Category: Pop. Genetics

Summary: Finding the mutation level where a modular quasispecies delocalizes on correlated rugged fitness landscapes.

The classic quasispecies error threshold marks when mutation overwhelms selection, but real fitness landscapes are often structured and rugged rather than analytically simple. This experiment asks a more specific question: on a block-modular correlated landscape, at what mutation rate does the leading population state stop concentrating in one module and spread out across genotype space?

The script constructs dense symmetric mutation-selection operators with modular correlations and then bisects mutation strength while increasing system size through iterative deepening. The threshold is defined through localization statistics rather than only total fitness.

That focus makes the output especially relevant to structured genotype spaces, where the first sign of breakdown may be loss of modular concentration before any simpler scalar criterion becomes obvious. It is a finite-size numerical map of a modular error threshold.

Method: Dense symmetric eigensolve with iterative deepening and bisection on mutation strength in a modular mutation-selection operator.

What is measured: Critical mutation threshold, localization score, peak module mass, system size reached, bracket width, and eigensolve count.