Work

Decoding assumptions and inference about a robustness preference in codon reassignments along sampled histories

Year: 2026
Type: preprint
Abstract: Catalogue-associated net transitions between genetic codes reach the silent third-position packing maximum more often than uniform targets do. Whether that signals selection depends on which targets decoding made accessible and which reassignments were detected. We prove that from a block-greedy code every single-box reassignment to an amino acid already in the codon’s box keeps the code optimal, that a robustness coefficient in a conditional-logit target-choice model is identified exactly when the within-pool contrasts span it, and that the class-conditional null stays exact when the decoding state is uncertain. On 200 sampled net-transition histories, the excess over uniform targets persists (35.5 against 20.0 units per history), but a recording ratio of 1.85 for optimum-attaining units would bring its merged reference statistic above 0.05. Within accessibility classes no repertoire’s test rejects once restrictive decoding states are treated as uncertain, and in conditional simulations pooled accessibility with no preference makes the fixed-state restrictive rule reject in all 60 replicates. In pointwise comparisons with fitted sequential generators, the observed statistic has upper-tail ranks below 0.05 for common preferences of odds 2, 2.5 and 3 under a restrictive repertoire (at most 0.025) and of odds 2.5 and 3 under pooled wobble rules, and above 0.05 at restrictive odds 1.5 (0.099) and pooled odds 2 (0.059); these ranks have no established exclusion coverage. Hierarchical intervals include zero under the primary priors. The data constrain several specified preference models but do not establish whether accessibility, a robustness preference or ascertainment generated the enrichment.
Language: English
Open Access status: green
OpenAlex ID: W7226014329
Cites: 23
Cited by:
Related to: 10