FOR NON-INVASIVE PREDICTION OF TUMOR PHENOTYPES AND SURVIVAL IN GLIOBLASTOMA

Abstract BACKGROUND DNA methylation profiling has become a pivotal tool in neuro-oncology. However, little is known about the impact of methylation profiling on radiological imaging in IDH-wildtype glioblastoma (GBM). This study explored the potential of radiomics to non-invasively predict molecular characteristics based on the methylation profiling of GBM. METHODS This study included a multi-sampling cohort of 32 patients with 113 samples and a single-sampling cohort of 87 patients with 87 samples, which underwent genome-wide methylation analysis and were classified as GBM by the DKFZ/Heidelberg CNS tumor classifier. We performed two different methylation-based deconvolution analyses. Deconvolution 1 estimated the fractions of tumor subtypes (RTK_I, RTK_II, MES_TYP, and MES_ATYP) and cell types in the microenvironment. Deconvolution 2 inferred the abundance of malignant cell states (stem-like and differentiated cell components) and cell types in the microenvironment. We derived 6084 radiomic features from preoperative multi-parametric MRI in each patient. RESULTS In the multi-sampling cohort, Deconvolution 1 and Deconvolution 2 demonstrated that tumor subclass components and immune components exhibited heterogeneous distribution across samples within patients; however, the stem-like to differentiated cell ratio was preserved across samples within patients. In the entire cohort, patients with tumors exhibiting a high proportion of the stem-like component, defined as stem-like tumors, had significantly shorter overall survival. Stem-like tumors exhibited significantly higher levels of the RTK_I component and lower proportions of the RTK_II component compared to non-stem-like tumors. A machine learning model that utilizes support vector machines to classify stem-like tumors using radiomic features achieved a mean AUC of 0.71 (95% confidence interval: 0.56–0.85) in nested cross-validation. CONCLUSIONS We showed that radiomethylomics can be used to predict tumor composition and patient survival from preoperative MRI in GBM. Based on these promising results we will increase our power with additional samples to facilitate a personalized approach to GBM management.

Prognosis prediction via histological evaluation of cellular heterogeneity in glioblastoma

Glioblastomas (GBMs) are the most aggressive types of central nervous system tumors. Although certain genomic alterations have been identified as prognostic biomarkers of GBMs, the histomorphological features that predict their prognosis remain elusive. In this study, following an integrative diagnosis of 227 GBMs based on the 2021 World Health Organization classification system, the cases were histologically fractionated by cellular variations and abundance to evaluate the relationship between cellular heterogeneity and prognosis in combination with O-6-methylguanine-DNA methyltransferase gene promoter methylation (mMGMTp) status. GBMs comprised four major cell types: astrocytic, pleomorphic, gemistocytic, and rhabdoid cells. t-distributed stochastic neighbor embedding analysis using the histological abundance of heterogeneous cell types identified two distinct groups with significantly different prognoses. In individual cell component analysis, the abundance of gemistocytes showed a significantly favorable prognosis but confounding to mMGMTp status. Conversely, the abundance of epithelioid cells was correlated with the unfavorable prognosis. Linear model analysis showed the favorable prognostic utility of quantifying gemistocytic and epithelioid cells, independent of mMGMTp. The evaluation of GBM cell histomorphological heterogeneity is more effective for prognosis prediction in combination with mMGMTp analysis, indicating that histomorphological analysis is a practical and useful prognostication tool in an integrative diagnosis of GBMs.

Integration of single-cell sequencing and bulk transcriptome data develops prognostic markers based on PCLAF+ stem-like tumor cells using artificial neural network in gastric cancer

BackgroundGastric cancer stem cells (GCSCs) are important tumour cells involved in tumourigenesis and gastric cancer development. However, their clinical value remains unclear due to the limitations of the available technologies. This study aims to explore the clinical significance of GCSCs, their connection to the tumour microenvironment, and their underlying molecular mechanisms. MethodsStem-like tumour cells were identified by mining single-cell transcriptomic data from multiple samples. Integrated analysis of single-cell and bulk transcriptome data was performed to analyse the role of stem-like tumour cells in predicting clinical outcomes by introducing the intermediate variable mRNA stemness degree (SD). Consensus clustering analysis was performed to develop an SD-related molecular classification strategy to assess the clinical characteristics in gastric cancer. A prognostic model was constructed using a customized approach that comprehensively considered SD-related gene signatures based on an artificial neural network. ResultsBy analysing single-cell data and validating immunofluorescence results, we identified a PCLAF+ stem-like tumour cell population in GC. By calculating SD, we observed that PCLAF+ stem-like tumour cells were associated with poor prognosis and certain clinical features. The SD was negatively correlated with the abundance of most immune cell types. Furthermore, we proposed an SD-related classification method and prognostic model. In addition, the customised prognostic model can be used to predict whether a patient respond to PD-1/PD-L1 immunotherapy. ConclusionWe identified a cluster of stem-like cells and elucidated their clinical significance, highlighting the possibility of their use as immunotherapeutic targets.

Variations in the germinal center response revealed by genetically diverse mouse strains.

The humoral response is complex and involves multiple cellular populations and signaling pathways. Bacterial and viral infections, as well as immunization regimens, can trigger this type of response, promoting the formation of microanatomical cellular structures called germinal centers (GCs). GCs formed in secondary lymphoid organs support the differentiation of high-affinity plasma cells and memory B cells. There is growing evidence that the quality of the humoral response is influenced by genetic variants. Using 12 genetically divergent mouse strains, we assessed the impact of genetics on GC cellular traits. At steady state, in the spleen, lymph nodes and Peyer's patches, we quantified GC B cells, plasma cells and follicular helper Tcells. These traits were also quantified in the spleen of mice following immunization with a foreign antigen, namely, sheep red blood cells, in addition to the number and size of GCs. We observed both strain- and organ-specific variations in cell type abundance, as well as for GC number and size. Moreover, we find that some of these traits are highly heritable. Importantly, the results of this study inform on the impact of genetic diversity in shaping the GC response and identify the traits that are the most impacted by genetic background.

Transcriptomic Correlates of State Modulation in GABAergic Interneurons: A Cross-Species Analysis.

GABAergic inhibitory interneurons comprise many subtypes that differ in their molecular, anatomical, and functional properties. In mouse visual cortex, they also differ in their modulation with an animal's behavioral state, and this state modulation can be predicted from the first principal component (PC) of the gene expression matrix. Here, we ask whether this link between transcriptome and state-dependent processing generalizes across species. To this end, we analysed seven single-cell and single-nucleus RNA sequencing datasets from mouse, human, songbird, and turtle forebrains. Despite homology at the level of cell types, we found clear differences between transcriptomic PCs, with greater dissimilarities between evolutionarily distant species. These dissimilarities arise from two factors: divergence in gene expression within homologous cell types and divergence in cell-type abundance. We also compare the expression of cholinergic receptors, which are thought to causally link transcriptome and state modulation. Several cholinergic receptors predictive of state modulation in mouse interneurons are differentially expressed between species. Circuit modelling and mathematical analyses suggest conditions under which these expression differences could translate into functional differences.

Molecular pathology, developmental changes and synaptic dysfunction in (pre-) symptomatic human C9ORF72-ALS/FTD cerebral organoids

A hexanucleotide repeat expansion (HRE) in C9ORF72 is the most common genetic cause of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). Human brain imaging and experimental studies indicate early changes in brain structure and connectivity in C9-ALS/FTD, even before symptom onset. Because these early disease phenotypes remain incompletely understood, we generated iPSC-derived cerebral organoid models from C9-ALS/FTD patients, presymptomatic C9ORF72-HRE (C9-HRE) carriers, and controls. Our work revealed the presence of all three C9-HRE-related molecular pathologies and developmental stage-dependent size phenotypes in cerebral organoids from C9-ALS/FTD patients. In addition, single-cell RNA sequencing identified changes in cell type abundance and distribution in C9-ALS/FTD organoids, including a reduction in the number of deep layer cortical neurons and the distribution of neural progenitors. Further, molecular and cellular analyses and patch-clamp electrophysiology detected various changes in synapse structure and function. Intriguingly, organoids from all presymptomatic C9-HRE carriers displayed C9-HRE molecular pathology, whereas the extent to which more downstream cellular defects, as found in C9-ALS/FTD models, were detected varied for the different presymptomatic C9-HRE cases. Together, these results unveil early changes in 3D human brain tissue organization and synaptic connectivity in C9-ALS/FTD that likely constitute initial pathologies crucial for understanding disease onset and the design of therapeutic strategies.

Deconvolution from bulk gene expression by leveraging sample-wise and gene-wise similarities and single-cell RNA-seq data

BackgroundThe widely adopted bulk RNA-seq measures the gene expression average of cells, masking cell type heterogeneity, which confounds downstream analyses. Therefore, identifying the cellular composition and cell type-specific gene expression profiles (GEPs) facilitates the study of the underlying mechanisms of various biological processes. Although single-cell RNA-seq focuses on cell type heterogeneity in gene expression, it requires specialized and expensive resources and currently is not practical for a large number of samples or a routine clinical setting. Recently, computational deconvolution methodologies have been developed, while many of them only estimate cell type composition or cell type-specific GEPs by requiring the other as input. The development of more accurate deconvolution methods to infer cell type abundance and cell type-specific GEPs is still essential.ResultsWe propose a new deconvolution algorithm, DSSC, which infers cell type-specific gene expression and cell type proportions of heterogeneous samples simultaneously by leveraging gene-gene and sample-sample similarities in bulk expression and single-cell RNA-seq data. Through comparisons with the other existing methods, we demonstrate that DSSC is effective in inferring both cell type proportions and cell type-specific GEPs across simulated pseudo-bulk data (including intra-dataset and inter-dataset simulations) and experimental bulk data (including mixture data and real experimental data). DSSC shows robustness to the change of marker gene number and sample size and also has cost and time efficiencies.ConclusionsDSSC provides a practical and promising alternative to the experimental techniques to characterize cellular composition and heterogeneity in the gene expression of heterogeneous samples.

Cognitive resilience to Alzheimer's disease characterized by cell-type abundance.

The molecular basis of cognitive resilience (CR) among pathologically confirmed Alzheimer's disease (AD) cases is not well understood. Abundance of 13 cell types and neuronal subtypes in brain bulk RNA-seq data from the anterior caudate, dorsolateral prefrontal cortex (DLPFC), and posterior cingulate cortex (PCC) obtained from 434 AD cases, 318 cognitively resilient AD cases, and 188 controls in the Religious Orders Study and Rush Memory and Aging Project was estimated by deconvolution. PVALB+ neuron abundance was negatively associated with cognitive status and tau pathology in the DLPFC and PCC (Padj<0.001) and the most reduced neuronal subtype in AD cases compared to controls in DLPFC (Padj=8.4×10-7) and PCC (Padj=0.0015). We identified genome-wide significant association of neuron abundance with TMEM106B single nucleotide polymorphism rs13237518 in PCC (p=6.08×10-12). rs13237518 was also associated with amyloid beta (p=0.0085) and tangles (p=0.0073). High abundance of PVALB+ neurons may be a marker of CR. TMEM106B variants may influence CR independent of AD pathology. Neuron retention and a lack of astrocytosis are highly predictive of Alzheimer's disease (AD) resilience. PVALB+ GABAergic and RORB+ glutamatergic neurons are associated with cognitive status. A TMEM106B single nucleotide polymorphism is related to lower AD risk, higher neuron count, and increased AD pathology.

Systematic transcriptomic analysis of childhood medulloblastoma identifies N6- methyladenosine-dependent lncRNA signatures associated with molecular subtype, immune cell infiltration, and prognosis.

Medulloblastoma, the most common malignant pediatric brain tumor, is classified into four main molecular subgroups, but group 3 and group 4 tumors are difficult to subclassify and have a poor prognosis. Rapid point-of-care diagnostic and prognostic assays are needed to improve medulloblastoma risk stratification and management. N6-methyladenosine (m6A) is a common RNA modification and long non-coding RNAs (lncRNAs) play a central role in tumor progression, but their impact on gene expression and associated clinical outcomes in medulloblastoma are unknown. Here we analyzed 469 medulloblastoma tumor transcriptomes to identify lncRNAs co-expressed with m6A regulators. Using LASSO-Cox analysis, we identified a five-gene m6A-associated lncRNA signature (M6LSig) significantly associated with overall survival, which was combined in a prognostic clinical nomogram. Using expression of the 67 m6A-associated lncRNAs, a subgroup classification model was generated using the XGBoost machine learning algorithm, which had a classification accuracy > 90%, including for group 3 and 4 samples. All M6LSig genes were significantly correlated with at least one immune cell type abundance in the tumor microenvironment, and the risk score was positively correlated with CD4+ naïve T cell abundance and negatively correlated with follicular helper T cells and eosinophils. Knockdown of key m6A writer genes METTL3 and METTL14 in a group 3 medulloblastoma cell line (D425-Med) decreased cell proliferation and upregulated many M6LSig genes identified in our in silico analysis, suggesting that the signature genes are functional in medulloblastoma. This study highlights a crucial role for m6A-dependent lncRNAs in medulloblastoma prognosis and immune responses and provides the foundation for practical clinical tools that can be rapidly deployed in clinical settings.

Systematic transcriptomic analysis of childhood medulloblastoma identifies N6-methyladenosine-dependent lncRNA signatures associated with molecular subtype, immune cell infiltration, and prognosis

Medulloblastoma, the most common malignant pediatric brain tumor, is classified into four main molecular subgroups, but group 3 and group 4 tumors are difficult to subclassify and have a poor prognosis. Rapid point-of-care diagnostic and prognostic assays are needed to improve medulloblastoma risk stratification and management. N6-methyladenosine (m6A) is a common RNA modification and long non-coding RNAs (lncRNAs) play a central role in tumor progression, but their impact on gene expression and associated clinical outcomes in medulloblastoma are unknown. Here we analyzed 469 medulloblastoma tumor transcriptomes to identify lncRNAs co-expressed with m6A regulators. Using LASSO-Cox analysis, we identified a five-gene m6A-associated lncRNA signature (M6LSig) significantly associated with overall survival, which was combined in a prognostic clinical nomogram. Using expression of the 67 m6A-associated lncRNAs, a subgroup classification model was generated using the XGBoost machine learning algorithm, which had a classification accuracy > 90%, including for group 3 and 4 samples. All M6LSig genes were significantly correlated with at least one immune cell type abundance in the tumor microenvironment, and the risk score was positively correlated with CD4+ naïve T cell abundance and negatively correlated with follicular helper T cells and eosinophils. Knockdown of key m6A writer genes METTL3 and METTL14 in a group 3 medulloblastoma cell line (D425-Med) decreased cell proliferation and upregulated many M6LSig genes identified in our in silico analysis, suggesting that the signature genes are functional in medulloblastoma. This study highlights a crucial role for m6A-dependent lncRNAs in medulloblastoma prognosis and immune responses and provides the foundation for practical clinical tools that can be rapidly deployed in clinical settings.

Spatial enrichment and genomic analyses reveal the link of NOMO1 with amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is a severe motor neuron disease with uncertain genetic predisposition in most sporadic cases. The spatial architecture of cell types and gene expression are the basis of cell-cell interactions, biological function and disease pathology, but are not well investigated in the human motor cortex, a key ALS-relevant brain region. Recent studies indicated single nucleus transcriptomic features of motor neuron vulnerability in ALS motor cortex. However, the brain regional vulnerability of ALS-associated genes and the genetic link between region-specific genes and ALS risk remain largely unclear. Here, we developed an entropy-weighted differential gene expression matrix-based tool (SpatialE) to identify the spatial enrichment of gene sets in spatial transcriptomics. We benchmarked SpatialE against another enrichment tool (multimodal intersection analysis) using spatial transcriptomics data from both human and mouse brain tissues. To investigate regional vulnerability, we analysed three human motor cortex and two dorsolateral prefrontal cortex tissues for spatial enrichment of ALS-associated genes. We also used Cell2location to estimate the abundance of cell types in ALS-related cortex layers. To dissect the link of regionally expressed genes and ALS risk, we performed burden analyses of rare loss-of-function variants detected by whole-genome sequencing in ALS patients and controls, then analysed differential gene expression in the TargetALS RNA-sequencing dataset. SpatialE showed more accurate and specific spatial enrichment of regional cell type markers than multimodal intersection analysis in both mouse brain and human dorsolateral prefrontal cortex. Spatial transcriptomic analyses of human motor cortex showed heterogeneous cell types and spatial gene expression profiles. We found that 260 manually curated ALS-associated genes are significantly enriched in layer 5 of the motor cortex, with abundant expression of upper motor neurons and layer 5 excitatory neurons. Burden analyses of rare loss-of-function variants in Layer 5-associated genes nominated NOMO1 as a novel ALS-associated gene in a combined sample set of 6814 ALS patients and 3324 controls (P = 0.029). Gene expression analyses in CNS tissues revealed downregulation of NOMO1 in ALS, which is consistent with a loss-of-function disease mechanism. In conclusion, our integrated spatial transcriptomics and genomic analyses identified regional brain vulnerability in ALS and the association of a layer 5 gene (NOMO1) with ALS risk.

Reconfiguration of the visual code and retinal cell type complement in closely related diurnal and nocturnal mice.

How does evolution act on neuronal populations to match computational characteristics to functional demands? We address this problem by comparing visual code and retinal cell composition in closely related murid species with different behaviours. Rhabdomys pumilio are diurnal and have substantially thicker inner retina and larger visual thalamus than nocturnal Mus musculus. High-density electrophysiological recordings of visual response features in the dorsal lateral geniculate nucleus (dLGN) reveals that Rhabdomys attains higher spatiotemporal acuity both by denser coverage of the visual scene and a selective expansion of elements of the code characterised by non-linear spatiotemporal summation. Comparative analysis of single cell transcriptomic cell atlases reveals that realignment of the visual code is associated with increased relative abundance of bipolar and ganglion cell types supporting OFF and ON-OFF responses. These findings demonstrate how changes in retinal cell complement can reconfigure the coding of visual information to match changes in visual needs.

Bulk and Single-cell Transcriptomic Brain Data Identify Overlapping Processes and Cell-types with Human AUD and Mammalian Models of Alcohol Use.

This study explores the neurobiological underpinnings of alcohol use disorder (AUD) by integrating bulk and single-cell transcriptomic data from humans, primates, and mice across three brain regions associated with addiction (i.e., prefrontal cortex (PFC), nucleus accumbens (NAc), and central amygdala (CeA)). We compared AUD RNA expression and cell-type abundance from 92 human brain to data from 53 primates and 90 mice engaged in diverse alcohol use paradigms. The findings revealed significant and reproducible correlations between human AUD and mammalian models of alcohol use that vary by tissue, species, and behavioral paradigm. The strongest correlations occurred between primate and mouse models of binge drinking (i.e., high drinking in the dark). Certain primate models demonstrated that the brain RNA correlations with human alcohol use disorder (AUD) were approximately 40% as strong as the correlations observed within human samples themselves. By integrating single-cell transcriptomic data, this study observed decreased oligodendrocyte proportions in the PFC and NAc of human AUD with similar trends in animal models. Gene co-expression network analyses revealed conserved systems associated with human AUD and animal models of heavy/binge alcohol consumption. Gene co-expression networks were enriched for pathways related to inflammation, myelination, and synaptic plasticity and the genes within them accounted for ~20% of the heritability in human alcohol consumption. Identified hub genes were associated with relevant traits (e.g., impulsivity, motivation) in humans and mice. This study sheds light on conserved biological entities underlying AUD and chronic alcohol use, providing insights into the cellular, genetic, and neuromolecular basis across species.

Metabolic pathway-based subtyping reveals distinct microenvironmental states associated with diffuse large B-cell lymphoma outcomes.

Diffuse large B-cell lymphoma (DLBCL) is a biologically and clinically heterogeneous disease that requires personalized clinical treatment. Assigning patients to different risk categories and cytogenetic abnormality and genetic mutation groups has been widely applied for prognostic stratification of DLBCL. Increasing evidence has demonstrated that dysregulated metabolic processes contribute to the initiation and progression of DLBCL. Metabolic competition within the tumor microenvironment is also known to influence immune cell metabolism. However, metabolism- and immune-related stratification has not been established. Here, 1660 genes involved in 84 metabolic pathways were selected and tested to establish metabolic clusters (MECs) of DLBCL. MECs established based on independent lymphoma datasets distinguished different survival outcomes. The CIBERSORT algorithm and EcoTyper were applied to quantify the relative abundance of immune cell types and identify variation in cell states for 13 lineages comprising the tumor micro environment among different MECs, respectively. Functional characterization showed that MECs were an indicator of the immune microenvironment and correlated with distinctive mutational characteristics and oncogenic signaling pathways. The novel immune-related MECs exhibited promising clinical prognostic value and potential for informing DLBCL treatment decisions.

Heterogeneous pseudobulk simulation enables realistic benchmarking of cell-type deconvolution methods

BackgroundComputational cell type deconvolution enables the estimation of cell type abundance from bulk tissues and is important for understanding tissue microenviroment, especially in tumor tissues. With rapid development of deconvolution methods, many benchmarking studies have been published aiming for a comprehensive evaluation for these methods. Benchmarking studies rely on cell-type resolved single-cell RNA-seq data to create simulated pseudobulk datasets by adding individual cells-types in controlled proportions.ResultsIn our work, we show that the standard application of this approach, which uses randomly selected single cells, regardless of the intrinsic difference between them, generates synthetic bulk expression values that lack appropriate biological variance. We demonstrate why and how the current bulk simulation pipeline with random cells is unrealistic and propose a heterogeneous simulation strategy as a solution. The heterogeneously simulated bulk samples match up with the variance observed in real bulk datasets and therefore provide concrete benefits for benchmarking in several ways. We demonstrate that conceptual classes of deconvolution methods differ dramatically in their robustness to heterogeneity with reference-free methods performing particularly poorly. For regression-based methods, the heterogeneous simulation provides an explicit framework to disentangle the contributions of reference construction and regression methods to performance. Finally, we perform an extensive benchmark of diverse methods across eight different datasets and find BayesPrism and a hybrid MuSiC/CIBERSORTx approach to be the top performers.ConclusionsOur heterogeneous bulk simulation method and the entire benchmarking framework is implemented in a user friendly package https://github.com/humengying0907/deconvBenchmarking and https://doi.org/10.5281/zenodo.8206516, enabling further developments in deconvolution methods.

Deep Cell-Type Deconvolution from Bulk Gene Expression Data Using DECODE

It is becoming clear that bulk gene expression measurements represent an average over very different cells. Elucidating the expression and abundance of each of the encompassed cells is key to disease understanding and precision medicine approaches. A first step in any such deconvolution is the inference of cell type abundances in the given mixture. Numerous approaches to cell-type deconvolution have been proposed, yet very few take advantage of the emerging discipline of deep learning and most approaches are limited to input data regarding the expression profiles of the cell types in question. Here we present DECODE, a deep learning method for the task that is data-driven and does not depend on input expression profiles. DECODE builds on a deep unfolded non-negative matrix factorization technique. It is shown to outperform previous approaches on a range of synthetic and real data sets, producing abundance estimates that are closer to and better correlated with the real values.

MDB-70. SYSTEMATIC TRANSCRIPTOMIC ANALYSIS OF CHILDHOOD MEDULLOBLASTOMA IDENTIFIES M6A-LNCRNA SIGNATURES ASSOCIATED WITH PROGNOSIS, MOLECULAR SUBTYPE, AND IMMUNE CELL INFILTRATION

Abstract BACKGROUND Medulloblastoma (MB) is the predominant pediatric brain tumor, subdivided into four molecular subgroups, with group 3 and group 4 tumors posing significant clinical challenges due to poor prognosis and classification. This study investigates the roles of N6-methyladenosine (m6A) and long non-coding RNAs (lncRNAs) in MB, aiming to develop diagnostic and prognostic tools. METHODS We analyzed transcriptome from 1236 samples to identify m6A-associated lncRNA signature (M6LSig). A multivariate-cox analysis identified a gene signature significantly associated with overall survival (OS) of patients. Risk score was calculated based on the sum of weighted gene expression of these genes. Nomograms were calculated for predicting OS and PFS probability. We did feature selection with Boruta to identify minimum gene signature for subgroup classification. We performed correlation analysis of gene signature and risk score with immune-cell abundance. We depleted METTL3 and METTL14 in G3-MB cell-line (D425) to measure impact on cell proliferation and dysregulation of M6LSig genes as a result of loss of m6A. RESULTS Nomogram was constructed using risk score derived from 24 M6LSig significantly associated with OS and PFS of the patients. A 171 gene signature was identified that can accurately classify samples into one of four major subgroups of MB with more than 90% accuracy including between Grp3/Grp4 samples. M6LSig genes and risk score showed significant correlation with immune cell type abundance in MB tumors. Expression of CD155 and 4-1BB was significantly different between high and low risk samples. m6A writer genes knockdown in D425 cells resulted in decreased cell proliferation and upregulation of many M6LSig genes. CONCLUSION M6LSig based risk can accurately predict OS and PFS and 171 genes signature can classify patients into subgroups with more than 90% accuracy. Risk score can also be used for determining high risk versus low-risk patient and design tailor made immune-therapy for better survival.