Statistical analysis

[NGS_user]

Hi folks,
I am a bit confused as to the possible statistical tools I could use in my analyses. Basically I have a number of observed genes for each gene family and a list of the number of known genes per family. the numbers of the observed are much much lower than the expected as I used universal primers and have smaller subsets of these gene families. If I do a chi square test it tells me I have significant differences. My distribution visually looks the same as the expected distribution so the of genes are similar relative to the other families, is there some statistical analysis where I can compare numbers of observed and expected with respect to other observed and expected numbers?

[colindaven]

Sounds like the Chi square test you've already done would be helpful. Check out the means, medians and distribution.

Also check the Chi square value. You might want to control for multiple testing with Bonferroni correction as you may indeed only have a tiny variation in your data which is "significant" because of many tests.