Hello,
I am observing mismatch between the Shannon Index calculated manually and Shannon Index provided by mothur as part of the command: summary.single. Command Synatx used is.
summary.single(sabund=sabundfile, calc=shannon)
To provide information, the Sabund file that I used has following Data:
unique 2 17 2
0.01 2 9 6
0.02 5 2 1 4 0 1
0.05 5 1 1 3 1 1
0.06 5 1 0 2 1 2
0.08 8 1 0 1 1 1 0 0 1
0.09 11 1 0 0 1 1 0 0 0 0 0 1
0.10 16 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1
0.11 21 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
For the Sabund file, when I performed the Shannon index diversity calculation(Performed using summary.single command) received the following
output.
abel shannon shannon_lci shannon_hci
unique 2.912494 2.609258 3.215731
0.01 2.648438 2.361207 2.935670
0.02 1.967530 1.711287 2.223774
0.05 1.860419 1.631457 2.089381
0.06 1.700178 1.483706 1.916650
0.08 1.448154 1.195361 1.700947
0.09 1.141226 0.856378 1.426075
0.10 0.548874 0.326945 0.770803
0.11 -0.000000 -0.000000 0.000000
When I performed the Shannon index calculations manually, the index values are correct from 0.08 cut off till 0.11 cut off. But the values from 0.01 till 0.006 were not correct.
For example in the case of 0.02 CUTOFF the value from manual calculation was: 1.109 but from mothur was : 1.967530
The calculation formula I have done is as follows
N = 21 - Total Number of Individuals
OTUUnit ni ni/N log(ni/N) (ni/N * log(ni/N))
1 2 0.095 -2.351 -0.224
2 2 0.095 -2.351 -0.224
3 12 0.571 -0.560 -0.320
4 5 0.238 -1.435 -0.342
Shannon Index = -1 * (ni/N * log(ni/N)), which is -1 * -1.109 = 1.109
Can you please confirm whether the mothur output is correct? If so, what is the deviation from the Shannon Index formula that has been used. When I
compared the data with a diversity excel, calculated values are same as the value which I have determined. This would essentially mean the data from
Sabund file that I have used has some issue.