[Daniel Fernandez]

I do that you tell me, but when I wrote ./configure the last line have a error

$ ./configure
checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
checking for gcc... gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for g++... g++
checking whether we are using the GNU C++ compiler... yes
checking whether g++ accepts -g... yes
configure: error: you must configure in a separate build directory

and make

$ make
Makeconfig:85: sysdeps/../config.make: No such file or directory
The GNU C library has not been configured.
Run `configure' to configure it before building.
Try `configure --help' for more details.
make: Failed to remake makefile `sysdeps/../config.make'.

[sklages]

I do that you tell me, but when I wrote ./configure the last line have a error

$ ./configure
checking build system type... i686-pc-linux-gnu
checking host system type... i686-pc-linux-gnu
checking for gcc... gcc
checking for suffix of object files... o
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking how to run the C preprocessor... gcc -E
checking for g++... g++
checking whether we are using the GNU C++ compiler... yes
checking whether g++ accepts -g... yes
configure: error: you must configure in a separate build directory

and make

$ make
Makeconfig:85: sysdeps/../config.make: No such file or directory
The GNU C library has not been configured.
Run `configure' to configure it before building.
Try `configure --help' for more details.
make: Failed to remake makefile `sysdeps/../config.make'.

You should find both libraries in your Ubuntu repos; have a look at the links I sent, in which one you wil have to search ...

If you really want to compile on your own, create an extra build dir:

mkdir build
cd build
../src/configure --help

or whatever ...

[Daniel Fernandez]

Well, when I run the tgicl, the first cluster have more than 1700 reads (and these are assembled in more than 200 contigs) and the rest have less than 100 reads.

I only run: tgicl AlpEST.fasta

I check the tgicl.cfg, may be here is the problem. I thought the tgicl compared with all Gene Index Database from TIGR by default, or I can wrote the database (all mammals for example)

-----------------------
# do not perform all-vs-all search, but search DB_FILE against
# TARGET_DB_FILE instead. Stops after the pairwise hits are generated.
# TARGET_DB_FILE =
-----------------------

[sklages]

Well, when I run the tgicl, the first cluster have more than 1700 reads (and these are assembled in more than 200 contigs) and the rest have less than 100 reads.

I only run: tgicl AlpEST.fasta

I check the tgicl.cfg, may be here is the problem. I thought the tgicl compared with all Gene Index Database from TIGR by default, or I can wrote the database (all mammals for example)

-----------------------
# do not perform all-vs-all search, but search DB_FILE against
# TARGET_DB_FILE instead. Stops after the pairwise hits are generated.
# TARGET_DB_FILE =
-----------------------

Not sure what you are trying to accomplish ... if you just want to cluster and assemble your sequences you did it the right way. Maybe you want to alter a few parameters?

[Kate.W]

TGICL installation

1. I thought you have commented out the DB_FILE entry in tgicl.cfg? Then it should work as intented, "tgicl AlpEST.fasta"



hth,
Sven

Hello,

I have been trying to make TGICL work. I get the same error as Daniel Fernandez, after commented out the DB_FILE entry in tgicl.cfg, when I try

$ tgicl test.fasta

Use of :locked is deprecated at .... /perl5/TGI/DBDrv.pm line 36.
tgicl (test.fasta) finished on machine
in /test, without a detectable error.

Seems to work for Daniel Fernandez, but not for me. Any idea? Or any alternative to TGICL? Thanks!

[sklages]

Hello,

I have been trying to make TGICL work. I get the same error as Daniel Fernandez, after commented out the DB_FILE entry in tgicl.cfg, when I try

$ tgicl test.fasta

Use of :locked is deprecated at .... /perl5/TGI/DBDrv.pm line 36.
tgicl (test.fasta) finished on machine
in /test, without a detectable error.

Seems to work for Daniel Fernandez, but not for me. Any idea? Or any alternative to TGICL? Thanks!

What is not working? What are you expecting? The message about ':locked' is just a (perl) warning, not an error.

And, alternatives? What kind of data do you have? What do you want to do?

Sven

[Kate.W]

Hi Sven,

Thnaks for your reply! Sorry I pasted the wrong message. So I got the "locked" warning message when I manually insert the fasta file in the config file :

DB_FILE = My.fasta

So doing this managed to make to program work, with a warning message.

When I commented the previous line by doing the following :

# DB_FILE =

and then, inside the folder where the fasta file is, type :

$ tgicl My.fasta

I get the following :
Use of :locked is deprecated at /usr/share/perl5/TGI/DBDrv.pm line 36.
Error: cannot find input file ''

That's what my problem really is. Sorry for the confusion.

I am merging several RNA-seq assemblies. So I have this huge file with multiple isoforms that I want to cluster...

[sklages]

$ tgicl My.fasta

Error: cannot find input file ''

Is this error message complete? I can only see one ".

What does 'ls -l My.fasta' spit out?

Sven

[zarthon59]

$ ls -l My.fasta

-rw-r--r-- 1 userk userk 287940826 Oct 10 17:18 My.fasta

[Kate.W]

Is this error message complete? I can only see one ".

Sven

Yes it is the only error message I get.

[sklages]

OK, next step, how does your config look like?

[Kate.W]

OK, next step, how does your config look like?

ubuntu 12.04
16 cpu
> 100 Gigs of Ram

Is that what you're asking?

[sklages]

No, I was asking for the tgicl.cfg file. But OK .. after years of abstinence I have downloaded and installed current TGICL 2.1 on my Ubuntu 12.04 (64bit) notebook :-)

After reading a few lines in

perldoc tgicl

I found pretty fast, that the program has to be called with the parameter '-F':

-F <fasta_db> the EST file in fasta format.

I just successfully clustered a small sanger dataset (7600 seqs) this way.

If you want to get rid of the warning for your TGICL installation, just add

no warnings 'deprecated';

to DBDrv.pm.

TGICL is not suited for huge (sanger) datasets IMHO; there is no active development.

An alternative may be (also for sanger) wcdest (http://code.google.com/p/wcdest/); that's what we were using some years ago after we ran into problems with TGICL. But it is just a clusterer; you still need an assembler and will still need some scripting.
You may also want to have a look at MIRA assembler for *assembling* EST datasets; it is not a clustering tool, but an assembler.

But there may be others.

hth,
Sven

[Kate.W]

Thanks a lot Sven!

[Raul Maia]

Hi,
I was having the same problem and I did all those steps.
But now,when I run tgicl :

tgicl -F teste.fasta

this appear in console:

not available in PATH!

This could be something wrong with environment variable? I'm confused.

Thanks.