New python toolkit automates single-cell lineage tracing, accelerates tumor drug resistance analysis

Researchers at Tsinghua University have released a novel Python toolkit, scLT-kit, which automates the processing and analysis of single-cell lineage tracing data, delivering clear insights into how individual cells develop, differentiate, and respond to treatments. Think of scLT-kit as a “GPS for cell genealogies”—it takes messy, high-dimensional data and plots each cell’s journey, making complex lineage relationships easy to follow. This tool addresses the growing need for flexible, user-friendly software to handle complex single-cell datasets.

Novel Model to Decode Cell Dynamics Energises Regenerative Medicine and Cancer Research

Understanding how cells change over time is crucial for fields ranging from regenerative medicine to cancer therapy. By making lineage tracing analyses more accessible, scLT-kit can accelerate research into tissue development, drug resistance, and disease progression. Its streamlined workflows could help biologists, pharmaceutical developers, and policymakers better evaluate how cells respond to treatments, ultimately informing new strategies for diagnosis and therapy.

“With scLT-kit, we’ve improved a previously labor-intensive, bespoke analysis into a seamless, reproducible workflow,” says Prof. Jin Gu. “Our goal was to empower every lab—whether focused on development, drug resistance, or disease progression—to extract lineage insights from single-cell data in minutes, not months.”

Automated Barcode Analysis Uncovers Predictable Developmental Clones—and Unruly Drug-Resistant Tumor Cells

Through a series of analyses, the researchers uncovered the following insights:

• scLT-kit reliably processes time-series single-cell RNA-seq data tagged with heritable barcodes, providing rapid quality checks and summary statistics of barcoding efficiency and clone sizes.
• In developmental datasets (e.g., blood progenitors and C. elegans embryos), cells sharing the same barcode showed higher similarity in gene expression than unrelated cells.
• In tumor cell lines treated with EGFR inhibitors (osimertinib or erlotinib), this within-clone similarity was less pronounced, reflecting the high heterogeneity of cancer persisters.
• The toolkit computes measures of cell fate diversity, showing that normal development features more predictable outcomes, whereas drug-treated cancer cells exhibit greater randomness in fate decisions.
• scLT-kit identifies subpopulations with distinct fate trajectories and uncovers genes linked to these fates through differential expression analysis.

All-in-One Python Package Delivers Barcode QC, Lineage Networks and Interactive Sankey Plots

scLT-kit combines standard single-cell RNA-seq processing with two specialized modules for lineage data. The scLT-statistics module calculates barcoding fractions and clone sizes at each time point. The scLT-analysis module builds lineage-based networks to infer how clusters of cells transition over time, visualizing dynamics with Sankey plots. Four quantitative indicators assess cell fate randomness and neighbor similarity. Finally, the package uses established statistical tests to link gene expression changes to specific fate outcomes, all within an easy-to-install Python package available on PyPI and GitHub.

Streamlined Workflows Lower the Barrier to Single-Cell Lineage Tracing in Development and Disease

scLT-kit brings robust, automated workflows to single-cell lineage tracing studies, lowering the barrier for labs to explore cell dynamics in development and disease. By integrating data quality checks, dynamic analysis, and gene-level insights, this toolkit promises to advance our understanding of how cells make fate decisions under both normal and perturbed conditions. The full study was published in Frontiers of Computer Science in April 2025 (https://doi.org/10.1007/s11704-025-41249-9).