MEDALT: single-cell copy number lineage tracing enabling gene discovery

Area of Science:

• Computational Biology
• Genomics
• Cancer Research

Background:

• Understanding cancer evolution is crucial for developing effective treatments.
• Single-cell copy number (SCCN) profiles offer insights into cellular heterogeneity and evolution.
• Existing phylogenetic methods may have limitations in reconstructing copy number lineage.

Purpose of the Study:

• To introduce a novel algorithm, Minimal Event Distance Aneuploidy Lineage Tree (MEDALT), for inferring cell population evolution history from SCCN profiles.
• To present a statistical routine, lineage speciation analysis (LSA), for discovering fitness-associated alterations and genes from SCCN lineage trees.
• To validate the accuracy and utility of MEDALT and LSA in a triple-negative breast cancer (TNBC) cohort.

Main Methods:

• Development and application of the MEDALT algorithm for lineage reconstruction.
• Implementation of lineage speciation analysis (LSA) for identifying key genetic alterations.
• Analysis of SCCN data from 20 triple-negative breast cancer patients.

Main Results:

• MEDALT demonstrates higher accuracy in reconstructing copy number lineage compared to traditional phylogenetics approaches.
• LSA effectively prioritizes genes critical for breast cancer cell fitness.
• The study identified genes implicated in convergent evolution and their association with patient survival.
• The developed methods successfully predicted patient survival outcomes in the TNBC cohort.

Conclusions:

• MEDALT and LSA provide a powerful framework for analyzing cancer evolution from SCCN data.
• These computational tools can identify novel therapeutic targets and biomarkers for triple-negative breast cancer.
• The findings highlight the importance of convergent evolution in driving cancer progression and patient outcomes.