Information Looking for Its Owner in DNA Repair

Ultraviolet light writes a mark into DNA: two neighbouring pyrimidines fuse into a cyclobutane pyrimidine dimer (CPD). The mark is information without an owner. Somewhere among billions of intact base pairs, a set of proteins has to find it, confirm it, cut it out and rewrite the gap. This application note borrows a way of asking questions, not a result, from a protocol written for a different system, the freezing of supercooled water (doi:10.5281/zenodo.23724281), and applies it to nucleotide excision repair. Three things transfer: a separation of five roles (source, carrier, processor, state builder, meaning-giver); the idea that what matters is the irreversible act, not the state it leaves behind; and a control that spends the same energy with the same machinery but produces no record. In repair, that control is the full excision system working on undamaged DNA. One thing does not transfer: the clock effect that Time-Gel Theory attaches to irreversible events is about 10^-11 s per cell, far below any clock that could sit inside a cell. The note proposes a biochemical experiment that measures what it costs, in ATP, for information to find its correct owner, and whether paying less produces more wrong owners. Conventional kinetic proofreading, proposed for this step by Reardon and Sancar (2003), predicts the same outcomes; the note claims no discrimination. It then takes a first look at a public CPD Damage-seq time course in human fibroblasts (Hu et al., PNAS 2017). Thirty-six hours after UV, about 17% of the genome sits in 100 kb bins whose damage has not gone down at all; these bins form blocks with a mean length of about 560 kb and are concentrated on chromosomes 21, X, 13 and 18. This note does not test Time-Gel Theory. Every claim is labelled (MEASURED, OBSERVED, MODEL, HYPOTHESIS, DECISION, OPEN, INTERPRETATION, ANALOGY).