Biology subjects

Giakountis, A.

Publications and source records attributed to Giakountis, A..

3 recordsLinked to original sources

bioRxiv2025

lncAPNet enables the deciphering of lncRNA and mRNA connections in patient transcriptomic data

MotivationLong non-coding RNAs (lncRNAs) regulate gene expression through chromatin remodeling, transcriptional control, and post-transcriptional modulation, influencing physiological cell homeostasis but also disease onset. Yet most transcriptomic and network-based studies rely on descriptive linear co-expression analyses, missing nonlinear and mechanistic insights. Emerging ML/DL methods offer promise but remain limited by data sparsity, noise, insufficient biological priors, and poor interpretability, constraining systems-level lncRNA-mRNA motif discovery. ResultsIn this manuscript, we introduce lncAPNet, an extended version of APNet workflow, which integrates graph-based nonlinear inference of lncRNA-mRNA interactions using NetBID2s activity logic within an lncRNA-focused SJARACNe co-expression network, coupled with PASNet, a biologically informed sparse deep learning model. This framework enables explainable identification of lncRNA drivers in two different cancer type case studies, Chronic Lymphocytic Leukemia (CLL) and Prostate Adenocarcinoma (PRAD), uncovering lncRNA drivers that illuminate lncRNA-mediated programs in cancer progression. Availability and implementationlncAPNets R scripts, Python scripts, and methodologies are available at github repository: https://github.com/BiodataAnalysisGroup/lncAPNet

bioinformatics↗

bioRxiv2025

STExplorer: Navigating the Micro-Geography of Spatial Omics Data

Spatial transcriptomics (ST) has the potential to provide unprecedented insights into gene expression across tissue architecture, but existing analytical methods often overlook the full complexity of the spatial dimension. We present STExplorer, an R package that adapts well-established computational geography (CG) methods to explore the micro-geography of spatial omics data. By incorporating techniques like Geographically Weighted Principal Component Analysis (GWPCA), Fuzzy Geographically Weighted Clustering (FGWC), Geographically Weighted Regression (GWR), and analyses of observation Spatial Autocorrelation (SA), STExplorer enables the uncovering of spatially resolved patterns that capture the spatial heterogeneity of biological data. STExplorer provides a complete suite of functions for spatial analyses and visualisations, supporting deeper biological understanding and inference. Built on the Bioconductor ecosystem, the package integrates with SpatialFeatureExperiment objects, ensuring compatibility with existing pipelines. It includes preprocessing capabilities such as data import, quality control, gene count normalisation, and variable gene selection, alongside tools for downstream analysis and detailed visualisations that quantify and map spatial heterogeneity and relationships. We demonstrate the utility of STExplorer through applications to spatial transcriptomics datasets, revealing that spatially varying gene expression and relationships are often masked by standard analyses. By bridging bioinformatics and CG, STExplorer provides a novel and informed approach to spatial transcriptomics analysis, with robust tools to address spatial heterogeneity and its associated underlying biology, thereby advancing our understanding of complex tissue biology without reinventing the wheel.

bioinformatics↗

bioRxiv2022

A species-specific lncRNA modulates the reproductive ability of the Asian tiger mosquito

Long non-coding RNA (lncRNA) research has emerged as an independent scientific field in recent years. Despite their association with critical cellular and metabolic processes in plenty of organisms, lncRNAs are still a largely unexplored area in mosquito research. We propose that they could serve as exceptional tools for pest management due to unique features they possess. These include low inter-species sequence conservation and high tissue specificity. In the present study, we investigated the role of ovary-specific lncRNAs in the reproductive ability of the Asian tiger mosquito, Aedes albopictus. Through the analysis of transcriptomic data, we identified several lncRNAs that were differentially expressed upon blood feeding; we called these genes Norma (NOn-coding RNA in Mosquito ovAries). We observed that silencing some of these Normas resulted in significant impact on mosquito fecundity and fertility. We further focused on Norma3 whose silencing resulted in 43% oviposition reduction and 53% hatching reduction of the laid eggs, compared to anti-GFP controls. Moreover, a significant downregulation of a neighboring ([~]100 Kb) mucins cluster was observed in smaller anti-Norma3 ovaries, indicating a potential mechanism of in-cis regulation between Norma3 and the mucins. Our work constitutes the first experimental proof-of-evidence connecting lncRNAs with mosquito reproduction and opens a novel path for pest management.

molecular biology↗