UV light leaves dimers in DNA. Searchers (XPC) walk the strand. When one stops, a verifier (TFIIH) checks the spot in steps, and every step burns ATP. A real lesion passes and is cut out. An intact spot should fail; when it passes anyway, the cell cuts good DNA. The energy spent per correct repair is the cost of finding the owner.
Time-Gel clock shift per repairing cell (estimate from the application note). Far below anything measurable. This simulator does not test Time-Gel.
Entropy produced by repair: σ ≈ rATP·|ΔGATP|/T. One ATP ≈ 20 kBT at 37 °C, so each ATP here adds about 20 kB of entropy.
What is a choice here. f = 0.2 is chosen so errors are visible; real NER discrimination is far sharper. Steps and timings are compressed. The shape of the result is the point: with no energy (g = 0) extra checks buy nothing; each energised step multiplies the error down by fg. Kinetic proofreading (Hopfield 1974, Ninio 1975; for NER, Reardon & Sancar 2003) predicts the same outcomes without Time-Gel. Paper: Information Looking for Its Owner in DNA Repair v1.0.
Model output, not experimental evidence. This page does not represent the views of Aziz Sancar or his laboratory.