Reader Comments

Post a new comment on this article

p values for spores

Posted by fstahl on 08 Dec 2008 at 18:52 GMT

However, in the spore analysis, only the LEU2-LYS2 interval (p = 0.014) was not significantly different from WT
http://plosgenetics.org/article/info:doi/10.1371/journal.pgen.1000188#article1.body1.sec2.sec9.p1

In Tables 1 and 2, certain calculations are based on “spores”, instead of on tetrads. These spore populations result from pooling the spores from all tetrads dissected. The error bars and claims of significance in these columns are suspect. The statistic used by the authors presumes that each spore is an independent sample, whereas the spores (parental or recombinant) originate in pairs. Thus, the calculated error limits in the tables are artificially small, which probably led to the authors’ statement, “In the analysis of complete tetrads, the URA3-LEU2 and ADE2-HIS3 intervals were significantly different from WT … but the LEU2-LYS2 … and the LYS2-ADE2 … were not … However, in the spore analysis, only the LEU2-LYS2 interval … was not significantly different from WT … Similarly, in isogenic SK1 strains, CO frequencies were increased in csm4Δ mutants at four out of eight analyzed intervals in complete tetrads and at six out of eight intervals in the spore analysis ….”