Biology subjects

Gambardella, G.

Publications and source records attributed to Gambardella, G..

8 recordsLinked to original sources

bioRxiv2026

ESMRank reveals a transferable axis of protein mutational constraint from overlapping variant effect assays

Proteome-wide interpretation of missense variation is constrained not only by predictive model performance but also by the absence of principled methods to reconcile heterogeneous multiplexed assays of variant effect (MAVEs) into a unified representation of mutational constraint. We show that redundancy among partially overlapping deep mutational scanning experiments encodes a reproducible ordinal signal that can be recovered despite differences in assay scale and readout. We introduce variant soundness, an overlap-aware framework that aligns within-assay rankings and aggregates them across experiments to derive an assay-agnostic, within-protein measure of mutational tolerance. Applying this approach to about 1,100 MAVEdb score sets spanning >2M variants reveals a coherent constraint landscape enriched for structural stability determinants, including residue burial, packing perturbation magnitude, and domain architecture. By aligning learning objectives with this intrinsic ordering, we develop ESMRank, a sequence-based learning-to-rank predictor integrating protein language model representations with physicochemical descriptors. Under strict protein-level partitioning, ESMRank outperforms widely used stability and fitness predictors across the Human Domainome, ProteinGym stability assays, and VariBench folding kinetics. Without clinical supervision, the reconstructed constraint axis is enriched for ClinVar pathogenic variants and stratifies genes by mechanistic disease classes. In CFTR, predicted constraint tracks folding efficiency, channel activity, and pharmacological rescue. These findings establish experimental overlap as a scalable resource for extracting transferable mutational ordering and for building mechanistically interpretable, proteome-wide variant effect predictors.

bioinformatics↗

bioRxiv2024

Deep-leaning and shallow hybrid Nanopore-Illumina sequencing for affordable and accurate detection of germline variants.

Despite the complementary strengths of short- and long-read sequencing approaches, variant calling methods still rely on a single data type. In this study, we collected and harmonized Nanopore dataset of the 7 GIAB healthy individuals across three independent consortia. Then, by leveraging these harmonized Nanopore data, we explore the benefits of using a hybrid DeepVariant model to jointly process Illumina and Nanopore data for germline variant detection. We show that a shallow hybrid long-short sequencing approach can match or surpass germline variant detection accuracy of state-of-the-art single-technology methods, potentially reducing overall sequencing costs and with the advantage of also enabling detection of large germline structural variations. These findings offer promising potential for molecular diagnostics in clinical settings, particularly for rare genetic disease screenings.

bioinformatics↗

bioRxiv2023

Predicting drug response from single-cell expression profiles of tumours

Drug response prediction at the single cell level is an emerging field of research that aims to improve the efficacy and precision of cancer treatments. Here, we introduce DREEP (Drug Response Estimation from single-cell Expression Profiles), a computational method that leverages publicly available pharmacogenomic screens and functional enrichment analysis to predict single cell drug sensitivity from transcriptomic data. We validated DREEP extensively in vitro using several independent single-cell datasets with over 200 cancer cell lines and showed its accuracy and robustness. Additionally, we also applied DREEP to molecularly barcoded breast cancer cells and identified drugs that can selectively target specific cell populations. DREEP provides an in-silico framework to prioritize drugs from single-cell transcriptional profiles of tumours and thus helps in designing personalized treatment strategies and accelerate drug repurposing studies. DREEP is available at https://github.com/gambalab/DREEP.

cancer biology↗

bioRxiv2023

Single cell lineage tracing reveals subclonal dynamics of anti-EGFR therapy resistance in triple negative breast cancer.

Epidermal growth factor receptor (EGFR)-targeted therapies have demonstrated variable and unpredictable clinical responses in triple negative breast cancer (TNBC). To elucidate the underlying mechanisms of this variability, we employ cellular barcoding and single-cell transcriptomics to reconstruct the subclonal dynamics of EGFR-amplified TNBC cells in response to afatinib, a tyrosine kinase inhibitor (TKI) that irreversibly inhibits EGFR. Integrated lineage tracing analysis revealed a rare pre-existing subpopulation of cells with distinct biological signature, including elevated expression levels of IGFBP2 (Insulin-Like Growth Factor Binding Protein 2). We show that IGFBP2 overexpression is sufficient to render TNBC cells tolerant to afatinib treatment by activating the compensatory IGF1-R signalling pathway. Finally, based on reconstructed mechanisms of resistance, we employ deep learning techniques to predict the afatinib sensitivity of TNBC cells. Our strategy proved effective in reconstructing the complex signalling network driving EGFR-targeted therapy resistance, offering new insights for the development of individualized treatment strategies in TNBC.

cancer biology↗

bioRxiv2022

Single-cell gene set enrichment analysis and transfer learning for functional annotation of scRNA-seq data

Although an essential step, the functional annotation of cells often proves particularly challenging in the analysis of single-cell transcriptional data. Several methods have been developed to accomplish this task. However, in most cases, these rely on techniques initially developed for bulk RNA sequencing or simply make use of marker genes identified from cell clustering followed by supervised annotation. To overcome these limitations and automatise the process, we have developed two novel methods, the single-cell gene set enrichment analysis (scGSEA) and the single cell mapper (scMAP). scGSEA combines latent data representations and gene set enrichment scores to detect coordinated gene activity at single-cell resolution. scMAP uses transfer learning techniques to repurpose and contextualise new cells into a reference cell atlas. Using both simulated and real datasets, we show that scGSEA effectively recapitulates recurrent patterns of pathways activity shared by cells from different experimental conditions. At the same time, we show that scMAP can reliably map and contextualise new single cell profiles on a breast cancer atlas we recently released. Both tools are provided in an effective and straightforward workflow providing a framework to determine cell function and significantly improve annotation and interpretation of scRNA-seq data.

bioinformatics↗

bioRxiv2021

Molecular phenotyping of breast cancer cell lines at the single-cell level for automated cancer diagnosis and prediction of drug sensitivity

Breast cancer patient stratification is mainly driven by tumour receptor status and histological grading and subtyping, with about twenty percent of patients for which absence of any actionable biomarkers results in no clear therapeutic intervention. Cancer cells within the same tumour have heterogeneous phenotypes and exhibit dynamic plasticity. However, how to evaluate such heterogeneity and its impact on outcome and drug response is still unclear. Here, we transcriptionally profiled 35,276 individual cells from 32 breast cancer cell lines covering all main breast cancer subtypes to yield a breast cancer cell line atlas. We found high degree of heterogeneity in the expression of clinically relevant biomarkers across individual cells within the same cell line; such heterogeneity is non-genetic and dynamic. We computationally mapped single cell transcriptional profiles of patients tumour biopsies to the atlas to determine their composition in terms of cell lines. Each tumour was found to be heterogenous and composed of multiple cell lines mostly, but not exclusively, of the same subtype. We then trained an algorithm on the atlas to determine cell line composition from bulk gene expression profiles of tumour biopsies, thus providing a novel approach to patient stratification. Finally, we linked results from large-scale in vitro drug screening1,2 to the single cell data to computationally predict responses to more than 450 anticancer agents starting from single-cell transcriptional profiles. We thus found that transcriptional heterogeneity enables cells with differential drug sensitivity to co-exist in the same population. Our work provides a unique resource and a novel framework to determine tumour heterogeneity and drug response in breast cancer patients.

cancer biology↗

bioRxiv2020

β-catenin perturbations control differentiation programs in mouse embryonic stem cells

The Wnt/{beta}-catenin pathway is involved in development, cancer and embryonic stem cell (ESC) maintenance; its dual role in stem cell self-renewal and differentiation is still controversial. Here, by applying an in vitro system enabling inducible gene expression control, we report that moderate induction of transcriptionally active exogenous {beta}- catenin in {beta}-catenin null mouse ESCs promotes epiblast-like cell (EpiLC) derivation in vitro. Instead, in wild type cells moderate chemical pre-activation of the Wnt/{beta}-catenin pathway promotes EpiLC in vitro derivation. Finally, we suggest that moderate {beta}- catenin levels in {beta}-catenin null mouse ESCs favour early stem cell commitment towards mesoderm if the exogenous protein is induced only in the ground state of pluripotency condition, or endoderm if the induction is maintained during the differentiation. Overall, our results confirm previous findings about the role of {beta}-catenin in pluripotency and differentiation, while indicating a role for its doses in promoting specific differentiation programs.

cell biology↗

bioRxiv2019

O-linked mucin-type glycosylation regulates the transcriptional programme downstream of EGFR in breast cancer

Aberrant mucin type O-linked glycosylation is a common occurrence in cancer. This type of O-linked glycosylation is not limited to mucins but can occur on many cell surface glycoproteins where only a small number of sites may be present. Upon EGF ligation, EGFR induces a signaling cascade but can also translocate to the nucleus where it can directly regulate gene transcription. Here we show that upon EGF binding, human breast cancer cells carrying different O-linked glycans respond by transcribing different gene expression signatures. This is not a result of changes in signal transduction but due to the differential nuclear translocation of EGFR in the two glyco-phenotypes. This is regulated by the formation of an EGFR/galectin-3/MUC1/{beta}-catenin complex at the cell surface that is present in cells carrying short core-1-based O-glycans characteristic of tumour cells but absent in core-2-carrying cells.

cancer biology↗