Clusters
Hello ADE users,
I have a big problem with the Clusters-program-module.
I did a COA with "biodata" 29 sites (in lines) and 198 species (in col.)
with ADE (Apple G3).
Then I did
Clusters
a) compute distances (input data file: *.fcli (29 sites); type of
distance:1 Eucle.)
b) compute hierarchy (input file: *.fcli.dist; type of hierarchy: 4)
c) prepare convexhulls (input hierarchy file: *.fcli.2mha; Number of
hierarchy levels: 5)
d) inertia-analysis-partition
(input data file "biodata":29 sites (in lines) and 198 species
input partition file: biodatafclidend
select a col. Number: 5 (for 5 clusters)
type of analysis: 1)
Now I have problems with the result of d (inertia analysis-partition).
Normally I get a specification of the Variable-(= species) numbers and
the corresponding values for each cluster.
f.e.
Var. 1,2,3,4,5,6,7,8,9,10,11,............196,197,198
Clust1 0,0,-1,1,..........
Clust2 .......
What I get is
Var. ....,69, 70, 71, 7, 73,........, 144, 1, 5, 146, 147.....196, 197, 198
There is 7 instead of 72 and no corresponding value (species) for each cluster,
and number 145 is missing; here are 2 columns: 1 and 5 with only 1 value.
So I always get 199 columns for 198 variables (species) and only 197
coresponding values in each cluster (the problems are at number 72 and 145).
Whatīs wrong? Who has a hint? Who can help me ?
1000 Thanks
Oskar Deichner
attached mail follows:
CoInertia-program-module for Apple ?
Hello ADE users,
I have a big problem with the CoInertia-program-module.
a) I made with my data-set a COA (samples/species)
b) I made with my data-set a PCA (samples/abiotic factors), rows
(=samples) weighted with
*.fcpl from COA
Everything functions very well.
c) I want to make a CoInertia-Analyses:
inputfiles
a) *.cnta from the PCA
b) *.fcta from the COA
c) Output file: xyz
At this point Ialways get the advice "Error, canīt open binary file
(ouvlec)".
I updated x-times the program-module-files (COA, PCA, CoInertia) from
the ftp-server, but the result is the same.
System:
Apple (PowerMac G3 blue/white); MacOs 8.6 German; ADE-4
Concerning the data-set. The Coinertia-analysis with the same data-set
runs at a friends PC (Windows98) ADE-version.
Is there something wrong with the program-module for Apple-PowerPc ?
Who can help me ?
1000 Thanks
This archive was generated by hypermail 2b30 : Thu Jul 04 2002 - 09:39:10 MEST