Estimated θ per kb for O. rufipogon, indica, and japonica under different schemes of site selection.

Only sites whose coverage in GA and SOLiD platform are both 6X or more are used. S is the number of segregating sites in a given region and S>1 counts the same sites but excludes singletons. S>2 excludes doubletons in addition. For the “combined” (GA plus SOLiD) data, S>1 represents keeping sites whose variant appear more than once in both GA and SOLiD data. For a comparison, estimates based on all polymorphic sites are also given (“All sites”). These estimates are greatly inflated due to the excesses in singletons and doubletons, many of which are sequencing errors (See Table S3 for more information).