Protocol Online logo
Top : New Forum Archives (2009-): : Microbiology

numerical taxonomy exercise - (Dec/08/2011 )

EXERCISE: express the similiarity between the isolated bacteria X and the ones cited for reference (1, 2, 3, 4) by means of a dendrogram.

................................SPECIES
TEST............1...........2..........3..........4......X
a...................+............-......... -..........+.......+
b...................+............-..........+..........-.......+
c...................+............-..........+..........+......+
d...................-............+..........+..........-.......-
e...................-............+..........+..........-.......+

ATTEMPT: construct the first similarity matrix

...........1.......2........3.......4
1........100
2..........0.....100
3.........40.....60.....100
4.........80.....20.......20....100
5.........80.....20.......60.....60

DOUBT: do I have to choose between (1-4) and (1-5) to construct the first cluster and then go on? Or should I construct a (1-4-5) cluster instead?

-kodos-

why not "number 5" (specie X) at the top too?

kodos on Thu Dec 8 12:27:47 2011 said:


EXERCISE: express the similiarity between the isolated bacteria X and the ones cited for reference (1, 2, 3, 4) by means of a dendrogram.

................................SPECIES
TEST............1...........2..........3..........4......X
a...................+............-......... -..........+.......+
b...................+............-..........+..........-.......+
c...................+............-..........+..........+......+
d...................-............+..........+..........-.......-
e...................-............+..........+..........-.......+

ATTEMPT: construct the first similarity matrix

...........1.......2........3.......4
1........100
2..........0.....100
3.........40.....60.....100
4.........80.....20.......20....100
5.........80.....20.......60.....60

DOUBT: do I have to choose between (1-4) and (1-5) to construct the first cluster and then go on? Or should I construct a (1-4-5) cluster instead?


look at 4 and 5 ... what do you see then?

And now look at 1 and 5 and at 1 and 4...


1 and 5 ==> 80% match
1 and 4 ==> 80 % match
but 4 and 5 ==> 60% match

Would you thus put 1, 4 and 5 in the same cluster?

-pito-

You're right. Now I chose between (1-4) or (1-5) arbitrarily and then made the other matrices and the dendrogram. So you can make different dendrograms with this data... Thank you.

-kodos-

kodos on Thu Dec 8 17:40:13 2011 said:


You're right. Now I chose between (1-4) or (1-5) arbitrarily and then made the other matrices and the dendrogram. So you can make different dendrograms with this data... Thank you.


yeah, you can make "different" ones... (well actually, they are not really different.. you just "re-arrange" them.. its not different really..)

-pito-