# ALFA-K: Local adaptive mapping of karyotype fitness landscapes

## TL;DR

The paper introduces ALFA-K, a new method to estimate karyotype fitness and predict future aneuploid karyotypes in tumor evolution.

## Contribution

ALFA-K is a novel method for inferring chromosome-level karyotype fitness landscapes from single-cell data.

## Key findings

- Whole genome doubling narrows the spectrum of deleterious copy-number changes, facilitating aneuploidy evolution.
- Environmental context and cisplatin treatment modulate the fitness impact of copy-number changes.
- Chromosome mis-segregation rates influence the predominant karyotypes in evolving populations.

## Abstract

Despite its critical role in tumor evolution, a detailed quantitative understanding of the evolutionary dynamics of aneuploidy remains elusive. Here we introduce ALFA-K (Adaptive Local Fitness landscapes for Aneuploid Karyotypes), a method that infers chromosome-level karyotype fitness landscapes from longitudinal single-cell data. ALFA-K estimates fitness of thousands of karyotypes closely related to observed populations, enabling robust prediction of emergent karyotypes not yet experimentally detected. We validate ALFA-K’s performance using synthetic data from an agent-based model and empirical data from in vitro and in vivo passaged cell lines. Analysis of fitted landscapes suggests several key insights: (1) Whole genome doubling facilitates aneuploidy evolution by narrowing the spectrum of deleterious copy-number changes; (2) Environmental context and cisplatin treatment significantly modulate the fitness impact of these changes; (3) Fitness effects of copy-number changes depend on parental karyotype; and (4) Chromosome mis-segregation rates strongly influence the predominant karyotypes in evolving populations.

The evolutionary dynamics of aneuploidy in solid tumors are challenging to study. Here the authors introduce a method, ALFA-K, which estimates karyotype fitness and predicts emergent karyotypes before experimental detection, and test its performance on synthetic and empirical data.