Zakeri, I. | bioRxiv Science

bioRxiv2024

Imbalance and Composition Correction Ensemble Learning Framework (ICCELF): A novel framework for automated scRNA-seq cell type annotation

Single-cell RNA sequencing (scRNA-seq) has gained broad utility and success in revealing novel biological insight in preclinical and clinical investigations. Cell type annotation remains a key analysis task with great influence on downstream interpretation of scRNA-seq data. Traditional machine learning approaches proposed for automated cell type annotation often overlook the inherent imbalance of cell type proportions within biological samples, and the compositional nature of sequencing-based gene expression quantification. In this study, we highlight the importance of accounting for cell type imbalance and compositionality of sequencing count data, and introduce the Imbalance and Composition Corrected Ensemble Learning Framework (ICCELF) as a novel approach to automated cell type annotation. We show via comprehensive evaluation on both simulated and real-world scRNA-seq data that by effectively addressing class imbalance and data compositionality. ICCELF offers a robust and efficient solution that facilitates accurate and reliable cell type annotation, paving the way for enhanced biological discoveries.

bioinformatics↗

bioRxiv2023

Mcadet: a feature selection method for fine-resolution single-cell RNA-seq data based on multiple correspondence analysis and community detection

Single-cell RNA sequencing (scRNA-seq) data analysis faces numerous challenges, including high sparsity, a high-dimensional feature space, technical biases, and biological noise. These challenges hinder downstream analysis, necessitating the use of feature selection methods to address technical biases, identify informative genes, and reduce data dimensionality. However, existing methods for selecting highly variable genes (HVGs) exhibit limited overlap and inconsistent clustering performance across benchmark datasets. Moreover, these methods often struggle to accurately select HVGs from fine-resolution scRNA-seq datasets and rare cell types, raising concerns about the reliability of their results. To overcome these limitations, we propose a novel feature selection framework for unique molecular identifiers (UMIs) scRNA-seq data called Mcadet. Mcadet integrates Multiple Correspondence Analysis (MCA), graph-based community detection, and a novel statistical testing approach. To assess the effectiveness of Mcadet, we conducted extensive evaluations using both simulated and real-world data, employing unbiased metrics for comparison. Our results demonstrate the superior performance of Mcadet in the selection of HVGs in scenarios involving fine-resolution scRNA-seq datasets and datasets containing rare cell populations. By addressing the challenges of feature selection in scRNA-seq analysis, Mcadet provides a valuable tool for improving the reliability and accuracy of downstream analyses in single-cell transcriptomics.

bioinformatics↗