Supplemental Material

This article contains the following supporting material:

• Figure S1 - Synchronization data for the pulse labeling experiments.

  A) Elutriation synchronized G2 cells were incubated with 500 nM BrdU and examined microscopically every 20 minutes for septation.
  B) Samples were taken 20 minutes for flow cytometry. We used a protocol that disrupts cell and releases whole nuclei, allowing us to observe the 1C G1 nuclei that are usually held together in 2C binucleate cells in fission yeast (Carlson et al., 1997). Most cells replicate between 140 and 180 minutes. We took samples for combing every 4 minutes from 130 to 170 minutes, and used 142 and 150 minutes for early and middle S phase.

• Figure S2 - Early-firing origin density.

  We estimated the density of origins that fire early by dividing the number of origin observed by the total length of DNA examined. If we include shorter fibers in this calculation, we over estimate the density of origin firing, because the shorter fibers have fewer large gaps. Therefore, we made the calculation for fibers of length greater than n, for each fiber of length n in the data set, and plotted the result versus n. The left axis is estimated early-firing origin per genome; the right axis is the average spacing between early-firing origins. For n = 250 kb, the estimated number of origins is about 150. This estimate decreases as shorter fibers are excluded and plateaus at about 120 for fibers greater than 600 to 700 kb. For fibers greater that 700 kb the data is noisy because of the small number of fibers.

• Figure S3 - Budding yeast origin firing is not randomly distributed.

  Data from budding yeast micro-array experiments have identified origins of replication (Raghuraman et al., 2001; Yabuki et al., 2002). We have reanalyzed this published data to examine the distribution of those origins. In addition, we have reanalyzed whole-genome HU-block combing data similar the data presented in Figure 1D). In each case, the best fit Gaussian and exponential curves are show in solid and dashed lines, respectively. The data from Raghuraman et al. (A), Yabuki et al. (B) and Lengronne et al. (C) all show approximately Gaussian distributions (R2 = 0.93, 0.87, 0.93), indicative of evenly spaced origin firing, with means of 35, 46 and 47 kb, respectively; none is close to a exponential distribution (R2 = 0.31, 0.15, 0.17).