| Alzheimer's disease is a complex and devastating neurological disorder, and there is a pressing need to discover effective treatments. Analyses in specific brain regions can provide valuable insights into the underlying pathology of the disease. Molecular biology techniques, such as single-cell analysis, can offer an in-depth analysis of cellular-level changes in the brain in Alzheimer's disease. However, integrating single-cell RNA sequencing (scRNA-seq) data from different studies can be challenging due to batch effects, which can lead to spurious results. The dominant approach for addressing this issue is SCALEX, an online single-cell data integration method that projects heterogeneous datasets into a common cell-embedding space. In this paper, we highlight the impact of SCALEX in detecting accurate biomarkers in the context of significant genes that are leading scRNA-seq data. We demonstrate the pitfalls of traditional data integration methods and show the protection offered by SCALEX in preserving the biological heterogeneity of the sample while minimizing the impact of technical artifacts introduced by batch effects. Our results show that integrating scRNA-seq data with SCALEX can lead to more accurate biomarkers and significant genes, offering insights into the underlying pathology of Alzheimer's disease. These findings have important implications for the development of new therapies and treatments for this devastating disease. |
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.