Repair-Mediated Path Filtering of Mutation Load in Multilocus Homeostatic Populations

Authors

Hardisty, B.

Abstract

We study a nonlinear multilocus mutation--selection--repair system for an asexual population maintained at exact homeostasis by state-dependent reproductive feedback. Each of $n$ identical biallelic damage loci mutates with probability $U$ at reproduction and is repaired at rate $nu$, with per-locus reproduction and mortality factors $zeta_rho<1$ and $zeta_gamma>1$. Permutation symmetry and feedback through damage-class frequencies reduce the $2^n$-state dynamics exactly to an $(n+1)$-class ODE system on a forward-invariant simplex. Without repair, the nonlinear equilibrium is obtained in closed form throughout the branch on which the undamaged class persists. The microstate distribution factorizes as a Bernoulli product measure, the damage number is binomially distributed, and the branch terminates at $U_c=1-zeta_rho/zeta_gamma$, independently of $n$. Under $ntoinfty$, $Uto0$, and $nUtoTheta$, the damage number converges to a Poisson distribution with mean $Theta/(zeta_gamma-zeta_rho)$. With repair, we prove existence and local asymptotic stability of an analytic rare-mutation equilibrium branch and derive a recursion for the factor $E_k$ suppressing the leading mutation load of class $k$. Its source weights follow a conditional $mathrm{Binomial}(k,zeta_rho/zeta_gamma)$ law independent of $n$ for every $nge k$. The resulting suppression lies between the product $prod_{j=1}^k S_j$, corresponding to sequential accumulation through every intermediate repair filter, and the terminal filter $S_k$ alone, corresponding to direct multilocus jumps that bypass them, where $S_j=lambda_j/(lambda_j+jnu)$ quantifies the balance between effective demographic depletion and repair. Repair also generates positive leading-order association among loci, breaking the no-repair product structure. Numerical integration of the full nonlinear system confirms the asymptotic predictions.

Preprint server: bioRxiv
The authors list and abstract were imported from bioRxiv on 17 Oct 2026.

Stats

1-terrible, 9-excellent. How would you rate this preprint? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.