This preprint (bioRxiv-style, Aug 2026) constructs a five-component bottom-up accounting model for the energy cost of building 1 kg of wet human skeletal muscle: stored tissue energy (5,668.5 kJ), biochemical synthesis cost (+670 kJ to 6,338 kJ), physiological deposition cost (4,488 kJ, at kP = 46%), resting maintenance during 84-day accretion (+2,285 kJ), and diet-induced thermogenesis (+1,457 kJ), yielding a central additional metabolizable energy intake of 14,570 kJ/kg (3,481 kcal/kg), range 13,410β15,520 kJ/kg .
kP dominates and is borrowed. Protein deposition efficiency (46%, range 42β52%) comes from infants and growing animals β no direct adult human data exist. The authors acknowledge this is the main uncertainty, but it is more than a caveat: roughly 90% of the deposition-cost component scales with kP. A kP of 30% would push the final estimate toward ~20,000 kJ/kg; a kP of 60% would pull it near 11,000 kJ/kg. The stated range likely understates true uncertainty.
Circularity risk in composition. The fixed composition (177 g protein/kg) excludes processes like connective-tissue and capillary remodeling that accompany real hypertrophy; these are absorbed implicitly into kP without independent verification.
No primary data or validation. The paper is pure calculation with zero new measurements; its triangulation studies (rats, cultured meat) differ in species and system. It is best read as a well-reasoned Fermi estimate β useful as a reference point, not as settled physiology .
A genuinely useful first quantitative reference with exemplary assumption transparency, but with a dominant borrowed parameter (kP) that means the central value should be treated as an order-of-magnitude anchor, not a precise constant. Would be strengthened by: adult-human deposition efficiency trials, Bayesian propagation of uncertainty beyond the kP range, and muscle-composition sensitivity analyses.
Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.