Results notes

Contents

[edit] 1 Alignment summary data

  • Ordered has much much more data than disordered; this could affect results. Ordered matrices are going to be more reliable and representative of the overall group. Fluctuations in families should bother them less.
  • What would the causes of there being less disordered data? The original number of families is roughly the same.
    • Ordered has longer sequences on average
    • BLOSUM62 used for the search, it is biased towards ordered proteins and thus might be more likely to give better scores between ordered proteins and not between disordered proteins

[edit] 2 Influence and scaling

  • Is that a good graph example?
  • Influences of smaller families rise because there are less substitutions over all.

[edit] 3 Amino acid compositions

  • All disorder promoting amino acids are hydrophobic--goes in results or conclusion?

[edit] 4 Log odds substitution matrices

  • Want to analyze which values are positive.
  • Want to analyze which areas are more likely in disorder to order and visa versa

[edit] 5 Jackknife analysis

  • Need more info on the t-tests?
  • 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. But probably does not affect this metric enough to make a difference in the results.
  • In 60%: 16 difference over 210 is probably not a biologically significant difference
  • Does the flipping of disordered being below ordered to disordered being above ordered in 40 indicate the structural constraints are less over time

[edit] 6 PAM250 extrapolation

  • Disordered changes more than ordered. Could be expected given the way it changes with regards to the nondiagonal sums.
  • The change from 85% to 40% is the most; sensible.
  • The change from 40% to 60% is much greater than the change from 60% to 85%. This means most of the change is taking place between the 60% and 85% level. The effects of long term evolution must be kicking in between 85 and 60 and not change too much from 60 to 40.

[edit] 7 Gaps

XHTML 1.1 CSS 2 Sec 508