Subtype-specific Gene Research Articles

RanBALL: An Ensemble Random Projection Model for Identifying Subtypes of B-Cell Acute Lymphoblastic Leukemia.

As the most common pediatric malignancy, B-cell acute lymphoblastic leukemia (B-ALL) has multiple distinct subtypes characterized by recurrent and sporadic somatic and germline genetic alterations. Identifying B-ALL subtypes can facilitate risk stratification and enable tailored therapeutic design. Existing methods for B-ALL subtyping primarily depend on immunophenotyping, cytogenetic tests and genomic profiling, which would be costly, complicated, and laborious. To overcome these challenges, we present RanBALL (an ensemble Ran dom projection-based model for identifying B - ALL subtypes), an accurate and cost-effective model for B-ALL subtype identification. By leveraging random projection (RP) and ensemble learning, RanBALL can preserve patient-to-patient distances after dimension reduction and yield robustly accurate classification performance for B-ALL subtyping. Benchmarking results based on >1700 B-ALL patients demonstrated that RanBALL achieved remarkable performance (accuracy: 0.93, F1-score: 0.93, and Matthews correlation coefficient: 0.93), significantly outperforming state-of-the-art methods like ALLSorts in terms of all performance metrics. In addition, RanBALL performs better than tSNE in terms of visualizing B-ALL subtype information. We believe RanBALL will facilitate the discovery of B-ALL subtype-specific marker genes and therapeutic targets to have consequential positive impacts on downstream risk stratification and tailored treatment design. To extend its applicability and impacts, a Python-based RanBALL package is available at https://github.com/wan-mlab/RanBALL .

In Silico Integrative scRNA Analysis of Human Colonic Epithelium Indicates Four Tuft Cell Subtypes.

This study integrated and analyzed human single-cell RNA sequencing data from four publicly available datasets to enhance cellular resolution, unveiling a complex landscape of tuft cell heterogeneity within the human colon. Four tuft subtypes (TC1-TC4) emerged as defined by unique gene expression profiles, indicating potentially novel biological function. Tuft cell 1 (TC1) was characterized by an antimicrobial peptide signature; TC2 had an increased transcription machinery gene expression profile consistent with a progenitor-like cell; TC3 expressed genes related to ganglion (neuronal) development; and TC4 expressed genes related to tight junctions. Our analysis of subtype-specific gene expression and pathway enrichment showed variances in tuft cell subtypes between healthy individuals and those with inflammatory bowel disease (IBD). The frequency of TC1 and TC2 differed between healthy controls and IBD. Relative to healthy controls, TC1 and TC2 in IBD-tissue showed an upregulation of gene expression, favouring increased metabolism and immune function. These findings provide foundational knowledge about the complexity of the human colon tuft cell population and hint at their potential contributions to gut health. They provide a basis for future studies to explore the specific roles these cells may play in gut function during homeostasis and disease. We demonstrate the value of in silico approaches for hypothesis generation in relation to the putative functions of low-frequency gut cells for subsequent physiological analyses.

TMIC-43. BIOLOGICAL AND MOLECULAR CHARACTERIZATION OF NOVEL GLIOBLASTOMA TME SUBTYPES USING TRANSCRIPTOMIC ANALYSIS AND NETWORK MODELLING TO IDENTIFY NOVEL TARGETS OF VULNERABILITY

Abstract New precision medicine therapies are urgently required for glioblastoma (GBM). Recently, we developed a novel GBM classification system, identifying three patient clusters uniquely characterized by tumor microenvironment (TME) composition: TMELow, TMEMedium, and TMEHigh1. Our objective now is to further investigate these subtypes employing transcriptomic analysis and network modelling approaches to identify novel subtype-specific targets of vulnerability. We analyzed transcriptomic data from >600 GBM samples from publicly available and in-house datasets. All samples underwent TME subtyping. Next, we performed differential gene expression (using DESeq2, edgeR) and pathway analysis (using PROGENy). The ‘TRANSFAC’ tool 2 was used to identify potential transcription factors enriched in each TME subtype. TMELow tumours manifested highly upregulated NLGN3 (P=5.777e-5) and HES5 (P=2.233e-3) expression compared to TMEHigh. Moreover, TMEHigh compared to TMELow tumours were enriched for genes associated with tissue remodelling, and showed elevated MMP7 (P=3.051e-6), CLCL13 (P= 3.843e-4), and CXCL5 expression (P= 1.592e-6). PROGENy analysis suggested that the WNT, TRAIL, and JAK/STAT pathways were significantly enriched in TMEHigh, while the PI3K pathway was significantly upregulated in TMEMedium. Additionally, VEGF pathway activity was significantly upregulated in TMELow. TRANSFAC analysis measured by regulatory score (a measure of a transcription factor effect on gene expression) revealed that in TMELow, key transcription factors includes ETS2, NANOG, and MECP2 (regulatory score: 1.97, 1.45, and 1.4, respectively). In TMEMedium, transcription factors HIF1A, PPARA, and CDX2 were identified (regulatory score: 2.52, 2.08 and 1.96, respectively). In TMEHigh, TRANSFAC indicated activiation of transcription factors SMAD3, ETS2, and NFKB1 regulatory score: 2.56, 2.41, 2.24, respectively). Overall these findings highlight distinct molecular signatures across TME subtypes, and provide a deeper understanding of associated molecular mechanisms. Ongoing work is focused on validating key subtype specific genes and associated pathways using spatial proteomics. Prioritised targets will ultimately be interrogated in vivo, as novel subtype-specific treatments. 1. White et al. Ann Oncol. 2023;34(3):300–314. 2. Matys et al. Nucleic Acids Res. 2003;31.

Glial 'omics in ischemia: Acute stroke and chronic cerebral small vessel disease.

Vascular injury and pathologies underlie common diseases including ischemic stroke and cerebral small vessel disease (CSVD). Prior work has identified a key role for glial cells, including microglia, in the multifaceted and temporally evolving neuroimmune response to both stroke and CSVD. Transcriptional profiling has led to important advances including identification of distinct gene expression signatures in ischemia-exposed, flow cytometrically sorted microglia and more recently single cell RNA sequencing-identified microglial subpopulations or clusters. There is a reassuring degree of overlap in the results from these two distinct methodologies with both identifying a proliferative and a separate type I interferon responsive microglial element. Similar patterns were later seen using multimodal and spatial transcriptomal profiling in ischemia-exposed microglia and astrocytes. Methodological advances including enrichment of specific neuroanatomic/functional regions (such as the neurovascular unit) prior to single cell RNA sequencing has led to identification of novel cellular subtypes and generation of new credible hypotheses as to cellular function based on the enhanced cell sub-type specific gene expression patterns. A ribosomal tagging strategy focusing on the cellular translatome analyses carried out in the acute phases post stroke has revealed distinct inflammation-regulating roles for microglia and astrocytes in this setting. Early spatial transcriptomics experiments using cerebral ischemia models have identified regionally distinct microglial cell clusters in ischemic core versus penumbra. There is great potential for combination of these methods for multi-omics approaches to further elucidate glial responses in the context of both acute ischemic stroke and chronic CSVD.

Human dorsal root ganglia are either preserved or completely lost after deafferentation by brachial plexus injury

BackgroundPlexus injury results in lifelong suffering from flaccid paralysis, sensory loss, and intractable pain. For this clinical problem, regenerative medicine concepts set high expectations. However, it is largely unknown how dorsal root ganglia (DRG) are affected by accidental deafferentation. MethodsHere, we phenotyped DRG of a clinically and MRI-characterised cohort of 13 patients with plexus injury. Avulsed DRG were collected during reconstructive nerve surgery. For control, we used DRG from forensic autopsy. The cellular composition of the DRG was analysed in histopathological slices with multicolour high-resolution immunohistochemistry, tile microscopy, and deep-learning-based bioimage analysis. We then sequenced the bulk RNA of corresponding DRG slices. ResultsIn about half of the patients we found loss of the typical DRG units consisting of neurones and satellite glial cells. The DRG cells were replaced by mesodermal/connective tissue. In the remaining patients, the cellular units were well preserved. Preoperative plexus MRI neurography was not able to distinguish the two types. Patients with ‘neuronal preservation’ had less maximum pain than patients with ‘neuronal loss’. Arm function improved after nerve reconstruction, but severe pain persisted. Transcriptome analysis of preserved DRGs revealed expression of subtype-specific sensory neurone marker genes, but downregulation of neuronal attributes. Furthermore, they showed signs of ongoing inflammation and connective tissue remodelling. ConclusionsPatients with plexus injury separate into two groups with either neuronal preservation or neuronal loss. The former could benefit from anti-inflammatory therapy. For the latter, studies should explore mechanisms of neuronal loss especially for regenerative approaches. Clinical trial registrationDRKS00017266.

Beyond the Spectrum: Subtype-Specific Molecular Insights into Autism Spectrum Disorder Via Multimodal Data Integration.

Some toddlers with autism spectrum disorder (ASD) have mild social symptoms and developmental improvement in skills, but for others, symptoms and abilities are moderately or even severely affected. Those with profound autism have the most severe social, language, and cognitive symptoms and are at the greatest risk of having a poor developmental outcome. The little that is known about the underlying biology of this important profound autism subtype, points clearly to embryonic dysregulation of proliferation, differentiation and neurogenesis. Because it is essential to gain foundational knowledge of the molecular biology associated with profound, moderate, and mild autism clinical subtypes, we used well-validated, data-driven patient subtyping methods to integrate clinical and molecular data at 1 to 3 years of age in a cohort of 363 ASD and controls representative of the general pediatric population in San Diego County. Clinical data were diagnostic, language, cognitive and adaptive ability scores. Molecular measures were 50 MSigDB Hallmark gene pathway activity scores derived from RNAseq gene expression. Subtyping identified four ASD, typical and mixed diagnostic clusters. 93% of subjects in one cluster were profound autism and 93% in a different cluster were control toddlers; a third cluster was 76% moderate ability ASD; and the last cluster was a mix of mild ASD and control toddlers. Among the four clusters, the profound autism subtype had the most severe social symptoms, language, cognitive, adaptive, social attention eye tracking, social fMRI activation, and age-related decline in abilities, while mild autism toddlers mixed within typical and delayed clusters had mild social symptoms, and neurotypical language, cognitive and adaptive scores that improved with age compared with profound and moderate autism toddlers in other clusters. In profound autism, 7 subtype-specific dysregulated gene pathways were found; they control embryonic proliferation, differentiation, neurogenesis, and DNA repair. To find subtype-common dysregulated pathways, we compared all ASD vs TD and found 17 ASD subtype-common dysregulated pathways. These common pathways showed a severity gradient with the greatest dysregulation in profound and least in mild. Collectively, results raise the new hypothesis that the continuum of ASD heterogeneity is moderated by subtype-common pathways and the distinctive nature of profound autism is driven by the differentially added profound subtype-specific embryonic pathways.

Multimodal Hox5 activity generates motor neuron diversity

Motor neurons (MNs) are the final output of circuits driving fundamental behaviors, such as respiration and locomotion. Hox proteins are essential in generating the MN diversity required for accomplishing these functions, but the transcriptional mechanisms that enable Hox paralogs to assign distinct MN subtype identities despite their promiscuous DNA binding motif are not well understood. Here we show that Hoxa5 modifies chromatin accessibility in all mouse spinal cervical MN subtypes and engages TALE co-factors to directly bind and regulate subtype-specific genes. We identify a paralog-specific interaction of Hoxa5 with the phrenic MN-specific transcription factor Scip and show that heterologous expression of Hoxa5 and Scip is sufficient to suppress limb-innervating MN identity. We also demonstrate that phrenic MN identity is stable after Hoxa5 downregulation and identify Klf proteins as potential regulators of phrenic MN maintenance. Our data identify multiple modes of Hoxa5 action that converge to induce and maintain MN identity.

Molecular programs guiding arealization of descending cortical pathways.

Layer 5 extratelencephalic (ET) neurons are present across neocortical areas and send axons to multiple subcortical targets1-6. Two cardinal subtypes exist7,8: (1) Slco2a1-expressing neurons (ETdist), which predominate in the motor cortex and project distally to the pons, medulla and spinal cord; and (2) Nprs1- or Hpgd-expressing neurons (ETprox), which predominate in the visual cortex and project more proximally to the pons and thalamus. An understanding of how area-specific ETdist and ETprox emerge during development is important because they are critical for fine motor skills and are susceptible to spinal cord injury and amyotrophic lateral sclerosis9-12. Here, using cross-areal mapping of axonal projections in the mouse neocortex, we identify the subtype-specific developmental dynamics of ET neurons. Whereas subsets of ETprox emerge by pruning of ETdist axons, others emerge de novo. We outline corresponding subtype-specific developmental transcriptional programs using single-nucleus sequencing. Leveraging these findings, we use postnatal in vivo knockdown of subtype-specific transcription factors to reprogram ET neuron connectivity towards more proximal targets. Together, these results show the functional transcriptional programs driving ET neuron diversity and uncover cell subtype-specific gene regulatory networks that can be manipulated to direct target specificity in motor corticofugal pathways.

PTBP1-mediated repression of neuron-specific CDC42 splicing constitutes a genomic alteration-independent, developmentally conserved vulnerability in IDH-wildtype glioblastoma.

Gene co-expression networks may encode hitherto inadequately recognized vulnerabilities for adult gliomas. By identifying evolutionally conserved gene co-expression modules around EGFR (EM) or PDGFRA (PM), we recently proposed an EM/PM classification scheme, which assigns IDH-wildtype glioblastomas (GBM) into the EM subtype committed in neural stem cell compartment, IDH-mutant astrocytomas and oligodendrogliomas into the PM subtype committed in early oligodendrocyte lineage. Here, we report the identification of EM/PM subtype-specific gene co-expression networks and the characterization of hub gene polypyrimidine tract-binding protein 1 (PTBP1) as a genomic alteration-independent vulnerability in IDH-wildtype GBM. Supervised by the EM/PM classification scheme, we applied weighted gene co-expression network analysis to identify subtype-specific global gene co-expression modules. These gene co-expression modules were characterized for their clinical relevance, cellular origin and conserved expression pattern during brain development. Using lentiviral vector-mediated constitutive or inducible knockdown, we characterized the effects of PTBP1 on the survival of IDH-wildtype GBM cells, which was complemented with the analysis of PTBP1-depedent splicing pattern and overexpression of splicing target neuron-specific CDC42 (CDC42-N) isoform. Transcriptomes of adult gliomas can be robustly assigned into 4 large gene co-expression modules that are prognostically relevant and are derived from either malignant cells of the EM/PM subtypes or tumor microenvironment. The EM subtype is associated with a malignant cell-intrinsic gene module involved in pre-mRNA splicing, DNA replication and damage response, and chromosome segregation, and a microenvironment-derived gene module predominantly involved in extracellular matrix organization and infiltrating immune cells. The PM subtype is associated with two malignant cell-intrinsic gene modules predominantly involved in transcriptional regulation and mRNA translation, respectively. Expression levels of these gene modules are independent prognostic factors and malignant cell-intrinsic gene modules are conserved during brain development. Focusing on the EM subtype, we identified PTBP1 as the most significant hub for the malignant cell-intrinsic gene module. PTBP1 is not altered in most glioma genomes. PTBP1 represses the conserved splicing of CDC42-N. PTBP1 knockdown or CDC42-N overexpression disrupts actin cytoskeleton dynamics, causing accumulation of reactive oxygen species and cell apoptosis. PTBP1-mediated repression of CDC42-N splicing represents a potential genomic alteration-independent, developmentally conserved vulnerability in IDH-wildtype GBM.

Investigation of ferroptosis-associated molecular subtypes and immunological characteristics in lupus nephritis based on artificial neural network learning

BackgroundLupus nephritis (LN) is a severe complication of systemic lupus erythematosus (SLE) with poor treatment outcomes. The role and underlying mechanisms of ferroptosis in LN remain largely unknown. We aimed to explore ferroptosis-related molecular subtypes and assess their prognostic value in LN patients.MethodsMolecular subtypes were classified on the basis of differentially expressed ferroptosis-related genes (FRGs) via the Consensus ClusterPlus package. The enriched functions and pathways, immune infiltrating levels, immune scores, and immune checkpoints were compared between the subgroups. A scoring algorithm based on the subtype-specific feature genes identified by artificial neural network machine learning, referred to as the NeuraLN, was established, and its immunological features, clinical value, and predictive value were evaluated in patients with LN. Finally, immunohistochemical analysis was performed to validate the expression and role of feature genes in glomerular tissues from LN patients and controls.ResultsA total of 10 differentially expressed FRGs were identified, most of which showed significant correlation. Based on the 10 FRGs, LN patients were classified into two ferroptosis subtypes, which exhibited significant differences in immune cell abundances, immune scores, and immune checkpoint expression. A NeuraLN-related protective model was established based on nine subtype-specific genes, and it exhibited a robustly predictive value in LN. The nomogram and calibration curves demonstrated the clinical benefits of the protective model. The high-NeuraLN group was closely associated with immune activation. Clinical specimens demonstrated the alterations of ALB, BHMT, GAMT, GSTA1, and HAO2 were in accordance with bioinformatics analysis results, GSTA1 and BHMT were negatively correlated with the severity of LN.ConclusionThe classification of ferroptosis subtypes and the establishment of a protective model may form a foundation for the personalized treatment of LN patients.

Lineage-dependence of the neuroblastoma surfaceome defines tumor cell state-dependent and independent immunotherapeutic targets.

Neuroblastoma is a heterogeneous disease with adrenergic (ADRN)- and therapy resistant mesenchymal (MES)-like cells driven by distinct transcription factor networks. Here, we investigate the expression of immunotherapeutic targets in each neuroblastoma subtype and propose pan-neuroblastoma and cell state specific targetable cell-surface proteins. We characterized cell lines, patient-derived xenografts, and patient samples as ADRN-dominant or MES- dominant to define subtype-specific and pan-neuroblastoma gene sets. Targets were validated with ChIP- sequencing, immunoblotting, and flow cytometry in neuroblastoma cell lines and isogenic ADRN-to-MES transition cell line models. Finally, we evaluated the activity of MES-specific agents in vivo and in vitro . Most immunotherapeutic targets being developed for neuroblastoma showed significantly higher expression in the ADRN subtype with limited expression in MES-like tumor cells. In contrast, CD276 (B7-H3) and L1CAM maintained expression across both ADRN and MES states. We identified several receptor tyrosine kinases (RTKs) enriched in MES-dominant samples and showed that AXL targeting with ADCT-601 was potently cytotoxic in MES-dominant cell lines and showed specific anti-tumor activity in a MES cell line-derived xenograft. Immunotherapeutic strategies for neuroblastoma must address the potential of epigenetic downregulation of antigen density as a mechanism for immune evasion. We identified several RTKs as candidate MES-specific immunotherapeutic target proteins for the elimination of therapy-resistant cells. We hypothesize that the phenomena of immune escape will be less likely when targeting pan-neuroblastoma cell surface proteins such as B7-H3 and L1CAM, and/or dual targeting strategies that consider both the ADRN- and MES-cell states. Cellular plasticity influences the abundance of immunotherapeutic targets.Subtype-specific targets may be susceptible to epigenetically-mediated downregulation.Immunotherapeutic targets in development, B7-H3 and L1CAM, show "pan-subtype" expression. Neuroblastoma is a lethal childhood malignancy that shows cellular plasticity in response to anti-cancer therapies. Several plasma membrane proteins are being developed as immunotherapeutic targets in this disease. Here we define which cell surface proteins are susceptible to epigenetically regulated downregulation during an adrenergic to mesenchymal cell state switch and propose immunotherapeutic strategies to anticipate and circumvent acquired immunotherapeutic resistance.

#1479 Single nucleus RNA-sequencing indicates functional heterogeneity of renal endothelial cells in chronic kidney disease

Abstract Background and Aims The burden of cardiovascular disease (CVD) in patients with chronic kidney disease (CKD) remains very high. Endothelial cell (EC) dysfunction is an important key player in the initiation and progression of both pathologies. More importantly, endothelial dysfunction occurs early in both disease courses and is potentially reversible, making it a promising target for effective preventive strategies. Recently, accumulating single-cell RNA sequencing-based evidence has highlighted EC heterogeneity in different pathophysiological contexts. However, while biological processes such as systemic inflammation, oxidative stress fibrosis are being defined as comm EC in CKD and CVD, a comprehensive understanding of at single-cell resolution is lacking. Here, we characterized renal EC subpopulations by single-nucleus RNA sequencing (snRNA-seq) to investigate alterations in CKD patients healthy controls and identify potential drivers of EC dysfunction in CKD. Method Droplet-based snRNA-seq (10X Genomics Chromium) was performed on cryopreserved kidney biopsy cores from 6 human patients with CKD (eGFR < 45 mL/min/1.73 m2) and 3 healthy controls. CKD patients included 2 patients with primary focal segmental glomerulosclerosis (FSGS) and 4 with secondary/maladaptive FSGS (2 patients with diabetes mellitus, 2 female patients in total); all patients had a Framingham 10-year cardiovascular risk score > 20, but no established CVD. Healthy control samples were pre-perfusion biopsies from patients undergoing living kidney donation (including 2 female patients). EC subclusters were annotated based on the expression of canonical EC marker genes. Differential gene expression analysis for patient samples vs. controls included all ECs and DEGs were filtered for log2FC > 0.25 or < −0.25 and adjusted P-value of < .05. Gene set enrichment analysis (GSEA) on the identified DEGs was performed using GO terms. Results A total of 4 081 ECs were present in the dataset, consisting of 5 EC subtypes in both the patient and control cohorts (Panel A). DEGs in CKD patients vs. controls included upregulated genes related to organogenesis and cell differentiation (MEF2C, MEIS2), vascular development (FLRT2), and angiogenesis regulators (VEGFC, PRCP, HDAC9) (Panel B). Notably, VEGFC and FLRT2, are involved in directing EC motility and vascular patterning. In accordance, GSEA showed enrichment for GO biological processes related to vascular endothelial growth factor signaling pathway, endothelial cell differentiation, endothelial development, regulation of angiogenesis, and endothelial cell migration (Panel C). This suggests a potential link between these genes and aberrant angiogenesis, analogous to previous findings in diabetic nephropathy. Some but not all of these changes were also reported for renal ECs of early diabetic nephropathy patients. Of note, the majority of patients studied in our cohort were non-diabetic, warranting further investigation into renal endothelial cell transcriptional alterations that are general to renal disease or specific to particular renal conditions such as diabetic nephropathy. Differential gene expression and GSEA analysis of the EC subclusters in CKD indicated involvement of the peritubular capillary and ascending vasa recta ECs in EC migration / angiogenesis-related responses, overlapping with the overall analysis described above. Glomerular capillary ECs showed upregulation of immune-related genes and processes, reflecting the inflammatory status of the kidney disease. Conclusion snRNA-seq analysis revealed heterogeneity and subtype-specific gene expression alterations of renal ECs from CKD patients compared to healthy controls. Differential functions include enhanced angiogenic (potentially compensatory) and immune-related responses. These findings suggest that specific biological functions may direct different roles of the renal EC subtypes in CKD and will be further validated for their cluster-specificity and for their therapeutic relevance.

Integrated Exploration of Epigenetic Dysregulation Reveals a Stemness/EMT Subtype and MMP12 Linked to the Progression and Prognosis in Hepatocellular Carcinoma.

Epigenetic dysregulation drives aberrant transcriptional programs playing a critical role in hepatocellular carcinoma (HCC), which may provide novel insights into the heterogeneity of HCC. This study performed an integrated exploration on the epigenetic dysregulation of miRNA and methylation. We discovered and validated three patterns endowed with gene-related transcriptional traits and clinical outcomes. Specially, a stemness/epithelial-mesenchymal transition (EMT) subtype was featured by immune exhaustion and the worst prognosis. Besides, MMP12, a characteristic gene, was highly expressed in the stemness/EMT subtype, which was verified as a pivotal regulator linked to the unfavorable prognosis and further proven to promote tumor proliferation, invasion, and metastasis in vitro experiments. Proteomic analysis by mass spectrometry sequencing also indicated that the overexpression of MMP12 was significantly associated with cell proliferation and adhesion. Taken together, this study unveils innovative insights into epigenetic dysregulation and identifies a stemness/EMT subtype-specific gene, MMP12, correlated with the progression and prognosis of HCC.

Network-based approach elucidates critical genes in BRCA subtypes and chemotherapy response in triple negative breast cancer

Breast cancers (BRCA) exhibit substantial transcriptional heterogeneity, posing a significant clinical challenge. The global transcriptional changes in a disease context, however, are likely mediated by few key genes which reflect disease etiology better than the differentially expressed genes (DEGs). We apply our network-based tool PathExt to 1,059 BRCA tumors across 4 subtypes to identify key mediator genes in each subtype. Compared to conventional differential expression analysis, PathExt-identified genes exhibit greater concordance across tumors, revealing shared and subtype-specific biological processes; better recapitulate BRCA-associated genes in multiple benchmarks, and are more essential in BRCA subtype-specific cell lines. Single-cell transcriptomic analysis reveals a subtype-specific distribution of PathExt-identified genes in multiple cell types from the tumor microenvironment. Application of PathExt to a TNBC chemotherapy response dataset identified subtype-specific key genes and biological processes associated with resistance. We described putative drugs that target key genes potentially mediating drug resistance.

Comprehensive analyses of the cancer-associated fibroblast subtypes and their score system for prediction of outcomes and immunosuppressive microenvironment in prostate cancer

BackgroundCancer-associated fibroblasts (CAFs) drive cancer progression and treatment failure on one hand, while their tumor-restraining functions are also observed on the other. Recent single cell RNA sequencing (scRNA-seq) analyses demonstrates heterogeneity of CAFs and defines molecular subtypes of CAFs, which help explain their different functions. However, it remains unclear whether these CAF subtypes have the same or different biological/clinical implications in prostate cancer (PCa) or other malignancies.MethodsPCa cells were incubated with supernatant from normal fibroblasts and CAFs to assess their effects on cell behaviors. Sequencing, genomic, and clinical data were collected from TCGA, MSKCC, CPGEA and GEO databases. CAF molecular subtypes and total CAF scores were constructed and grouped into low and high groups based on CAF-specific gene expression. Progression free interval (PFI), clinicopathological features, telomere length, immune cell infiltration, drug treatment and somatic mutations were compared among CAF molecular subtypes and low/high score groups.ResultsThe PCa CAF-derived supernatant promoted PCa cell proliferation and invasion. Based on differentially expressed genes identified by scRNA-seq analyses, we classified CAFs into 6 molecular subtypes in PCa tumors, and each subtype was then categorized into score-high and low groups according to the subtype-specific gene expression level. Such score models in 6 CAF subtypes all predicted PFI. Telomeres were significantly shorter in high-score tumors. The total CAF score from 6 CAF subtypes was also associated with PFI in PCa patients inversely, which was consistent with results from cellular experiments. Immunosuppressive microenvironment occurred more frequently in tumors with a high CAF score, which was characterized by increased CTLA4 expression and indicated better responses to CTLA4 inhibitors. Moreover, this model can also serve as a useful PFI predictor in pan-cancers.ConclusionBy combining scRNA-seq and bulk RNA-seq data analyses, we develop a CAF subtype score system as a prognostic factor for PCa and other cancer types. This model system also helps distinguish different immune-suppressive mechanisms in PCa, suggesting its implications in predicting response to immunotherapy. Thus, the present findings should contribute to personalized PCa intervention.

Abstract 4907: RanBALL: Identifying B-cell acute lymphoblastic leukemia subtypes based on an ensemble random projection model

Abstract As the most common pediatric malignancy, B-cell acute lymphoblastic leukemia (B-ALL) has multiple distinct subtypes characterized by recurrent and sporadic somatic and germline genetic alterations like chromosomal alteration, transcription factor rearrangement or kinase inhibition. The treatment of B-ALL patients is personalized based on specific subtypes, as the treatment responses for different B-ALL subtypes may vary considerably. Identification of B-ALL subtypes can facilitate risk stratification and enable tailored therapeutic approaches. Existing methods for B-ALL subtyping primarily depend on immunophenotypic, cytogenetic and genomic analyses, which would be costly, complicated, and laborious in clinical practice applications. To overcome these challenges, we present RanBALL (an Ensemble Random Projection-Based Model for Identifying B-Cell Acute Lymphoblastic Leukemia Subtypes), an accurate and cost-effective model for B-ALL subtype identification based on transcriptomic profiling only. RanBALL leverages random projection (RP) to construct an ensemble of dimension-reduced multi-class classifiers for B-ALL subtyping. Specifically, the transcriptomic profiling features were projected onto low-dimensional spaces by random projection matrices whose elements conform to a distribution characterized by zero mean and unit variance. To ensure reliable and robust performance, we selected 20 subspace dimensions ranging from 600 to 2500, with intervals of 100. The transformed low dimensional data matrix was used for training an ensemble of multi-class support vector machine (SVM) classifiers, each corresponding to one of the RP matrices of various dimensions. The predicted probabilistic scores of each B-ALL subtype were integrated for determining the final decision. Results based on 10 times 10-fold cross validation tests for >1700 B-ALL patients demonstrated that the proposed model achieved an accuracy of 93.7%, indicating promising prediction capabilities of RanBALL for B-ALL subtyping. Furthermore, the 30% held-out tests suggested that the model was robust and consistent to maintain high confidence levels for accurate predictions. The high accuracies of RanBALL suggested that our model could effectively capture underlying patterns of transcriptomic profiling for accurate B-ALL subtype identification. To extend the impact of RanBALL, we have established a free and publicly available python package for RanBALL available at https://github.com/wan-mlab/RanBALL. We believe RanBALL will facilitate the discovery of B-ALL subtype-specific marker genes and therapeutic targets, and eventually have consequential positive impacts on downstream risk stratification and tailored treatment design. Citation Format: Lusheng Li, Hanyu Xiao, Joseph D. Khoury, Jieqiong Wang, Shibiao Wan. RanBALL: Identifying B-cell acute lymphoblastic leukemia subtypes based on an ensemble random projection model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4907.

Abstract 3500: Refining diagnosis of renal cell carcinoma subtypes through single-cell resolution transcriptomic signatures

Abstract A remaining challenge in the field of renal cell carcinoma (RCC) is diagnosis of rare or mixed subtypes, which hampers the delivery of the most effective healthcare to patients. While a new classification scheme, integrating molecular characteristics of tumors, is emerging, it requires a deep understanding of the tumor heterogeneity and etiopathogenesis of each subtype. We have generated and analyzed single-cell transcriptome profiles of 38 human RCCs and normal kidneys 16 RCCs with diverse histological subtypes and established gene signatures specific to each cell type. We have applied these gene signatures to bulk RNA-sequencing data of 1000 RCC tumors with different histological subtypes and normal kidneys from The Cancer Genome Atlas (TCGA) to characterize different cancer types and cellular origins. Unsupervised clustering analysis of TCGA data shows diagnostic discrepancies for 9.3% cases and subclassifies each major subtype (clear cell, papillary and chromophobe) into clusters with differential clinical outcomes.[KM1]. The pathological marker expression in misdiagnosed tumors is consistent with clustering-based classification, not their original diagnosis. Notably, using unique and subtype-specific gene signature patterns, we identified clear cell papillary renal cell tumor (ccpRCT), which is a newly classified benign entity and the main tumor type representing the mis-diagnosed tumors. These observations have further been confirmed by reviewing histological images of the tumors by an experienced genitourinary pathologist. Interestingly, our gene signatures reveal that ccpRCT does not originate from proximal tubule cells unlike clear cell and papillary RCCs. Taken together, our approach suggests a potential new classification method connected to clinical behaviors and drug responses based on overall transcriptomic differences. Citation Format: Minjun Kim, Ariel Madrigal, Zohreh Mehrjoo, Kate I. Glennon, Madeleine Arseneault, Morag Park, Fadi Brimo, Simon Tanguay, Hamed S. Najafabadi, Yasser Riazalhosseini. Refining diagnosis of renal cell carcinoma subtypes through single-cell resolution transcriptomic signatures [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3500.

Genetic Optimization in Uncovering Biologically Meaningful Gene Biomarkers for Glioblastoma Subtypes

Background: Glioblastoma multiforme (GBM) is a highly aggressive brain cancer known for its challenging survival rates; it is characterized by distinct subtypes, such as the proneural and mesenchymal states. The development of targeted therapies is critically dependent on a thorough understanding of these subtypes. Advances in single-cell RNA-sequencing (scRNA-seq) have opened new avenues for identifying subtype-specific gene biomarkers, which are essential for innovative treatments. Methods: This study introduces a genetic optimization algorithm designed to select a precise set of genes that clearly differentiate between the proneural and mesenchymal GBM subtypes. By integrating differential gene expression analysis with gene variability assessments, our dual-criterion strategy ensures the selection of genes that are not only differentially expressed between subtypes but also exhibit consistent variability patterns. This approach enhances the biological relevance of identified biomarkers. We applied this algorithm to scRNA-seq data from GBM samples, focusing on the discovery of subtype-specific gene biomarkers. Results: The application of our genetic optimization algorithm to scRNA-seq data successfully identified significant genes that are closely associated with the fundamental characteristics of GBM. These genes show a strong potential to distinguish between the proneural and mesenchymal subtypes, offering insights into the molecular underpinnings of GBM heterogeneity. Conclusions: This study introduces a novel approach for biomarker discovery in GBM that is potentially applicable to other complex diseases. By leveraging scRNA-seq data, our method contributes to the development of targeted therapies, highlighting the importance of precise biomarker identification in personalized medicine.

Identifying functional subtypes of IgA nephropathy based on three machine learning algorithms and WGCNA

BackgroundIgA nephropathy (IgAN) is one of the most common primary glomerulonephritis, which is a significant cause of renal failure. At present, the classification of IgAN is often limited to pathology, and its molecular mechanism has not been established. Therefore we aim to identify subtypes of IgAN at the molecular level and explore the heterogeneity of subtypes in terms of immune cell infiltration, functional level.MethodsTwo microarray datasets (GSE116626 and GSE115857) were downloaded from GEO. Differential expression genes (DEGs) for IgAN were screened with limma. Three unsupervised clustering algorithms (hclust, PAM, and ConsensusClusterPlus) were combined to develop a single-sample subtype random forest classifier (SSRC). Functional subtypes of IgAN were defined based on functional analysis and current IgAN findings. Then the correlation between IgAN subtypes and clinical features such as eGFR and proteinuria was evaluated by using Pearson method. Subsequently, subtype heterogeneity was verified by subtype-specific modules identification based on weighted gene co-expression network analysis(WGCNA) and immune cell infiltration analysis based on CIBERSORT algorithm.ResultsWe identified 102 DEGs as marker genes for IgAN and three functional subtypes namely: viral-hormonal, bacterial-immune and mixed type. We screened seventeen genes specific to viral hormonal type (ATF3, JUN and FOS etc.), and seven genes specific to bacterial immune type (LIF, C19orf51 and SLPI etc.). The subtype-specific genes showed significantly high correlation with proteinuria and eGFR. The WGCNA modules were in keeping with functions of the IgAN subtypes where the MEcyan module was specific to the viral-hormonal type and the MElightgreen module was specific to the bacterial-immune type. The results of immune cell infiltration revealed subtype-specific cell heterogeneity which included significant differences in T follicular helper cells, resting NK cells between viral-hormone type and control group; significant differences in eosinophils, monocytes, macrophages, mast cells and other cells between bacterial-immune type and control.ConclusionIn this study, we identified three functional subtypes of IgAN for the first time and specific expressed genes for each subtype. Then we constructed a subtype classifier and classify IgAN patients into specific subtypes, which may be benefit for the precise treatment of IgAN patients in future.

Signature construction and molecular subtype identification based on liver-specific genes for prediction of prognosis, immune activity, and anti-cancer drug sensitivity in hepatocellular carcinoma

BackgroundLiver specific genes (LSGs) are crucial for hepatocyte differentiation and maintaining normal liver function. A deep understanding of LSGs and their heterogeneity in hepatocellular carcinoma (HCC) is necessary to provide clues for HCC diagnosis, prognosis, and treatment.MethodsThe bulk and single-cell RNA-seq data of HCC were downloaded from TCGA, ICGC, and GEO databases. Through unsupervised cluster analysis, LSGs-based HCC subtypes were identified in TCGA-HCC samples. The prognostic effects of the subtypes were investigated with survival analyses. With GSVA and Wilcoxon test, the LSGs score, stemness score, aging score, immune score and stromal score of the samples were estimated and compared. The HCC subtype-specific genes were identified. The subtypes and their differences were validated in ICGC-HCC samples. LASSO regression analysis was used for key gene selection and risk model construction for HCC overall survival. The model performance was estimated and validated. The key genes were validated for their heterogeneities in HCC cell lines with quantitative real-time PCR and at single-cell level. Their dysregulations were investigated at protein level. Their correlations with HCC response to anti-cancer drugs were estimated in HCC cell lines.ResultsWe identified three LSGs-based HCC subtypes with different prognosis, tumor stemness, and aging level. The C1 subtype with low LSGs score and high immune score presented a poor survival, while the C2 subtype with high LSGs score and immune score indicated an enduring survival. Although no significant survival difference between C2 and C3 HCCs was shown, the C2 HCCs presented higher immune score and stroma score. The HCC subtypes and their differences were confirmed in ICGC-HCC dataset. A five-gene prognostic signature for HCC survival was constructed. Its good performance was shown in both the training and validation datasets. The five genes presented significant heterogeneities in different HCC cell lines and hepatocyte subclusters. Their dysregulations were confirmed at protein level. Furthermore, their significant associations with HCC sensitivities to anti-cancer drugs were shown.ConclusionsLSGs-based HCC subtype classification and the five-gene risk model might provide useful clues not only for HCC stratification and risk prediction, but also for the development of more personalized therapies for effective HCC treatment.