RepeatSeq for accurate genotyping of microsatellite repeats

[oiiio]

Published in NAR today, RepeatSeq is a tool for genotyping tandem repeats from sequencing data. RepeatSeq is stable, very easy to install and use, with very accurate results compared to existing tools and even in Sanger validation of our genotype calls.

Open Article from NAR:

http://nar.oxfordjournals.org/content/early/2012/10/22/nar.gks981.abstract

GitHub Download:

https://github.com/adaptivegenome/repeatseq

I am very happy to answer any questions in this thread or by private message/email ([email protected]). Any feedback is widely welcome, and is a critical part to RepeatSeq's development.

[deyler]

I am able to install and run RepeatSeq. However - how does one generate a regions file for input? It does not appear to be a simple BED file of repeat regions.

[deyler]

I received a reply from Gareth Highnam via another channel and will repost here.

Quote:

chr:start-stop<tab>2.3_3_100_0_14_0_57_42_0_0.99_GCC

2nd column string doesnt affect compute so use that

End quote

[basalganglia]

Hello,

I have done all instructions in Repatseq website. But I can not run repatseq when I write repeatseq command?

Also how can I prepare region file? Is it fasta file including region? There is an example above but I dont understand whats the 2.3_3_100_0_14_0_57_42_0_0.99_GCC ??

Thanks

BG

[deyler]

more info needed

The regions file is the reformated output from Gary Benson's Tandem Repeat Finder. You can find it here:

http://tandem.bu.edu/trf/trf.html

You want to run trf using the -ngs and -h options and then parse the output. It gives you a output formatted like this:

@chr1
2 20 some other stuff
100 120 some other stuff
@chr2
56 69 some other stuff
etc.

RepeatSeq only uses the chromosome, start, and end positions. So you need to reformat the file to look like this:

[chrom]:[start]-[end]<tab>some_other_stuff

example:
chr1:2-20 some_other_stuff

That should be easy, if it's not then PM me.

I have no idea why you can't run repeatseq. If you post your command and the error it generates when you try and run it, perhaps I'll be able to help you.

[basalganglia]

Thank you for your help

I have download repeatseq, bamtools and fastahack from website with zip file.

I have built repeatseq but when I write;

[basalganglia@abc2 repeatseq-master]$ repeatseq
-bash: repeatseq: command not found

is written.

I think I have problem with LD_LIB-PATH. I have used this command ;

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/home/basalganglia/repeatseq-master/bamtools

Is it right or not ?

Thanks again

[GenoMax]

You probably need to add execute permissions to the executable. You can do that by

$ chmod u+x repeatseq

Then you can run the application by

$ ./repeatseq

or

$ /path_to/repeatseq

[deyler]

echo @GenoMax

@GenoMax is right. repeatseq needs to be 1) executable and 2) on your $PATH when you try to run it.

chmod u+x

makes a file executable.

If you are in the same directory as repeatseq, you should then be able to just type repeatseq or ./repeatseq. If you're not, then you need to provide the /full/path/to/repeatseq when you run it.

$PATH is the list of directories where your computer looks for programs to run. You can see what's currently in it by typing

echo $PATH

To make your life easier you might want to make a directory where all your executables live. Then add that to your $PATH for all login shells - I think in ~/.bash_profile or some such place. Once you do that you don't need to remember/type full path names for your executables anymore.

[basalganglia]

Thank you for your help, but

I have got same problem again ;

$ chmod u+x repeatseq
$ ./repeatseq
./repeatseq: error while loading shared libraries: libbamtools.so.2.3.0: cannot open shared object file: No such file or directory

[deyler]

Can you post the contents of the repeatseq makefile? The start of mine looks like this:

# $* is prefix shared by target and dependent; $@ is name of target file
CFLAGS = -c -O3 -Ibamtools/src
OBJS= repeatseq.o structures.o CLParse.o
NAME= repeatseq

$(NAME): $(OBJS)
g++ -o $@ $(OBJS) fastahack/Fasta.cpp fastahack/split.cpp -lpthread -lbamtools -Lbamtools/lib

It looks like you are setting LD_LIBRARY_PATH the right way from above, but I can tell you mine's not set at all and repeatseq still runs.

Also, what happens when you try

which bamtools

?

[basalganglia]

Dear deyler,

Makefile including


# $* is prefix shared by target and dependent; $@ is name of target file
CFLAGS = -c -O3 -Ibamtools/src
OBJS= repeatseq.o structures.o CLParse.o
NAME= repeatseq

$(NAME): $(OBJS)
g++ -o $@ $(OBJS) fastahack/Fasta.cpp fastahack/split.cpp -lpthread -lbamtools -Lbamtools/lib

# Suffix rules: tell how to take file with first suffix and make it into
# file with second suffix

.cpp.o:
g++ $(CFLAGS) $*.cpp
clean:
rm *.o



When I wrote which bamtools;

repeatseq-master]$ which bamtools
/usr/bin/which: no bamtools in (/usr/lib64/qt-3.3/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/sbin:/opt/jdk1.7.0_71/bin:/home/basalganglia/bin)

Many thanks for your help

[deyler]

OK. This result means that your system cannot find your installation of bamtools. Let's fix that first, and hopefully the other problem goes away too.

The thing in which there is no bamtools is your PATH. I can tell you that, by looking at your PATH - it's the thing in parentheses. Any program that you want to be able to run just by typing its name needs to be located in a directory listed in the PATH. So you need to add the directory where bamtools lives to your PATH variable.

There are several ways of doing this.
1) the stupid way

$> PATH="/full/path_to/bamtools:$PATH"
$> export PATH

If you do it the stupid way, you'll have to do it again every time you start a new login shell.

2) the smarter way
Go to your home directory. Usually cd ~ will do this.
Then open .bash_rc or .bash_profile. Use whatever text editor you like. Note that neither file may exist before you do this as they are user-specific configuration files.
Now, put the two lines from #1 into the configuration file. Save it and go back to the terminal. Then run this command:
. ~/.bash_profile (or .bash_rc, whichever one you used)
or just log out and log back in again.

If you did it right, you should be able to type "echo $PATH" and see your PATH that includes the directory where bamtools lives. You should also be able to type "which bamtools" and have it spit out the full path to bamtools.

Note that you can do this for repeatseq too, so that you can run it from any directory within your system. All of my programs live in a deyler/bin directory, which is in my PATH - so from wherever I am, I can say "myprogram3" and it runs.

I'm going to be offline for the next 12-14 hours but I'll check this thread when I get back on.

[basalganglia]

I have same trouble again

repeatseq-master]$ chmod u+x repeatseq
repeatseq-master]$ ./repeatseq
./repeatseq: error while loading shared libraries: libbamtools.so.2.3.0: cannot open shared object file: No such file or directory
repeatseq-master]$ PATH="/home/bg/repeatseq-master/bamtools:$PATH"
repeatseq-master]$ export PATH
repeatseq-master]$ which bamtools
/usr/bin/which: no bamtools in (/home/bg/repeatseq-master/bamtools:/home/bg/repeatseq-master/bamtools:/home/bg/repeatseq-master/bamtools:/usr/lib64/qt-3.3/bin:/usr/local/bin:/bin:/usr/bin:/usr/local/sbin:/usr/sbin:/sbin:/opt/jdk1.7.0_71/bin:/home/bg/bin)

[GenoMax]

Do you see anything when you try following command:

$ ls -lh /home/bg/repeatseq-master/b*

It sounds like you could use some help with basic unix. It may be useful to spend a couple of hours going through this part: http://korflab.ucdavis.edu/Unix_and_...ent.html#part1

Did you compile repeatseq program yourself or download a pre-compiled binary?

[basalganglia]

/home/bg/repeatseq-master/bamtools:
total 36K
drwxrwsr-x 2 bg bg 4.0K May 28 00:08 bin
drwxrwsr-x 4 bg bg 4.0K May 28 00:06 build
-rw-rw-r-- 1 bg bg 1.8K Jul 30 2013 CMakeLists.txt
drwxrwsr-x 2 bg bg 4.0K Jul 30 2013 docs
drwxrwsr-x 4 bg bg 4.0K May 28 00:06 include
drwxrwsr-x 2 bg bg 4.0K May 28 00:07 lib
-rw-rw-r-- 1 bg bg 1.2K Jul 30 2013 LICENSE
-rw-rw-r-- 1 bg bg 2.0K Jul 30 2013 README
drwxrwsr-x 7 bg bg 4.0K Jul 30 2013 src

I have download from website then transfer to server with Filezilla