Biology subjects

Ribas, G. T.

Publications and source records attributed to Ribas, G. T..

3 recordsLinked to original sources

bioRxiv2026

Housekeeping Gene Expression Normalization in Transcriptomics Mitigates Data Leakage in Machine Learning Models

BackgroundInappropriate normalization can lead to data leakage and overfitting in machine learning models. Accurately identifying housekeeping genes (HKGs) can overcome this problem and is crucial for normalizing gene expression data, particularly in RNA-Seq experiments. ResultsFirst, we demonstrate that the gene expression of commonly used HKGs significantly changes over time due to immunosuppressive treatments in transplant recipients. Using large public transcriptomic datasets of kidney transplantation, we developed a pipeline based on the genes coefficient of variation, stability, and Gini coefficient, and identified nine stable and better-suitable HKG candidates. Our results demonstrate that these HKGs improve the robustness and generalizability of machine learning models by minimizing data leakage, as evidenced by superior performance compared to benchmark methods like median ratio normalization and trimmed mean of M values. ConclusionsThis approach enables more accurate comparison of gene expression datasets across different clinical scenarios, improving the reliability of biomarker identification and enhancing personalized treatment strategies.

bioinformatics↗

bioRxiv2024

a2iHelper: a Python toolkit for a differential editing site analysis of RNA-Seq data

BackgroundA-to-I RNA editing, mediated by ADAR enzymes, plays a crucial role in cancer and autoimmune disorders but lacks standardized tools for differential analysis. After reviewing 55 studies, it highlights significant methodological heterogeneity, hindering result comparability and reproducibility. To address this, we developed a2iHelper, a Python package that streamlines RNA editing analysis by filtering noise, performing statistical analyses, and generating visualizations. a2iHelper integrates seamlessly with Python machine learning tools, aiming to standardize and simplify RNA editing research. ResultsWe applied our methods to analyze A-to-I editing in a public dataset comparing wild-type and ADAR knockout. The source code is open, freely available on GitHub, and organized in a well-documented Python package. Using Snakemake for preprocessing, we conducted differential editing analysis on the top 104 most edited genes. The results include p-values from statistical tests, Odds ratios for Manhattan plots, and correlations between editing frequencies and gene expression, visualized in various plots. ConclusionsWe developed a2iHelper, a Python-based package for analyzing and visualizing A-to-I RNA editing data. It allows researchers with minimal programming experience to perform organized, reproducible editing analyses. Novice users can easily detect editing sites, filter noise, and generate figures, while advanced users can integrate functionalities into their workflows. a2iHelper runs on personal computers without needing High-Performance Computing resources.

bioinformatics↗

bioRxiv2023

Immune Response of Transplanted Kidney Tissues Assembled from Organoid Building Blocks

The increasing scarcity of organs and the significant morbidity linked to dialysis requires the development of engineered kidney tissues from human-induced pluripotent stem cells. To accomplish this, integrative approaches that synergize scalable kidney organoid differentiation, tissue biomanufacturing, and comprehensive assessment of their immune response and host integration are essential. Here, we create engineered human kidney tissues composed of kidney organoid building blocks (OBBs) and transplant them into mice reconstituted with allogeneic human immune cells. We assess their host vascular integration, in vivo maturation, and their ability to trigger human immune responses. Tissue-infiltrating human immune cells are composed of effector T cells and innate cells. This immune infiltration leads to kidney tissue injury characterized by reduced microvasculature, enhanced kidney cell apoptosis, and a unique inflammatory gene signature comparable to kidney organ transplant rejection in humans. Upon treatment with the immunosuppressive agent Rapamycin, the induced immune response is greatly suppressed. Our model serves as a translational platform to study engineered kidney tissue immunogenicity and develop novel therapeutic targets for kidney rejection.

bioengineering↗