dendrogram

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

dendrogram

Zhesi He-2
Dear list,

Sorry if my question is too easy.
I now have a class list like

1.01.01.01
1.01.01.02
1.02.01.01
1.03.01.01
...
9.09.09

I have no problem transferring it to a matrix without those zeros. But
I really want to have a dengrogram class object so that I can have a
hierarchical plot.

Also... is there any package that contains interactive dendrogram? for
example, i can select the levels of cutting the  tree on the plot?
something like that would be useful.

Thanks a lot.

___________________________________________________

Zhesi He
Computational Biology Laboratory, University of York
York YO10 5YW, U.K.
Phone:  +44-(0)1904-328279
Email:  [hidden email]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: dendrogram

Sean Davis



On 1/4/06 6:03 AM, "Zhesi He" <[hidden email]> wrote:

> Dear list,
>
> Sorry if my question is too easy.
> I now have a class list like
>
> 1.01.01.01
> 1.01.01.02
> 1.02.01.01
> 1.03.01.01
> ...
> 9.09.09
>
> I have no problem transferring it to a matrix without those zeros. But
> I really want to have a dengrogram class object so that I can have a
> hierarchical plot.

You didn't really explain what kind of clustering you want to do.  However,
you probably want to do help.search('cluster').  That will give you many
methods for clustering.  Most of them work with either a matrix or a
distance object.  See ?dist for help with the latter.

Sean

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: dendrogram

Zhesi He-2
Thanks for reply.
The problem is I don't need to do clustering. What I meant was the
matrix was already in a hierarchical format.
each number before the dot is one layer of the tree
so 1>1.01 > 1.01.01 > 1.01.01.01

so part of lowest nodes of the tree might look like this

1.01
|
|__1.01.01
| |______1.01.01.01
| |______1.01.01.02
|
|__1.01.02
        |______1.01.02.01
        |______1.01.02.02


so how can I convert the list into such format? each one is a node of
the tree and there are 4 layers.
Also. I'm looking for any interactive dendrogram visualisation.

Thanks in advance.
Zhesi.

On 4 Jan 2006, at 11:48, Sean Davis wrote:

>
> On 1/4/06 6:03 AM, "Zhesi He" <[hidden email]> wrote:
>
>> Dear list,
>>
>> Sorry if my question is too easy.
>> I now have a class list like
>>
>> 1.01.01.01
>> 1.01.01.02
>> 1.02.01.01
>> 1.03.01.01
>> ...
>> 9.09.09
>>
>> I have no problem transferring it to a matrix without those zeros. But
>> I really want to have a dengrogram class object so that I can have a
>> hierarchical plot.
>
> You didn't really explain what kind of clustering you want to do.  
> However,
> you probably want to do help.search('cluster').  That will give you
> many
> methods for clustering.  Most of them work with either a matrix or a
> distance object.  See ?dist for help with the latter.
>
> Sean
>
>

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: dendrogram

Sean Davis
In reply to this post by Sean Davis



On 1/4/06 6:48 AM, "Sean Davis" <[hidden email]> wrote:

>
>
>
> On 1/4/06 6:03 AM, "Zhesi He" <[hidden email]> wrote:
>
>> Dear list,
>>
>> Sorry if my question is too easy.
>> I now have a class list like
>>
>> 1.01.01.01
>> 1.01.01.02
>> 1.02.01.01
>> 1.03.01.01
>> ...
>> 9.09.09
>>
>> I have no problem transferring it to a matrix without those zeros. But
>> I really want to have a dengrogram class object so that I can have a
>> hierarchical plot.
>
> You didn't really explain what kind of clustering you want to do.  However,
> you probably want to do help.search('cluster').  That will give you many
> methods for clustering.  Most of them work with either a matrix or a
> distance object.  See ?dist for help with the latter.

You could probably try creating your own dendrogram object from scratch.
See help(dendrogram) for what might be involved.  However, if you are
looking to browse your list and you have a webserver available, you might
consider using DHTML to create a nice collapsible list like those shown
here:

http://www.oreillynet.com/pub/a/javascript/2002/02/22/hierarchical_menus.htm
l

Doing a google search for hierarchical list javascript might get you what
you need.

Sean

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html