Research by the authors has demonstrated the effect of day/night temperature difference (Tdiff) on plant growth is as substantive as the effect of daily average temperature (DAT). Dependence of plant primary productivity on temperature cannot be assessed with fewer than two data per 24 hours. Thus, the same experimental approach was applied to time to anthesis in Delphinium cultorum Voss `Magic Fountains' and Stokesia laevis L. `White Parasols', and to survival in D. cultorum. Two hundred and seventy seedlings of D. cultorum and 72 plantlets of S. laevis were grown for 56 days in growth chambers under eighteen 12 hour day/12 hour night combinations of six day and six night temperatures (10, 15, 20, 25, 30, or 35 °C). Ninety plants of D. cultorum were harvested after 13, 34, or 56 days, and 36 plants of S. laevis after 34 or 56 days. For each event of interest (anthesis or death), one datum per plant was recorded, consisting of time elapsed when either the event occurred, or the plant was harvested, whichever came first. Each datum was paired with an indicator of whether the plant was harvested prior to the event being observed. Data were analyzed using time—to—event data analysis procedures. Several parametric distributions fitted the data equally well, and both day and night temperature had strong effects on time to anthesis and survival time. However, in contrast with biomass production, DAT was quite sufficient to account for timing of these developmental events in relation to temperature. Addition of Tdiff contributed marginally to the fit to the data, but the magnitude of the effect was considerably smaller. Within the range of temperatures likely to be encountered in cultivation, the effect was negligible.
A Fast and Accurate Method for Genome-Wide Time-to-Event Data Analysis and Its Application to UK Biobank
