Jackknifes
Steady trend in increasing nondiagonal sums for all classes; makes sense because an increase in PI means an increase in mutations off of the diagonal.
Steady trend in disorder increasing compared to order. Disordered starts out lower and ends up higher.
PROBLEM W/ 85%: lots of low counted substitutions. Could affect this metric strongly, as substitutions with 0 substitutions end up having very very low values.
Contents |
[edit] 1 Welch Two Sample t-tests
[edit] 1.1 85% T-test
- t = -198.4535, df = 327.386, p-value < 2.2e-16
- alternative hypothesis: true difference in means is not equal to 0
- 95 percent confidence interval: -112.3374 -110.1321
- sample estimates:
- x: -1908.610
- y: -1797.376
[edit] 1.2 60% T-test
- t = -75.6752, df = 378.127, p-value < 2.2e-16
- alternative hypothesis: true difference in means is not equal to 0
- 95 percent confidence interval: -16.33282 -15.50556
- sample estimates:
- x: -623.4058
- y: -607.4866
DISCUSSION: THESE SPREAD OVER 210 DIFFERENCES IS NOT BIOLOGICALLY SIGNIFICANT
[edit] 1.3 40% T-test
- t = 541.8361, df = 227.407, p-value < 2.2e-16
- alternative hypothesis: true difference in means is not equal to 0
- 95 percent confidence interval: 82.13297 82.73253
- sample estimates:
- x: -96.58791
- y: -179.02066