Biology subjects

Andreatta, M.

Publications and source records attributed to Andreatta, M..

17 recordsLinked to original sources

bioRxiv2026

Evaluating cell type annotations in single-cell omics in the absence of ground truth

Accurate cell type annotation is essential for single-cell transcriptomics, directly shaping downstream analyses and biological interpretations. Yet, objective evaluation of annotation quality remains a major challenge. Here, we argue that a cell type or cell state label has practical utility only if it captures a molecular pattern that is reproducible across biological replicates. Based on this principle, we introduce inter-sample consistency (ISC), a quantitative framework to assess annotation quality in single-cell RNA-seq datasets. Unlike existing cluster validation approaches, ISC distinguishes annotations that generalize across samples and individuals from those driven by technical or unwanted variation, thereby providing principled criteria for annotation quality and transferability. When applied to published single-cell atlases, ISC reveals widespread reproducibility gaps and provides actionable guidance for repairing inconsistent annotations. Notably, ISC enables benchmarking of automated cell type annotation tools even when ground-truth labels are unavailable, providing interpretable metrics to guide their development and evaluation. Implemented as the scTypeEval Bioconductor package, this framework offers a broadly applicable resource for evaluating and improving cell type annotations in single-cell RNA-seq experiments.

bioinformatics↗

bioRxiv2026

Cell type composition drives patient stratification in single-cell RNA-seq cohorts

Early transcriptomic studies demonstrated that unsupervised analysis of bulk gene expression can reveal clinically meaningful patient subgroups. Single-cell RNA sequencing (scRNA-seq) provides high-resolution characterization of cellular heterogeneity and therefore enables more refined patient stratification. Several computational approaches have been proposed to summarize single-cell data into sample-level representations for cohort-level exploratory analyses. However, these methods generally do not explicitly account for the compositional nature of cell-type proportions. Based on eleven scRNA-seq cohorts across different biological conditions, we evaluated several state-of-the-art sample representation methods for their ability to recover known biological groupings in an unsupervised setting. Surprisingly, we found that baseline approaches based on cell-type composition and pseudobulk gene expression consistently matched or outperformed more complex methods while requiring orders of magnitude fewer computational resources. In particular, centered log-ratio-transformed cell-type proportions achieved the highest stratification performance and demonstrated robustness to batch effects. The stratification signal was frequently concentrated in a small subset of highly variable cell types, and performance was robust across diverse cell type annotation strategies. Altogether, these results suggest that clinically relevant inter-sample variation in scRNA-seq cohorts is largely driven by differences in cell-type composition. Importantly, compositional representations directly link cohort-level structure to specific cell populations, enabling mechanistic interpretation and facilitating clinical translation. We provide scECODA, an open-source R package for scalable and interpretable cohort-level Exploratory COmpositional Data Analysis of scRNA-seq data, and establish cell-type compositional representations as a powerful and interpretable baseline for unsupervised patient stratification.

bioinformatics↗

bioRxiv2025

Metabolic reinvigoration of NK cells by IL-21 enhances immunotherapy against MHC-I deficient solid tumors

Natural killer (NK) cells, a type of potent cytotoxic lymphocytes, are particularly promising for the treatment of cancers that lose or downregulate major histocompatibility complex class I (MHC-I) expression to evade T cell-mediated immunotherapy. However, the hostile and immune suppressive tumor microenvironment (TME) greatly hinders the function of tumor-infiltrating NK cells limiting the therapeutic efficacy against solid tumors. Here, we show that a fusion protein of interleukin-21 (IL-21-Fc), as a direct in vivo intervention, can safely and effectively reprogram NK cell metabolism and enhance their effector function in the TME. Our research demonstrates that combining IL-21-Fc with IL-15 superagonist (IL-15SA) or NK cell transfer leads to the eradication of MHC-I-deficient tumors and confers durable protection in syngeneic and xenograft tumor models. Mechanistically, we uncover that IL-21-Fc enhances NK cell effector function by upregulating glycolysis in a lactate dehydrogenase A (LDHA)-dependent manner. These findings not only underscore the considerable potential of IL-21-Fc as an in vivo therapeutic intervention to bolster NK cell-based immunotherapy, but also unveil an innovative strategy of metabolic reprogramming for NK cell rejuvenation within tumors. GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=183 SRC="FIGDIR/small/643961v2_ufig1.gif" ALT="Figure 1"> View larger version (57K): org.highwire.dtl.DTLVardef@daf24forg.highwire.dtl.DTLVardef@5a5029org.highwire.dtl.DTLVardef@144c978org.highwire.dtl.DTLVardef@106ed1c_HPS_FORMAT_FIGEXP M_FIG C_FIG HIGHLIGHTSO_LINK cells display functional exhaustion in MHC-I deficient solid tumors C_LIO_LIIL-21-Fc as an in vivo-applicable and safe immunotherapy reinvigorates intratumoral NK cells for enhanced effector function C_LIO_LIIL-21-Fc enhances NK cell function by elevating glycolysis in a LDHA-dependent manner C_LIO_LICombining IL-21-Fc with low-dose IL-15SA or NK cell transfer eradicates syngeneic and xenografted solid tumors C_LI

cancer biology↗

bioRxiv2025

ECM Signatures Reveal Quiescent Stem Cell Diversity in the Colonic Niche

Colonic stem cells have a key role in the continuous regeneration of a healthy intestinal epithelium. Despite considerable advances in single-cell omics technologies, the transcriptional heterogeneity of rare cell types such as colonic stem cells, as well as their functional states and niche-specific behaviors, remain poorly characterised. In this study, we leverage a meta-analysis of scRNA-seq and spatial transcriptomic datasets to comprehensively map the heterogeneity of colonic stem cells. We identify multiple, previously underappreciated stem cell states, including distinct quiescent subtypes marked by CDKN1A (P21), CDKN1B (P27), and CDKN1C (P57), proliferative populations defined by MKI67 (Ki67) and LRIG1, and a lineage-committed intermediate subpopulation expressing MUC2. Strikingly, we find that these states can be robustly identified solely by their extracellular matrix (ECM) gene expression signatures, revealing ECM composition as a critical determinant of stem cell identity. Notably, LAMA1 expression is highly specific to the P57+ quiescent population, linking laminin-mediated microenvironments to the active maintenance of deep quiescence, consistent with our recent findings associating LAMA1 with quiescent cell survival. By applying these ECM gene signatures, we delineate discrete "micro-niches" of quiescent stem cells in healthy tissue and provide evidence that analogous states persist in the colorectal cancer (CRC) environment. Extending our approach to an unrelated tissue, the pancreas, we detect parallel quiescent cell subtypes, illustrating the broader applicability of ECM-based signatures. Taken together, our findings redefine the concept of stem cell heterogeneity in the colon, establish ECM-driven gene signatures as a powerful tool for characterizing stem cell states, and offer new perspectives on the niche-dependent regulation of both healthy and cancerous stem cell populations.

developmental biology↗

bioRxiv2024

Wounding triggers invasive progression in human basal cell carcinoma

The interconnection between wound healing and cancer has long been recognized, as epitomized by the expression "cancer is a wound that does not heal". However, the impact of inducing a wound, such as through biopsy collection, on the progression of established tumors remains largely unknown. In this study, we apply single-cell spatial transcriptomics to characterize the heterogeneity of human basal cell carcinoma (BCC) and identify a wound response gene program as the most prominent feature of highly invasive BCC. To explore the causal relationship between wounding and cancer invasive progression, we perform a longitudinal experiment to compare human tumors at baseline and one week post-biopsy. Our results demonstrate that biopsy collection triggers, in proximity of the wound, the same transcriptional cancer cell state observed in highly invasive BCC. This cancer cell transcriptional switch is coupled with morphological changes and the transcriptional reprogramming of cancer-associated fibroblasts (CAFs). This study provides evidence that wounding triggers invasive progression of established human tumors and warrants further research on the potentially harmful effects of biopsies and wound-inducing treatments.

cancer biology↗

bioRxiv2023

Semi-supervised integration of single-cell transcriptomics data

Single-cell sequencing technologies offer unprecedented opportunities to characterize the complexity of biological samples with high resolution. At the same time, variations in sample processing and experimental protocols introduce technical variability - or "batch effects" - in the molecular readouts, hindering comparative analyses across samples and individuals. Although batch effect correction methods are routinely applied in single-cell omics analyses, data integration often leads to overcorrection, resulting in the loss of true biological variability. In this study, we present STACAS v2, a semi-supervised scRNA-seq data integration method that leverages prior knowledge in the form of cell type annotations to preserve biological variance. Through an open and reproducible benchmarking pipeline, we show that semi-supervised STACAS outperforms popular unsupervised methods such as Harmony, FastMNN, Seurat v4, scVI, and Scanorama, as well as supervised methods such as scANVI and scGen. Notably, STACAS is robust to incomplete and imprecise cell type annotations, which are commonly encountered in real-life integration tasks. Highlighting its scalability, we successfully applied semi-supervised STACAS to construct a high-resolution map of tumor-infiltrating CD8 T cells encompassing over 500,000 cells from 265 individuals. Based on our findings, we argue that the incorporation of prior cell type information should be a common practice in single-cell data integration, and we provide a flexible framework for semi-supervised batch effect correction. STACAS seamlessly integrates with Seurat pipelines and can be run with one command: Run.STACAS(seurat.list, cell.labels).

bioinformatics↗

bioRxiv2022

NFAT5 induction by the tumor microenvironment enforces CD8 T cell exhaustion

Persistent exposure to antigen during chronic infection or cancer renders T cells dysfunctional. The molecular mechanisms regulating this state of exhaustion are thought to be common in infection and cancer, despite obvious differences in their microenvironments. We discovered that NFAT5, an NFAT family member lacking an AP-1 docking site, is highly expressed in exhausted T cells from murine and human tumors and is a central player in tumor-induced exhaustion. While NFAT5 overexpression reduced tumor control, NFAT5 deletion improved tumor control by promoting the accumulation of tumor-specific CD8+ T cells that expressed less TOX and PD-1 and produced more cytokines particularly among precursor exhausted cells. Conversely, NFAT5 had no effect on chronic infection-induced T cell exhaustion. Mechanistically we found that TCR triggering induced NFAT5 expression and that hyperosmolarity stimulated transcriptional activity of NFAT5. We propose that NFAT5 takes over NFAT1/2 to promote exhaustion specifically in tumor-infiltrating CD8+ T cells.

immunology↗

bioRxiv2021

scGate: marker-based purification of cell types from heterogeneous single-cell RNA-seq datasets

AO_SCPLOWBSTRACTC_SCPLOWA common bioinformatics task in single-cell data analysis is to purify a cell type or cell population of interest from heterogeneous datasets. Here we present scGate, an algorithm that automatizes marker-based purification of specific cell populations, without requiring training data or reference gene expression profiles. scGate purifies a cell population of interest using a set of markers organized in a hierarchical structure, akin to gating strategies employed in flow cytometry. In our benchmark for blood-derived and tumor-infiltrating immune cells, scGate outperforms SingleR, a state-of-the-art classifier for single-cell data. scGate is implemented as an R package and integrated with the Seurat framework, providing an intuitive tool to isolate cell populations of interest from complex scRNA-seq datasets. AvailabilityR package source code and reproducible tutorials are available at https://github.com/carmonalab/scGate

bioinformatics↗

bioRxiv2021

A single-cell reference atlas delineates CD4+ T cell subtype-specific adaptation during acute and chronic viral infections

CD4+ T cells are critical orchestrators of immune responses against a large variety of pathogens, including viruses. The multifaceted roles of CD4+ T cells, including their help to innate cells, CD8+ T and B cells and their support for long-lived immunity rely on a profound functional heterogeneity. While multiple CD4+ T cell subtypes and their key transcriptional regulators have been identified, there is a lack of consistent definition for CD4+ T cell transcriptional states. In addition, the progressive changes affecting CD4+ T cell subtypes during and after immune responses remain poorly defined. Taking advantage of single-cell transcriptomics, efficient computational methods, and robust animal models, we characterize the transcriptional landscape of CD4+ T cells responding to self-resolving and chronic viral infections. We build a comprehensive atlas of virus-specific CD4+ T cells and their evolution over time, and identify six major distinct cell states that are consistently observed in acute and chronic infections. During the course of acute infection, T cell composition progressively changes from effector to memory states, with subtype-specific gene modules and kinetics. Conversely, T cells in persistent infections fail to transition from effector to memory states, and acquire distinct, chronicity-associated transcriptional programs. By single-cell T cell receptor (TCR) sequencing analysis, we characterize the clonal structure of virus-specific CD4+ T cells across individuals and T cell subtypes. We find that virus-specific CD4+ T cell responses are mainly private across individuals and that most T cells differentiate into all subtypes independently of their TCR, in both acute and chronic infections. Finally, we show that our CD4+ T cell atlas can be used as a reference to accurately interpret cell states in external single-cell datasets. Overall, this study describes a previously unappreciated level of adaptation of the transcriptional states of CD4+ T cells responding to viruses and provides a new computational resource for CD4+ T cell analysis, available online at https://spica.unil.ch.

immunology↗

bioRxiv2021

UCell: robust and scalable single-cell gene signature scoring

UCell is an R package for evaluating gene signatures in single-cell datasets. UCell signature scores, based on the Mann-Whitney U statistic, are robust to dataset size and heterogeneity, and their calculation demands less computing time and memory than other available methods, enabling the processing of large datasets in a few minutes even on machines with limited computing power. UCell can be applied to any single-cell data matrix, and includes functions to directly interact with Seurat objects. The UCell package and documentation are available on GitHub at https://github.com/carmonalab/UCell

bioinformatics↗

bioRxiv2020

Projecting single-cell transcriptomics data onto a reference T cell atlas to interpret immune responses

Single-cell RNA-sequencing (scRNA-seq) has emerged as a revolutionary technology for characterizing the heterogeneity of cell populations. However, robust reference atlases that can be used to systematically interpret cellular states across studies and diseases are currently lacking. Here, we generated the first cross-study T cell atlases for cancer and viral infection and developed a novel algorithm, ProjecTILs, that enables the projection of new scRNA-seq data onto these reference atlases. ProjecTILs accurately predicted the effects of multiple perturbations, including the ablation of immunoregulatory targets controlling T cell differentiation, such as Tox, Ptpn2, miR-155 and Regnase-1, and suggested novel gene programs that were altered in these cells. Moving beyond mouse models, we used ProjecTILs to conduct a meta-analysis of human tumor-infiltrating T lymphocytes (TILs), revealing a remarkable conservation of TIL subtypes between human and mouse and across cancer types. Clonotype analysis supported a model in which rare human tumor-specific effector-memory (EM)-like CD8 TILs that resemble blood-circulating EM cells, differentiate into proliferative terminal exhausted/dysfunctional effector TILs through a progenitor subtype that upregulates the exhaustion master regulator TOX. Our novel computational method allows exploring the effect of human and murine T cell perturbations (e.g. as the result of therapy or genetic engineering) in terms of reference cellular states, altered genetic programs and clonotype structure, revealing mechanisms of action behind immunotherapies and opening opportunities for their improvement.

bioinformatics↗

bioRxiv2020

STACAS: Sub-Type Anchor Correction for Alignment in Seurat to integrate single-cell RNA-seq data

Computational tools for the integration of single-cell transcriptomics data are designed to correct batch effects between technical replicates or different technologies applied to the same population of cells. However, they have inherent limitations when applied to heterogeneous sets of data with moderate overlap in cell states or sub-types. STACAS is a package for the identification of integration anchors in the Seurat environment, optimized for the integration of datasets that share only a subset of cell types. We demonstrate that by i) correcting batch effects while preserving relevant biological variability across datasets, ii) filtering aberrant integration anchors with a quantitative distance measure, and iii) constructing optimal guide trees for integration, STACAS can accurately align scRNA-seq datasets composed of only partially overlapping cell populations. We anticipate that the algorithm will be a useful tool for the construction of comprehensive single-cell atlases by integration of the growing amount of single-cell data becoming available in public repositories. Code availabilityR package: https://github.com/carmonalab/STACAS Docker image: https://hub.docker.com/repository/docker/mandrea1/stacas_demo

bioinformatics↗

bioRxiv2020

The role of intolerance of uncertainty in the acquisition and extinction of reward

Individuals, who score high in self-reported intolerance of uncertainty (IU), tend to find uncertainty anxiety-provoking. IU has been reliably associated with disrupted threat extinction. However, it remains unclear whether IU would be related to disrupted extinction to other arousing stimuli that are not threatening (i.e., rewarding). We addressed this question by conducting a reward associative learning task with acquisition and extinction training phases (n = 58). Throughout the associative learning task, we recorded valence ratings (i.e. liking), skin conductance response (SCR) (i.e. sweating), and corrugator supercilii activity (i.e. brow muscle indicative or negative and positive affect) to learned reward and neutral cues. During acquisition training with partial reward reinforcement, higher IU was associated with greater corrugator supercilii activity to neutral compared to reward cues. IU was not related to valence ratings or SCRs during the acquisition or extinction training phases. These preliminary results suggest that IU-related deficits during extinction may be limited to situations with threat. The findings further our conceptual understanding of IUs role in the associative learning and extinction of reward, and in relation to the processing of threat and reward more generally.

neuroscience↗

bioRxiv2019

NNAlign_MA; semi-supervised MHC peptidome deconvolution for accurate characterization of MHC binding motifs and improved T cell epitope prediction

Antigen presentation by Major Histocompatibility Complex (MHC) is a cornerstone of the cellular immune system, and characterization of the rules defining which peptides are suitable for MHC presentation is of key interest for understanding T cell immunogenicity and the development of T cell driven therapeutics. The collective set of peptides presented on a cell surface by MHC molecules is known as the immunopeptidome. Due to the vast MHC polymorphism, this peptidome is unique between individuals. Current state-of-the-art liquid chromatography mass spectrometry (LC-MS) technologies allow the identification of large peptidomes specific for a given host or cell line, and numerous studies have proven this data to be a rich source of information for learning the rules of MHC-mediated antigen presentation. The data obtained with MS techniques is usually poly-specific - i.e. it contains multiple sequence motifs matching the different MHC molecules expressed in the system under investigation. Deconvolution of this poly-specificity has proven a challenge, and imposes a substantial limitation for learning MHC specific rules for antigen presentation from immunopeptidome data. This since each ligand first must be unambiguously associated to its presenting MHC molecule(s) within the haplotype of the cell line. Here, we describe NNAlign_MA, a method that is able to address this challenge. NNAlign_MA is capable of taking a mixed training set composed of single-allele (SA) data (peptides assigned to single MHCs) and multi-allele (MA) data (peptides with multiple options for MHCs assignments) as input, and in a fully automated manner deconvolute the individual MHC restriction of all MA sequences while learning the binding specificities of all the MHCs present in the training set. The NNAlign_MA method was benchmarked on a large and diverse dataset, covering human MHC class I and class II, and bovine class I (BoLA) data. For all cases, the method was demonstrated to have unprecedented performance compared to state-of-the-art methods, achieving a complete deconvolution of binding motifs contained within poly-specificity MS eluted ligand data and an improved predictive performance for identification of both eluted ligands and T cell epitopes. Given its very high flexibility and ease of use, we expect NNAlign_MA to serve as an effective tool to increase our understanding of the rules for MHC antigen presentation and guide the development of novel T cell-based therapeutics.

bioinformatics↗

bioRxiv2018

Footprints of antigen processing boost MHC class II natural ligand binding predictions

Major Histocompatibility complex class II (MHC-II) molecules present peptide fragments to T cells for immune recognition. Current predictors for peptide:MHC-II binding are trained on binding affinity data, generated in-vitro and therefore lacking information about antigen processing. For the first time, we here describe prediction models of peptide:MHC-II binding trained directly on naturally eluted peptides, and show that these, in addition to peptide binding to the MHC, incorporate identifiable rules of antigen processing. In fact, we observed detectable signals of protease cleavage at defined positions of the peptides. We also hypothesize a role of the length of the terminal ligand protrusions for trimming the peptide to the epitope presented. The results of integrating binding affinity and eluted ligand data in a combined model demonstrate improved performance for the prediction of MHC-II ligands, and foreshadow a new generation of improved peptide:MHC-II prediction tools of considerable importance for understanding and manipulating immune responses.

immunology↗

bioRxiv2017

Computational Tools for the Identification and Interpretation of Sequence Motifs in Immunopeptidomes

Recent advances in proteomics and mass-spectrometry have widely expanded the detectable peptide repertoire presented by major histocompatibility complex (MHC) molecules on the cell surface, collectively known as the immunopeptidome. Finely characterizing the immunopeptidome brings about important basic insights into the mechanisms of antigen presentation, but can also reveal promising targets for vaccine development and cancer immunotherapy. In this report, we describe a number of practical and efficient approaches to analyze immunopeptidomics data, discussing the identification of meaningful sequence motifs in various scenarios and considering current limitations. We address the issue of filtering false hits and contaminants, and the problem of motif deconvolution in cell lines expressing multiple MHC alleles, both for the MHC class I and class II systems. Finally, we demonstrate how machine learning can be readily employed by non-expert users to generate accurate prediction models directly from mass-spectrometry eluted ligand data sets.

bioinformatics↗

bioRxiv2017

NetMHCpan 4.0: Improved peptide-MHC class I interaction predictions integrating eluted ligand and peptide binding affinity data

Cytotoxic T cells are of central importance in the immune systems response to disease. They recognize defective cells by binding to peptides presented on the cell surface by MHC (major histocompatibility complex) class I molecules. Peptide binding to MHC molecules is the single most selective step in the antigen presentation pathway. On the quest for T cell epitopes, the prediction of peptide binding to MHC molecules has therefore attracted large attention.\n\nIn the past, predictors of peptide-MHC interaction have in most cases been trained on binding affinity data. Recently an increasing amount of MHC presented peptides identified by mass spectrometry has been published containing information about peptide processing steps in the presentation pathway and the length distribution of naturally presented peptides. Here, we present NetMHCpan-4.0, a method trained on both binding affinity and eluted ligand data leveraging the information from both data types. Large-scale benchmarking of the method demonstrates an increased predictive performance compared to state-of-the-art when it comes to identification of naturally processed ligands, cancer neoantigens, and T cell epitopes.

bioinformatics↗