Fwd: Need help with running BLAT on CentOS
Aneesha Das
From: Aneesha Das <dasan...@gmail.com>
Date: Thu, Sep 25, 2014 at 8:48 PM
Subject: Re: Need help with running BLAT on CentOS
To: Jim Kent <ke...@soe.ucsc.edu>, cbs...@soe.ucsc.edu
Looks like you need to setenv MACHTYPE, and also do a make in the lib directory. Instructions for this should be in the README.On Sun, Sep 7, 2014 at 7:28 AM, Aneesha Das <dasan...@gmail.com> wrote:Dear Dr.Kent,I work in Dr.Saikat Chakrabarti's lab at the CSIR-Indian Institute of Chemical Biology. I have been trying to install BLAT in CentOS 5.8 server. I have downloaded the compressed version of the program (blatSrc35.zip), uncompressed the folder. But when I enter the blatSrc directory and type make (as user root), the process displays the following messages:make[1]: Entering directory`/hpchome1/saikat/SOFTWARE/EVM/PASA_r20140417/blatSrc/blat'make[1]: *** No rule to make target `../lib//jkweb.a', needed by `blat'. Stop.make[1]: Leaving directory `/hpchome1/saikat/SOFTWARE/EVM/PASA_r20140417/blatSrc/blat'make: *** [all] Error 2I had made a directory using the $MACHTYPE variable (x86_64-redhat-linux-gnu) in the lib directory, but the file jkweb.a could not be found anywhere.I badly need help with the installation and running of this program (which is a part of the PASA pipeline, which I am trying to run).Help!!Aneesha Das.
Steve Heitner
Hello, Aneesha.
Please run the following commands:
rpm -qa > someFileName
gcc -v
uname -a
Then please send us the file output from the first command and the output from the other two commands. You can send it directly to me if you would prefer that it doesn’t go to the list.
Please contact us again at gen...@soe.ucsc.edu if you have any further questions. Questions sent to that address will be archived in a publicly-accessible forum for the benefit of other users. If your question contains sensitive data, you may send it instead to genom...@soe.ucsc.edu.
---
Steve Heitner
UCSC Genome Bioinformatics Group
--
Aneesha Das
Thanks for replying. I am on vacation in Jakarta right now, so I will be able to send you the files when I return to work on 7th October. Do let me know if this is okay with you.
Regards,
Aneesha.
Steve Heitner
Hello, Aneesha.
Yes, this would be fine. We’ll look to hear from you when you return.
Please contact us again at gen...@soe.ucsc.edu if you have any further questions. All messages sent to that address are archived on a publicly-accessible Google Groups forum. If your question includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.
Aneesha Das
Jonathan Casper
Hello Aneesha,
This is definitely a puzzle. Your environment is quite similar to our development machines, but we can't seem to reproduce your errors. You could instead try building BLAT using the source code in our userApps package at http://hgdownload.soe.ucsc.edu/admin/exe/. There are also some precompiled binaries for several architectures in that directory. If your system really is that similar to our development environment, you might be able to use those and skip the compilation step altogether.
I hope this is helpful. If you have any further questions, please reply to gen...@soe.ucsc.edu or genome...@soe.ucsc.edu. Questions sent to those addresses will be archived in publicly-accessible forums for the benefit of other users. If your question contains sensitive data, you may send it instead to genom...@soe.ucsc.edu.
--
Jonathan Casper
UCSC Genome Bioinformatics Group
--
Aneesha Das
Jonathan Casper
Hello Aneesha,
You do not have to install gfServer and gfClient to run BLAT. You may find, however, that gfServer and gfClient are a better way to use BLAT depending on your needs. The gfServer daemon keeps some data in memory, which can result in faster alignments if you plan to run a series of alignments against the same target sequence. Otherwise, BLAT will run fine on a single compute node if that is what you want to do.
I am still curious about your problems compiling BLAT from the source code in blatSrc35.zip. Can you please go into the blatSrc/lib/ directory and run the following command?
env CFLAGS="-E" make net.o
Then send me the resulting net.o file. It might help us figure out where the problem is coming from.
I hope this is helpful. If you have any further questions, please reply to gen...@soe.ucsc.edu or genome...@soe.ucsc.edu. Questions sent to those addresses will be archived in publicly-accessible forums for the benefit of other users. If your question contains sensitive data, you may send it instead to genom...@soe.ucsc.edu.
--
Jonathan Casper
UCSC Genome Bioinformatics Group
Aneesha Das
Jonathan Casper
Hello Aneesha,
I'm somewhat confused by your error output. Running env CFLAGS="-E" make net.o should have added -E to the list of options for gcc. In your output, however, it's clear that the -E option is not being used. If you are using one of the C shell variants (csh, tcsh), you can instead try running
setenv CFLAGS "-E" make net.o unsetenv CFLAGS
The goal of this command would be to see what how the gcc preprocessor is defining the socklen_t variables, as you may have conflicting library definitions. The output of the preprocessor should appear in the net.o file instead of its usual contents (object code).
That said, there is no guarantee that this will lead anywhere useful for you. If you are able to run one of our precompiled BLAT binaries from http://hgdownload.soe.ucsc.edu/admin/exe/, that is a better plan.
You will have to check with the PASA developers about where it expects to find the blat binary - we didn't make PASA, and we can't support it.
I hope this is helpful. If you have any further questions, please reply to gen...@soe.ucsc.edu or genome...@soe.ucsc.edu. Questions sent to those addresses will be archived in publicly-accessible forums for the benefit of other users. If your question contains sensitive data, you may send it instead to genom...@soe.ucsc.edu.
--
Jonathan Casper
UCSC Genome Bioinformatics Group