structure for linux 64bit

[jos matejus]

Dear list members,

Can anyone point me in the right direction to find either source or
binary version of structure (both with and without the front end)
which will run on a 64 bit ubuntu system. I have read through some of
the previous posts, but couldnt find too much information.

I have tried to install the linux binary for structure both with and
without the front end but keep getting errors (segmentation fault) or
i am unable to run structure on the testdata1 (see below)

Cheers
Jos

%:~/structure/console$ ./structure -i testdata1


----------------------------------------------------
STRUCTURE by Pritchard, Stephens and Donnelly (2000)
and Falush, Stephens and Pritchard (2003)
Code by Pritchard, Falush and Hubisz
Version 2.3.2.1 (Oct 2009)
----------------------------------------------------


Reading file "mainparams".
datafile is
infile
Reading file "extraparams".
Note: RANDOMIZE is set to 1. The random number generator will be
initialized using the system clock, ignoring any specified value of
SEED.
Reading file "testdata1".
readlociEOF


WARNING: Unexpected end of input file. The details of the
input file are set in mainparams. I ran out of data while reading
the data for individual 16.

----------------------------------
There were errors in the input file (listed above). According to
"mainparams" the input file should contain one row of markernames with
100 entries,
200 rows with 102 entries .

There are 400 rows of data in the input file, with an average of 8.00
entries per line. The following shows the number of entries in each
line of the input file:

# Entries: Line numbers
8: 1--400
----------------------------------

Exiting the program due to error(s) listed above.

%:~/structure/console$ uname -mrs
Linux 2.6.32-38-server x86_64

[Vikram Chhatre]

Hi Jos,

For the 64 bit version for linux/mac systems, Dr. Pritchard will have
a better answer for you. However, based on the program output you
provided, it seems that the version you installed is running properly.
The error message you are getting is due to problems with file
formatting.

Can you provide more information about your data set, e.g. number of
individuals and markers, and the parameters you are using? This will
help us troubleshoot the problem better.

Vikram

> --
> You received this message because you are subscribed to the Google Groups "structure-software" group.
> To post to this group, send email to structure...@googlegroups.com.
> To unsubscribe from this group, send email to structure-softw...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/structure-software?hl=en.
>

[Vikram Chhatre]

Jos -

About the 64 bit version, you could simply download the source code
and compile it on your 64bit machine.

On the following page, scroll down to Source Code download:
http://pritch.bsd.uchicago.edu/structure_software/release_versions/v2.3.3/html/structure.html

Hope that helps.
V

[jos matejus]

Hi Vikram,

Many thanks for your reply. I am using the sample data 'testdata1'
file that came with the installation of structure as I am just testing
the install.

%:~/structure/console$ head testdata1
1 1 0 0 1 3 8 9
1 1 0 1 -1 -1 7 -3
2 1 0 -1 2 2 6 7
2 1 0 0 5 0 9 7
3 1 0 -1 2 0 2 8
3 1 0 0 2 4 10 9
4 1 0 -1 2 -1 8 6
4 1 0 -1 -2 0 9 9
5 1 0 0 1 3 7 8
5 1 0 -1 2 1 5 7

I thought that structure is setup to automatically run this testdata1
file with the mainparam file by using
./structure

(installation instructions from the website
http://pritch.bsd.uchicago.edu/structure_software/release_versions/v2.3.3/html/install.html)

As such, I haven't altered the mainparam file. Can you let me know
what to alter if needed?

thanks again
Jos

On Mar 5, 9:09 pm, Vikram Chhatre <crypticline...@gmail.com> wrote:
> Jos -
>
> About the 64 bit version, you could simply download the source code
> and compile it on your 64bit machine.
>

> On the following page, scroll down to Source Code download:http://pritch.bsd.uchicago.edu/structure_software/release_versions/v2...

[Vikram Chhatre]

Jos -

The mainparams you're using is not set up for using testdata1 data
file. The testdata1 contains data for 200 individuals typed at 5
loci, whereas the default unedited mainrams is set up for 100
individuals typed at 100 loci. Hence the problem.

You will need to edit mainparams to reflect information in testdata1.

V

[jos matejus]

Hi Vikram, thanks for your patience. I have amended the mainparam file
to


#define INFILE testdata1 //(str) name of input data file
#define OUTFILE outfile //(str) name of output data file

Data file format

#define NUMINDS 200 //(int) number of diploid indiviuals in data
file
#define NUMLOCI 5 // (int) number of loci in data file
#define PLOIDY 2 // (int) ploidy of data
#define MISSING -9 // (int) value given to missing genotype
data
#define ONEROWPERIND 0 // (B) store data for individuals in a
single line


#define LABEL 1 // (B) Input file contains individual labels
#define POPDATA 0 // (B) Input file contains a population
identifier
#define POPFLAG 0 // (B) Input file contains a flag which says
whether to use popinfo when
USEPOPINFO==1
#define LOCDATA 0 // (B) Input file contains a location
identifier

#define PHENOTYPE 0 // (B) Input file contains phenotype
information
#define EXTRACOLS 0 // (int) Number of additional columns of data
before the genotype data start.

#define MARKERNAMES 1 // (B) data file contains row of marker
names
#define RECESSIVEALLELES 0 // (B) data file contains dominant markers
(eg AFLPs)
// and a row to indicate which alleles are
recessive
#define MAPDISTANCES 0 // (B) data file contains row of map
distances
// between loci


Advanced data file options

#define PHASED 0 // (B) Data are in correct phase (relevant
for linkage model only)
#define PHASEINFO 0 // (B) the data for each individual
contains a line
indicating phase (linkage model)

but i still get an error

----------------------------------------------------
STRUCTURE by Pritchard, Stephens and Donnelly (2000)
and Falush, Stephens and Pritchard (2003)
Code by Pritchard, Falush and Hubisz
Version 2.3.2.1 (Oct 2009)
----------------------------------------------------


Reading file "mainparams".
datafile is

testdata1

Reading file "extraparams".
Note: RANDOMIZE is set to 1. The random number generator will be
initialized using the system clock, ignoring any specified value of
SEED.
Reading file "testdata1".

WARNING: There may be more data in the input file
than indicated by the program constants. Check the values
entered for NUMLOCI and NUMINDS, etc, in the program constants.

----------------------------------
There were errors in the input file (listed above). According to
"mainparams" the input file should contain one row of markernames with

5 entries,
400 rows with 6 entries .

There are 400 rows of data in the input file, with an average of 8.00
entries per line. The following shows the number of entries in each
line of the input file:

# Entries: Line numbers
8: 1--400
----------------------------------

Exiting the program due to error(s) listed above.

Sorry for the confusion before but I interpreted the install
instructions to mean that structure was already setup to run the
testdata1 file

Thanks (yet again)
Jos

On Mar 5, 9:30 pm, Vikram Chhatre <crypticline...@gmail.com> wrote:
> Jos -
>

> The mainparams you're using is not set up for usingtestdata1data

> file.  Thetestdata1contains data for 200 individuals typed at 5

> loci, whereas the default unedited mainrams is set up for 100
> individuals typed at 100 loci.  Hence the problem.
>
> You will need to edit mainparams to reflect information intestdata1.
>
> V
>
>
>
>
>
>
>
> On Mon, Mar 5, 2012 at 3:22 PM, jos matejus <matejus...@googlemail.com> wrote:
> > Hi Vikram,
>
> > Many thanks for your reply. I am using the sample data  'testdata1'
> > file that came with the installation of structure as I am just testing
> > the install.
>
> > %:~/structure/console$ headtestdata1
> > 1 1 0 0 1 3 8 9
> > 1 1 0 1 -1 -1 7 -3
> > 2 1 0 -1 2 2 6 7
> > 2 1 0 0 5 0 9 7
> > 3 1 0 -1 2 0 2 8
> > 3 1 0 0 2 4 10 9
> > 4 1 0 -1 2 -1 8 6
> > 4 1 0 -1 -2 0 9 9
> > 5 1 0 0 1 3 7 8
> > 5 1 0 -1 2 1 5 7
>
> > I thought that structure is setup to automatically run thistestdata1
> > file with the mainparam file by using
> > ./structure
>
> > (installation instructions from the website

> >http://pritch.bsd.uchicago.edu/structure_software/release_versions/v2...)

[Vikram Chhatre]

Jos -

If you look at how the data is arranged in testdata1, you will notice
that before the genotype begins and after the individual label, there
are two extra columns. I believe one of them is POPDATA and another
is LOCDATA. So you will need to TURN ON both of those parameters in
the mainparams/extraparams.

After that edit, the analysis should work properly. If not, let us know.

V

[jos matejus]

Hi again

So I have set the mainparams file to

#define NUMINDS 200 //(int) number of diploid indiviuals in data
file
#define NUMLOCI 5 // (int) number of loci in data file
#define PLOIDY 2 // (int) ploidy of data
#define MISSING -9 // (int) value given to missing genotype
data
#define ONEROWPERIND 0 // (B) store data for individuals in a
single line


#define LABEL 1 // (B) Input file contains individual labels

#define POPDATA 1 // (B) Input file contains a population

identifier
#define POPFLAG 0 // (B) Input file contains a flag which says
whether to use popinfo when
USEPOPINFO==1

#define LOCDATA 1 // (B) Input file contains a location

identifier

#define PHENOTYPE 0 // (B) Input file contains phenotype
information
#define EXTRACOLS 0 // (int) Number of additional columns of data
before the genotype data start.

#define MARKERNAMES 1 // (B) data file contains row of marker
names
#define RECESSIVEALLELES 0 // (B) data file contains dominant markers
(eg AFLPs)
// and a row to indicate which alleles are
recessive
#define MAPDISTANCES 0 // (B) data file contains row of map
distances
// between loci

but am still receiving an error message

WARNING: Unexpected end of input file. The details of the
input file are set in mainparams. I ran out of data while reading

the data for individual 200.

----------------------------------
There were errors in the input file (listed above). According to
"mainparams" the input file should contain one row of markernames with
5 entries,

400 rows with 7 entries .

There are 400 rows of data in the input file, with an average of 8.00
entries per line. The following shows the number of entries in each
line of the input file:

# Entries: Line numbers
8: 1--400
----------------------------------

Thanks for your time
Jos

On Mar 5, 10:25 pm, Vikram Chhatre <crypticline...@gmail.com> wrote:
> Jos -
>

> If you look at how the data is arranged intestdata1, you will notice

> that before the genotype begins and after the individual label, there
> are two extra columns.  I believe one of them is POPDATA and another
> is LOCDATA.  So you will need to TURN ON both of those parameters in
> the mainparams/extraparams.
>
> After that edit, the analysis should work properly.  If not, let us know.
>
> V
>
>
>
>
>
>
>
> On Mon, Mar 5, 2012 at 4:20 PM, jos matejus <matejus...@googlemail.com> wrote:
> > Hi Vikram, thanks for your patience. I have amended the mainparam file
> > to
>

> > #define INFILEtestdata1 //(str) name of input data file

[Vikram Chhatre]

Try to set

EXTRACOLS 1

V

[jos matejus]

Sorry again:

WARNING: Unexpected end of input file. The details of the
input file are set in mainparams. I ran out of data while reading

the data for individual 178.

----------------------------------
There were errors in the input file (listed above). According to
"mainparams" the input file should contain one row of markernames with
5 entries,

400 rows with 8 entries .

There are 400 rows of data in the input file, with an average of 8.00
entries per line. The following shows the number of entries in each
line of the input file:

# Entries: Line numbers
8: 1--400
----------------------------------

I'm not really sure what the problem is to be honest. Just to
reiterate, the file testdata1 is the data file which comes with
structure.

Best
Jos

> ...
>
> read more »

[Vikram Chhatre]

Ha, it was much simpler than that. Turn OFF the markernames
parameter. We don't have that info in testdata1.

V

[jos matejus]

Champion Vikram.

Just in case anyone else might find this useful, here is the mainparam
file

Basic Program Parameters

#define MAXPOPS 2 // (int) number of populations assumed
#define BURNIN 10000 // (int) length of burnin period
#define NUMREPS 20000 // (int) number of MCMC reps after burnin

Input/Output files

#define INFILE testdata1 //(str) name of input data file

#define OUTFILE outfile //(str) name of output data file

Data file format

#define NUMINDS 200 //(int) number of diploid indiviuals in data
file
#define NUMLOCI 5 // (int) number of loci in data file
#define PLOIDY 2 // (int) ploidy of data
#define MISSING -9 // (int) value given to missing genotype
data
#define ONEROWPERIND 0 // (B) store data for individuals in a
single line


#define LABEL 1 // (B) Input file contains individual labels

#define POPDATA 1 // (B) Input file contains a population
identifier
#define POPFLAG 0 // (B) Input file contains a flag which says
whether to use popinfo when
USEPOPINFO==1
#define LOCDATA 1 // (B) Input file contains a location
identifier

#define PHENOTYPE 0 // (B) Input file contains phenotype
information
#define EXTRACOLS 0 // (int) Number of additional columns of data
before the genotype data start.

#define MARKERNAMES 0 // (B) data file contains row of marker

names
#define RECESSIVEALLELES 0 // (B) data file contains dominant markers
(eg AFLPs)
// and a row to indicate which alleles are
recessive
#define MAPDISTANCES 0 // (B) data file contains row of map
distances
// between loci


Advanced data file options

#define PHASED 0 // (B) Data are in correct phase (relevant
for linkage model only)
#define PHASEINFO 0 // (B) the data for each individual
contains a line
indicating phase (linkage model)

#define MARKOVPHASE 0 // (B) the phase info follows a Markov
model.
#define NOTAMBIGUOUS -999 // (int) for use in some analyses of
polyploid data

Thanks very much for your time and help Vikram. Much appreciated
Jos