Problem in summary.single file

Commands in mothur
1

Hello,

I am getting this result after doing mothur > summary.single(shared=stability.an.shared, calc=nseqs-coverage-sobs-invsimpson, subsample=14272) [14271 is the no. of sequence of the smallest samlple]


label group method nseqs coverage sobs invsimpson invsimpson_lci invsimpson_hci 0.03 Sample1 ave 14742 0.99221 224.895 8.860295 8.640853 9.091174 0.03 Sample12 ave 14742 0.99630 141.139 9.128217 8.969688 9.292452 0.03 Sample13 ave 14742 0.99626 148.776 9.978423 9.793687 10.170263 0.03 Sample18 ave 14742 0.99378 205.971 14.276668 13.974664 14.592015 0.03 Smple1 std 0 0.00062 9.09835 0.103303 0.10143 0.105278 0.03 Sample12 std 0 0.00046 6.498744 0.075838 0.07445 0.077307 0.03 Sample13 std 0 0.00043 6.295222 0.090122 0.088274 0.092076 0.03 Sample18 std 0 0.00056 8.082336 0.139745 0.137408 0.142195

Why am I getting 0 no. of sequences in the std method ?


looking forward.
2

Did you possibly run sub.sample before running summary.single? Do all of your samples have 14742 reads in them?

3

Hello Dr. Schloss,

Yes I did sub. sample before summary.single. 14742 is the no. of sequence of my smallest sample (sample with least no.of unique sequences).


looking forward,

Richa

4

You don’t want to run summary.single on subsampled data. The reason the std (standard deviation) is zero is because there’s no variation since all of the samples have the same number of reads.

pat

5

Hi Dr. Schloss,

Do you mean we should remove the subsample=# option from summary.single command? Or using the original (before subsampled) .shared file not the .subsample.shared file?
The sub.sample command will generate a separate .subsample.shared file and the original .shared file is not modified, am I right?

Also, how informative those “std” values are? Can I just use the “ave” values to make a table and put it in my dissertation? Thanks!

Deng Pan

6

Hello Dr. Schloss,

Thanks for your reply.

I follow Miseq SOP, where I see

“mothur > summary.single(shared=stability.an.shared, calc=nseqs-coverage-sobs-invsimpson, subsample=2441)”

What is subsample here ? Could you please explain, the correct command line that I should use at summary.single step ??

Thanks

7

Hello Dr. Schloss,

Thanks for your reply.

I follow Miseq SOP, where I see

“mothur > summary.single(shared=stability.an.shared, calc=nseqs-coverage-sobs-invsimpson, subsample=2441)”

What is subsample here ? Could you please explain, the correct command line that I should use at summary.single step ??

Thanks and looking forward

8

Earlier you said…

Yes I did sub. sample before summary.single. 14742 is the no. of sequence of my smallest sample (sample with least no.of unique sequences).

Where did you do that?

if you are following the SOP, the correct command is what you posted last:

summary.single(shared=stability.an.shared, calc=nseqs-coverage-sobs-invsimpson, subsample=2441)

9

Hello

If I understood correctly, in mothur > summary.single(shared=stability.an.shared, calc=nseqs-coverage-sobs-invsimpson, subsample=2441), 2441 is the total no. of sequences of the smallest sample in SOP.

But you wrote: " You don’t want to run summary.single on subsampled data". Why so ?

Are you suggesting that I should not run summary.single on subsmapled data, if I have to work with Standard deviation values ??

Thanks and looking forward.

10

But you wrote: " You don’t want to run summary.single on subsampled data". Why so ?

Are you suggesting that I should not run summary.single on subsmapled data, if I have to work with Standard deviation values ??

Sorry, I’m lost. you just quoted what I said and asked if that was what I said. The input to summary.single should be the file that is generated by make.shared and should not be subsampled prior to running summary.single.