Re: Cluster analysis for beginners

On Mar 29, 4:38 pm, David Winsemius <doe_s...@xxxxxxxxxxx> wrote:
Sidney <milan_y...@xxxxxx> wrote innews:24466740.1175159875339.JavaMail.jakarta@xxxxxxxxxxxxxxxxxxxxxx:

Dear all,
I ran into a problem that I can't solve with my basic statistics high
school knowledge - apologies if this is too trivial to some of you.

Assume you have 5000 proteins that are ordered by their molecular
weight from 1000 Daltons to 100000 Daltons (the numbers don't matter).
If you now find that a certain motif (e.g. a specific phosphorylation
motif) which is only found within a certain molecular weight range,
e.g. only between 77000-81000 Daltons, how do you determine if this
'clustering' is significant? At this point I have no idea what to do
and where to start at. Your input is very much appreciated. Thanks a
lot in advance. Sidney

I did see illywacker's reply, but I disagree. I thought your scientific
question was reasonably clear for one thing. If your null hypothesis is
that there is no association between MW and presence of the motif, you
could start by arranging the proteins in deciles of weight and testing for
uniformity of motif-ication in a multinomial model with 9 degrees of
freedom. Unless most of your proteins are in that specified range above,
with 5000 data points it seems reasonably clear that you will get a
'significant' result using that approach. I would argue that such a test
does not represent one with "strong assumptions".

Check it out, and the Prartt and Savage references therein.

illywhacker;

.

Relevant Pages

  • Re: Cluster analysis for beginners
    ... weight from 1000 Daltons to 100000 Daltons. ... If you now find that a certain motif (e.g. a specific phosphorylation ... "Compared to a uniform distribution of proteins" or "Compared to a unimodal distribution of proteins with a peak of XXX Daltons..." ...
    (sci.stat.math)
  • Re: Cluster analysis for beginners
    ... weight from 1000 Daltons to 100000 Daltons. ... If you now find that a certain motif (e.g. a specific phosphorylation ... Classical hypothesis testing is fatally flawed. ...
    (sci.stat.math)
  • Re: Cluster analysis for beginners
    ... weight from 1000 Daltons to 100000 Daltons (the numbers don't ... If you now find that a certain motif (e.g. a specific ... molecular weight range, e.g. only between 77000-81000 Daltons, how ... Unless most of your proteins are in ...
    (sci.stat.math)
  • Re: Cluster analysis for beginners
    ... weight ranges, and second, the sizes of the clusters also vary a lot ... proteins to the mass spectrum of motif-bearers. ... difference in the median molecular weight of motif bearing proteins and ...
    (sci.stat.math)
  • Re: Cluster analysis for beginners
    ... weight from 1000 Daltons to 100000 Daltons. ... If you now find that a certain motif (e.g. a specific phosphorylation ... Unless most of your proteins are in that specified range above, ...
    (sci.stat.math)