Question about Ace estimator

Hi All,

I am analyzing some data from 454, and I think that is something wrong with Ace richness estimator.
My data are some Milk samples, with low diversity. I have 3 samples with around 2 thousand sequences each one.
The sample A has a higher diversity than the samples B and C, and this is demonstrated by the number of OTUs, Chao, Jackkinife and Shannon and also by the phylogenetic assignment. However, for Ace richness, the pattern is completely different, and the sample B has the richness higher than sample A (see below).

label OTUs chao jackknife shannon ace
A 134 242 264 2,25 245
B 23 99 117 0,17 560
C 13 35 36 0,11 68

Can someone explain me why is it happening? What is Ace taking in account to be so different than all other estimators? Thank you very much. Caio Rachid

Hi there,

Have you taken the data in your rabund file for these samples and calculated the ACE parameter by hand? This hasn’t changed any since way, way back when so I’m pretty confident the the calculator works. I’d encourage you to follow the tutorial on the ACE page with your own data and see if you can get the numbers by hand.


Hi Pat,

I used the summary.single comand to calculate Chao from a .list file.

The values that I showed here in this post was obtained when I did the alignment, dist matrix calculation and clustering of each treatment indivídually.

But, when I repeated the analisys doing all theses steps with a single file contained the sequencences of all three treatments, and puted the summary.single from the .shared file, the Ace estimator was not different than the pattern of the others estimators.

Can you tell me why it happens?

Thanks a lot
Caio Rachid