MEDALT

MEDALT reconstructs evolutionary lineages of cell populations from single-cell copy number (SCCN) profiles to trace aneuploidy-driven tumor evolution.

Key Features:

• Minimal Event Distance Aneuploidy Lineage Tree (MEDALT): Implements a minimal event distance framework to model copy number changes between single cells.
• Lineage tracing via SCCN profiles: Uses single-cell sequencing-derived copy number profiles to infer lineage relationships among cells.
• Rooted Directed Minimal Spanning Tree (RDMST): Constructs an RDMST to represent directed evolutionary trajectories of aneuploid cells.
• Identification of copy number alterations: Detects focal and broad copy number alterations associated with lineage expansion.
• Lineage Speciation Analysis (LSA): Provides a statistical routine to discover fitness-associated alterations and genes within SCCN lineage trees.
• Enhanced accuracy over traditional phylogenetics: Demonstrates improved accuracy in reconstructing copy number lineages compared to traditional phylogenetic approaches.

Scientific Applications:

• Gene discovery: Identifies fitness-associated genetic alterations and candidate genes from SCCN lineage trees.
• Cancer research: Prioritizes genes essential for breast cancer cell fitness and, in a cohort of 20 triple-negative breast cancer patients, predicted patient survival and identified convergent evolutionary events.
• Predictive modeling: Infers lineage evolution to provide predictive insights into tumor progression and potential therapeutic targets.

Methodology:

Analyzes single-cell copy number (SCCN) profiles, applies a minimal event distance framework to construct a rooted directed minimal spanning tree (RDMST), and uses Lineage Speciation Analysis (LSA) to identify focal and broad copy number alterations associated with lineage expansion and fitness.