Traditional mutation models often apply simplistic substitution matrices or random walk heuristics. The CAS framework, however, integrates a multi-factorial probability model that adjusts mutation likelihood based on:
Local structural context (buried vs. exposed residues),
Functional significance (W) of the residue or domain,
Thermodynamic tolerability (G threshold),
Systemic stability feedback (S) from previous simulation cycles.
This logic is formalized as:
P(mi)=f(Wi)f(Si)1+eGi/kTP(m_i) = \frac{f(W_i) \cdot f(S_i)}{1 + e^{\Delta \Delta G_i / kT}}
Where:
P(mi)P(m_i) is the mutation probability at residue i,
f(Wi)f(W_i) is a function assigning weight-based mutation flexibility (highly functional sites mutate less),
f(Si)f(S_i) maps current systemic robustness to local mutability (less stable systems restrict mutation),
