fastSTRUCTURE popfile?

1,450 views
Skip to first unread message

Adam Cardilini

unread,
Jul 6, 2014, 8:49:23 AM7/6/14
to structure...@googlegroups.com
G'day All,

I am trying to include a --popfile= when running distruct.py in fastSTRUCTURE. I haven't been able to find a description of what the file should look like. If someone knows what this file should like like could you please let me know.

I thought it might just be a list of the populations in order of the samples, but this didn't seem to work. Do I need to include the sample names as well?

Sorry if this is really simple and I have overlooked something in the README file, thanks for your help.

Cheers,
Adam

Vikram Chhatre

unread,
Jul 7, 2014, 8:43:20 PM7/7/14
to structure-software
Adam -

I haven't yet used the distruct script that comes with fastSTRUCTRUE, but I suppose the format is similar to that needed by DISTRUCT.  You can check distruct manual for the file format.  

Basically, popfile (called .popq by Distruct) format is as follows.  The example contains membership coefficients of 8 populations in 2 clusters (K=2).  The last column is number of individuals in the given population.

  1:     0.926  0.074              28
  2:     0.907  0.093              27
  3:     0.914  0.086               4
  4:     0.922  0.078              29
  5:     0.634  0.366               4
  6:     0.077  0.923              10
  7:     0.059  0.941              18
  8:     0.159  0.841              16

HTH
V



--
You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.
To post to this group, send email to structure...@googlegroups.com.
Visit this group at http://groups.google.com/group/structure-software.
For more options, visit https://groups.google.com/d/optout.

Adam Cardilini

unread,
Jul 28, 2014, 9:48:12 PM7/28/14
to structure...@googlegroups.com
Thanks Vikram,

I will check out the distruct manual and see how I go.

Cheers,
Adam

Tommy Carstensen

unread,
Sep 14, 2015, 4:04:13 PM9/14/15
to structure-software
Did you solve this Adam? What did your file look like in the end? Thanks.

Vikram Chhatre

unread,
Sep 14, 2015, 4:04:56 PM9/14/15
to structure-software
Check the distruct2 project on github for an example of popfile.


V

Tommy Carstensen

unread,
Sep 14, 2015, 4:09:02 PM9/14/15
to structure-software
Thanks much. I also found the answer by reading the function get_admixture_proportions in the distruct.py code:
https://github.com/rajanil/fastStructure/blob/master/distruct.py

To anyone reading this. It's just a single column file with population IDs in the same order as the samples in the input file.

On Monday, September 14, 2015 at 9:04:56 PM UTC+1, Vikram Chhatre wrote:
Check the distruct2 project on github for an example of popfile.


V
On Mon, Sep 14, 2015 at 4:03 PM, Tommy Carstensen <tommy.ca...@gmail.com> wrote:
Did you solve this Adam? What did your file look like in the end? Thanks.

On Tuesday, July 29, 2014 at 2:48:12 AM UTC+1, Adam Cardilini wrote:
> Thanks Vikram,
>
>
> I will check out the distruct manual and see how I go.
>
>
> Cheers,
> Adam
>
> On Sunday, July 6, 2014 10:49:23 PM UTC+10, Adam Cardilini wrote:G'day All,
> I am trying to include a --popfile= when running distruct.py in fastSTRUCTURE. I haven't been able to find a description of what the file should look like. If someone knows what this file should like like could you please let me know.
> I thought it might just be a list of the populations in order of the samples, but this didn't seem to work. Do I need to include the sample names as well?
> Sorry if this is really simple and I have overlooked something in the README file, thanks for your help.
> Cheers,
> Adam

--
You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-software+unsub...@googlegroups.com.

Ingrid Jordon-Thaden

unread,
Jul 18, 2016, 10:43:29 PM7/18/16
to structure-software
Hi All,
Does this mean that you need an entry for every sample/accession? For example, I have 656 samples, but 33 populations. Do I need 656 lines? or 33? 
I can't seem to figure out how to get it to read my pop file correctly.
Best,
Ingrid

Ingrid Jordon-Thaden

unread,
Jul 18, 2016, 10:43:29 PM7/18/16
to structure-software
Where do you find the "membership coefficients" from the faststructure out put? 
Thanks!
Ingrid


On Monday, July 7, 2014 at 5:43:20 PM UTC-7, Vikram Chhatre wrote:
Adam -

I haven't yet used the distruct script that comes with fastSTRUCTRUE, but I suppose the format is similar to that needed by DISTRUCT.  You can check distruct manual for the file format.  

Basically, popfile (called .popq by Distruct) format is as follows.  The example contains membership coefficients of 8 populations in 2 clusters (K=2).  The last column is number of individuals in the given population.

  1:     0.926  0.074              28
  2:     0.907  0.093              27
  3:     0.914  0.086               4
  4:     0.922  0.078              29
  5:     0.634  0.366               4
  6:     0.077  0.923              10
  7:     0.059  0.941              18
  8:     0.159  0.841              16

HTH
V
On Sun, Jul 6, 2014 at 8:49 AM, Adam Cardilini <a.car...@gmail.com> wrote:
G'day All,

I am trying to include a --popfile= when running distruct.py in fastSTRUCTURE. I haven't been able to find a description of what the file should look like. If someone knows what this file should like like could you please let me know.

I thought it might just be a list of the populations in order of the samples, but this didn't seem to work. Do I need to include the sample names as well?

Sorry if this is really simple and I have overlooked something in the README file, thanks for your help.

Cheers,
Adam

--
You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-software+unsub...@googlegroups.com.

Vikram Chhatre

unread,
Jul 18, 2016, 10:43:46 PM7/18/16
to structure-software
Check the meanQ file.

To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.
To post to this group, send email to structure...@googlegroups.com.
Visit this group at http://groups.google.com/group/structure-software.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.

To post to this group, send email to structure...@googlegroups.com.
Visit this group at https://groups.google.com/group/structure-software.

Vikram Chhatre

unread,
Jul 18, 2016, 10:44:48 PM7/18/16
to structure-software
The popfile has as many lines as there are samples in your data. 

pop1
pop1
pop1
pop2
pop2
pop2
pop2
etc...

Ingrid
To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.

To post to this group, send email to structure...@googlegroups.com.
Visit this group at http://groups.google.com/group/structure-software.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "structure-software" group.
To unsubscribe from this group and stop receiving emails from it, send an email to structure-softw...@googlegroups.com.

To post to this group, send email to structure...@googlegroups.com.
Visit this group at https://groups.google.com/group/structure-software.
Reply all
Reply to author
Forward
0 new messages