fastSTRUCTURE running but how to build .str or .bed file

[sno...@googlemail.com]

Hey anyone,

finally i got Faststructure working on Debian correctly with the test files but i am ashamed because i cannot create an input file for fastSTRUCTURE. In old Structure software everything worked fine wit my .txt file. BUT how to create a STR or BED file ????
PLEASE Help or give me a hint.....

[Vikram Chhatre]

Fast structure requires two rows per individual data if using 'str' format.  It's basically a structure file

- without the first row containing locus names

- with an extra four empty columns at the beginning, so six total columns before the locus data begins if you have popdata column (5 if not).

Alternatively, if you have nextgen data in hapmap format, you might be able to convert to .bed format using PLINK.

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.

[sno...@googlemail.com]

Ok... so am i right when i say : i can use my txt file, rename it to .str and format it as shown ? :

"first row empty"
"col1 empty" "col2 empty" "col3 empty" "col4 empty" "col5 indiv.Data" "col6 popdata " Col7-296 genetic information"

I do not have nextgen data....

Am Freitag, 28. November 2014 02:48:42 UTC+1 schrieb Vikram Chhatre:

Fast structure requires two rows per individual data if using 'str' format.  It's basically a structure file

- without the first row containing locus names

- with an extra four empty columns at the beginning, so six total columns before the locus data begins if you have popdata column (5 if not).

Alternatively, if you have nextgen data in hapmap format, you might be able to convert to .bed format using PLINK.

V

On Thu, Nov 27, 2014 at 3:46 PM, <sno...@googlemail.com> wrote:

Hey anyone,

finally i got Faststructure working on Debian correctly with the test files but i am ashamed because i cannot create an input file for fastSTRUCTURE. In old Structure software everything worked fine wit my .txt file. BUT how to create a STR or BED file ????
PLEASE Help or give me a hint.....

--
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]

Yes for .str extension.

Don't use spaces for items in any of the first empty columns.  Space and tab are each used to delimit data.  Otherwise, the formatting is correct.

V

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

[sno...@googlemail.com]

Hi again,

at first thank you for your Help :)

today i tested the formattin without success.....
i opend my .txt in Excel, copyed my data in a new sheet with first row empty and first four colums empty.
Saved it as txt and rename to .str.
==> Faststructure run: IndexError: tuple index out of range 

Am Freitag, 28. November 2014 20:51:52 UTC+1 schrieb Vikram Chhatre:

Yes for .str extension.

Don't use spaces for items in any of the first empty columns.  Space and tab are each used to delimit data.  Otherwise, the formatting is correct.

V

On Fri, Nov 28, 2014 at 2:46 PM, <sno...@googlemail.com> wrote:

Ok... so am i right when i say : i can use my txt file, rename it to .str and format it as shown ? :

"first row empty"
"col1 empty" "col2 empty" "col3 empty" "col4 empty" "col5 indiv.Data" "col6 popdata " Col7-296 genetic information"

I do not have nextgen data....

Am Freitag, 28. November 2014 02:48:42 UTC+1 schrieb Vikram Chhatre:

Fast structure requires two rows per individual data if using 'str' format.  It's basically a structure file

- without the first row containing locus names

- with an extra four empty columns at the beginning, so six total columns before the locus data begins if you have popdata column (5 if not).

Alternatively, if you have nextgen data in hapmap format, you might be able to convert to .bed format using PLINK.

V

On Thu, Nov 27, 2014 at 3:46 PM, <sno...@googlemail.com> wrote:

Hey anyone,

finally i got Faststructure working on Debian correctly with the test files but i am ashamed because i cannot create an input file for fastSTRUCTURE. In old Structure software everything worked fine wit my .txt file. BUT how to create a STR or BED file ????
PLEASE Help or give me a hint.....

--
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+unsubscribe@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.

[Vikram Chhatre]

How big is your data file?  If not too big (less than a few Mb), feel free to send it to me off the list.  

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

[Ruijuan li]

Vikram,

Is it fine to have first row with individual name followed by 4 rows of all 0 and then genetic information for the rest of the file? This file runs fine using faststructure now. 

Thanks!

Ruijuan 

On Thursday, November 27, 2014 at 7:48:42 PM UTC-6, Vikram Chhatre wrote:

Fast structure requires two rows per individual data if using 'str' format.  It's basically a structure file

- without the first row containing locus names

- with an extra four empty columns at the beginning, so six total columns before the locus data begins if you have popdata column (5 if not).

Alternatively, if you have nextgen data in hapmap format, you might be able to convert to .bed format using PLINK.

V

On Thu, Nov 27, 2014 at 3:46 PM, <sno...@googlemail.com> wrote:

Hey anyone,

finally i got Faststructure working on Debian correctly with the test files but i am ashamed because i cannot create an input file for fastSTRUCTURE. In old Structure software everything worked fine wit my .txt file. BUT how to create a STR or BED file ????
PLEASE Help or give me a hint.....

--
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.

[thúy uyên hồng vũ]

Dear Vikram Chhatre,

I would like to ask whether a difference between the input in structure format and .bed format.

I have done for 2 formats and seen that

1. structure format took a longer time to run than .bed format. 

2. the marginal Likelihood in the same K of .str is different from .bed

I saw your answer from another post about "multiple K and repetition on fast structure"  (https://groups.google.com/forum/#!searchin/structure-software/fast$20structure%7Csort:date/structure-software/sT4X12K6UMo/wYv3A7gTDAAJ)

> This is unnecessary in FastStructure, which automatically performs iterations behind the scene

From your reply, I think you mean that we don't need to run many replicates as we usually do for STRUCTURE. I don't know if I am right or not. Could you explain a little bit more, please?

Thank you so much!

Best regards,

Uyen

On Friday, November 28, 2014 at 12:48:42 PM UTC+11, Vikram Chhatre wrote:

Fast structure requires two rows per individual data if using 'str' format.  It's basically a structure file

- without the first row containing locus names

- with an extra four empty columns at the beginning, so six total columns before the locus data begins if you have popdata column (5 if not).

Alternatively, if you have nextgen data in hapmap format, you might be able to convert to .bed format using PLINK.

V

On Thu, Nov 27, 2014 at 3:46 PM, <sno...@googlemail.com> wrote:

Hey anyone,

finally i got Faststructure working on Debian correctly with the test files but i am ashamed because i cannot create an input file for fastSTRUCTURE. In old Structure software everything worked fine wit my .txt file. BUT how to create a STR or BED file ????
PLEASE Help or give me a hint.....

--
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.