DevKidCC allows for robust classification and direct comparisons of kidney organoid datasets

Area of Science:

• Developmental Biology
• Stem Cell Biology
• Bioinformatics

Background:

• Accurate cell classification in single-cell transcriptional profiling is challenging, especially in complex systems like pluripotent stem cell-derived organoids.
• Kidney organoids are valuable models for studying kidney development and disease, but comparing different protocols is difficult due to analytical variations.
• Existing clustering algorithms and marker genes often fail to precisely identify cellular identities in organoid datasets.

Purpose of the Study:

• To develop an unbiased computational approach for accurate cell classification in kidney organoids.
• To enable robust comparison of cellular composition across different kidney organoid differentiation protocols.
• To identify variations in cell type proportions and maturation states influenced by experimental protocols.

Main Methods:

• Trained an R package, scPred, using human fetal kidney single-cell RNA-seq data to create a hierarchical classification model.
• Developed the DevKidCC R package incorporating this model to predict cell identity in published kidney organoid datasets.
• Utilized custom functions within DevKidCC for differential gene expression analysis and data visualization.

Main Results:

• DevKidCC successfully classified cellular subtypes (nephron, stroma, ureteric epithelium) across diverse kidney organoid datasets.
• Identified significant differences in cell type proportions, including stromal and unassigned cells, and nephron progenitor prevalence between protocols.
• Distinguished between closely related cell types like distal nephron and ureteric epithelium, and detected protocol-specific effects, such as retinoic acid's depletion of nephron progenitors.

Conclusions:

• DevKidCC provides a reproducible method for classifying cellular identity within single-cell datasets of kidney organoids.
• The tool facilitates consistent and rapid comparison of kidney organoid protocols, aiding in optimizing differentiation strategies.
• Enables validation of new approaches and drives improvements in generating kidney organoids for research and therapeutic applications.