Selective Retention and Adaptive Reuse of Biological Information
A Computational Analogy for Persistence, Genetic Loss, and Evolutionary Change
John Swygert
November 16, 2026
DOI: To be assigned
Abstract
This companion conceptual paper explores an analogy between evolutionary change and the maintenance of a computational system that retains useful code, discards costly or dispensable components, and develops new functions under altered operating conditions. The analogy motivates a distinction between material persistence, hereditary retention, functional relevance, and subsequent innovation. Biological evolution does not possess an intentional code auditor or foresight: deletion, duplication, recombination, mutation, and regulatory changes generate variation, while selection and drift influence its fate. We propose comparative tests of whether jointly tracking retained, lost, reactivated, and newly functional sequences improves predictions of descendant outcomes beyond established models. The paper complements “Biological Persistence Through Destruction: Survival, Selection, and the Evolutionary Expression of Genetic Information” without claiming a new molecular mechanism.
1. Relation to the companion paper
The companion paper, “Biological Persistence Through Destruction: Survival, Selection, and the Evolutionary Expression of Genetic Information” (Swygert, 2026), separates physical persistence P, conditional functional recovery F, and conditional heritable transmission T after disturbance. It asks whether independently measuring those stages improves prediction over viable-survivor and fitness baselines. The present paper considers a related but distinct question: once biological material or lineages persist, how do retention, loss, reuse, and new function alter the information represented in later generations? The two papers should be read together, but neither requires the other to establish a new evolutionary force.
2. The software-maintenance analogy
Imagine a computational agent inheriting an old codebase built for a former operating environment. Some modules are essential now; some are dormant but potentially useful under future conditions; some impose costs; and others are redundant. A deliberate engineer might inspect dependencies, remove unused modules, retain robust components, and write replacements. Biological systems have no corresponding general-purpose agent choosing which DNA or RNA to keep. Rather, variants arise through molecular processes and their frequencies change through inheritance, selection, drift, and other established evolutionary forces. The analogy is therefore a guide to classifying outcomes, not a literal account of biological agency.
3. Four fates of inherited information
Retention: a sequence or regulatory arrangement remains in descendants, whether or not its present function is known. Loss: deletions, genome reduction, or lineage extinction remove material from a descendant lineage. Reuse: an existing sequence or regulatory element acquires a new context or function, including co-option (exaptation). Innovation: duplication, recombination, sequence change, and regulatory evolution can yield new or modified functions. These categories can overlap and must be assigned using pre-specified evidence. An apparently unused region is not automatically dispensable; it may have context-dependent, regulatory, structural, or as-yet-unmeasured effects.
4. Environmental change and biological directionality
A changing environment can alter which variants have higher reproductive success. This can produce a directional population-level response without implying that mutations are designed for future needs. Stress can affect DNA damage, repair, mutation rates, gene expression, and horizontal transfer, but whether a particular mutation is advantageous is established by its effect in a defined context. Historical contingency also matters: different populations can arrive at different solutions under similar pressures. Thus, environmental change influences the selection and expression of possibilities; it does not generally instruct genomes to write a specific useful sequence.
5. A testable retention–reuse framework
For each tracked variant or sequence class j, distinguish R_j: retention in the post-disturbance heritable pool; A_j(E): measured activity or functional contribution in recovery environment E; H_j(E): heritable representation among descendants; and N_j(E): evidence of newly acquired or repurposed function relative to the preregistered ancestral baseline. These are descriptive observables, not independent probabilities by definition, and must not be multiplied without checking conditional dependence. The candidate analysis will treat them as potentially correlated indicators; dependence will be assessed using preregistered joint-frequency summaries and model diagnostics rather than assumed away. For N_j(E), a newly acquired function requires a prespecified change relative to the ancestral sequence or regulatory baseline, a reproducible phenotypic effect in environment E, and evidence that the relevant change or functional state is heritably represented. Repurposing of an unchanged inherited sequence instead requires a documented shift in functional context, a measurable effect, and heritable association; sequence mutation is not required for co-option. In particular, the detection of a fragment does not establish viability, and the loss of a sequence does not prove that it was useless. The hypothesis to test is whether adding separately measured loss, reactivation, and reuse categories improves held-out prediction beyond survivor abundance, fitness, and known genetic-change covariates.
6. Experimental comparison
Use a nonpathogenic, genetically characterized model organism and established sequence-tracking methods. Compare matched populations under stable versus changed recovery environments, with independent replication and preregistered sampling windows. Measure genetic presence, expression, functional effects, and descendant frequencies separately. Use suitable negative controls, account for detection limits, and classify putative loss or reuse only after confirmatory genetic and phenotypic evidence. For loss, distinguish verified sequence deletion from nondetection and from loss of an entire lineage; for reuse, document the ancestral function or context, the altered context, a measurable functional effect, and its heritable association. A primary comparator should include initial and surviving variant frequencies, measured fitness, and established population-genetic predictors. A candidate model adds preregistered retention/loss/reuse indicators. Evaluate both using the same held-out populations, proper predictive scoring, uncertainty intervals, and a prespecified minimum meaningful gain. No gain means the added classification has not established incremental predictive value.
7. What the analogy does and does not predict
Prediction A: sequences detectable after a disturbance but not functionally or heritably represented will contribute little to descendant outcomes; this is largely expected under established biology. Prediction B: sequences with minimal function in one environment may become consequential after environmental change, including through regulatory shifts or co-option. Such potential may already be present as standing or cryptic genetic variation and need not arise from a new mutation after disturbance. Prediction C: accounting for measured functional reactivation and loss may improve prediction of lineage composition over abundance-only baselines, but any claimed advance must also exceed survival-plus-fitness models. These are hypotheses for comparative measurement, not evidence that organisms optimize genomes as software engineers do. Genome reduction can reflect selection, drift, mutational bias, or population history; streamlining for efficiency is not presumed to be the default explanation, nor is any intentional drive toward efficiency implied.
8. RNA, destruction, and preservation
The initiating intuition includes the possibility that environmental extremes destroy much of a biological system while leaving particular RNA or other molecular fragments. Such persistence can matter evolutionarily only where a credible route to replication, transfer, or heritable incorporation exists. RNA is often less chemically stable than DNA, and isolated RNA fragments are not presumed to survive longest or to direct future mutation. The companion paper’s distinction between molecular integrity, viable recovery, and heritable transmission therefore remains essential. Water activity, salinity, and ionic conditions may influence these stages differently, without implying a single optimal environment for all life.
9. Limits, novelty, and conclusion
Genome streamlining, gene loss, exaptation, regulatory evolution, standing variation, and historical contingency are established research topics. This manuscript claims no discovery of those phenomena. Its possible contribution is a disciplined analogy and a proposed comparative measurement scheme linking survival of biological information to later loss, reuse, and innovation. Whether that scheme offers predictive value beyond existing evolutionary models is an empirical question. The guiding distinction is simple: what remains after change is not necessarily what is functional; what is functional now is not necessarily what will remain useful; and new function can emerge from inherited material without foresight or design.
Selected scientific context
Swygert, J. (2026). Biological Persistence Through Destruction: Survival, Selection, and the Evolutionary Expression of Genetic Information. Companion conceptual manuscript.
Gould, S. J., & Vrba, E. S. (1982). Exaptation—a missing term in the science of form. Paleobiology, 8, 4–15.
Lynch, M. (2007). The Origins of Genome Architecture. Sinauer Associates.
Lenski, R. E. et al. (1991). Long-term experimental evolution in Escherichia coli. I. Adaptation and divergence. American Naturalist, 138, 1315–1341.
Blount, Z. D., Borland, C. Z., & Lenski, R. E. (2008). Historical contingency and the evolution of a key innovation in an experimental population of Escherichia coli. PNAS, 105, 7899–7906.
No comments:
Post a Comment