|
 CBRG - FAQs
CBRG accounts (molbiol):
Error Messages:
Bioinformatics on Unix: - SEE ALSO: bioinformatics faq
What is a FAQ?
A FAQ is a Frequently Asked Question together with the answer to that question. Including a list of answers
to the most common queries received at the CBRG should help both new and experienced users alike.
top | back
How can I contact the CBRG?
You can contact the CBRG via email (genmail@molbiol.ox.ac.uk).
Alternatively, the members of CBRG can be found at either the Weatherall Institute of Molecular Medicine or at
the Sir William Dunn School of Pathology (Room 214.00.14). More...
top | back
CBRG accounts (molbiol):
How do I get access the bioinformatics tools provided by CBRG?
An account with the Computational Biology Research Group,
allows you to log on to our server to use the bioinformatics tools that we provide.
top | back
How do I acknowledge CBRG?
It is a condition of the service that any work done using the CBRG computer systems or with the
assistance of the CBRG team should be acknowledged in publications or presentations in the following way:
"I/We wish to acknowledge the Computational Biology Research Group, Medical Sciences Division, Oxford
for use of their services in this project.".
More...
top | back
How do I connect to my account?
Some of the bioinformatics tools provided by CBRG are available on our UNIX server (called "orac").
A secure connection to orac can be made via SSH using the settings for PC's or for macs.
top | back
How can I speed up my Xwindow displays?
Problems have been reported with the speed of Xwindow applications for molbiol account holders when logging on to the
server orac from Microsoft Windows machines. These applications are generally those that rely upon graphics for their interface
rather than a command line, and include nedit, jalview, jemboss and the Staden group of
programs. This problem does not seem to affect connections from macs.
For Windows users, the speed of these applications may be increased by reducing the level of security for these Xwindow applications
(after the important secure user log on has taken place). You can set up your account to do this using the command:
mydisplay
This command will reset the level of security for X applications for all connections to orac from now on and so only has to
be carried out once (although you will be prompted by your pc for permission each time you start an X application).
Note: You must be aware that although the act of logging on to orac will remain secure after the display has been set up
using this command, any information transmitted via the X application thereafter will not. Your data will be transmitted
with the same low level of security associated with telnet connections.
top | back
Why doesn't my old alias work/How do I set up an alias?
On enterprise, it was possible to set up an alias in your account to avoid repeatedly typing the same command. These were
generally implemented by editing either the .cshrc or .login files. On the new machine, Orac, the
default login shell is "Bash" (rather than C shell) and so you must now edit the .profile file in order to
set up an alias. This file is not found by default in your new account and so you will have to create it.
The syntax required is also slightly different - for example an alias to override the behaviour of the ls
command would require the following addition to the .profile file (note the addition of the 'equals' sign which was
not required on enterprise):
alias ls="ls -F"
top | back
Error Messages:
How do I overcome this error: "Gtk-WARNING **: cannot open display:"?
When logged into orac from a PC, you may see this error message when attempting to use any X-application software (Staden, Stars,
nedit etc.). This can be cured by starting eXceed before attempting to run the application again (Note, occasionally it might also be
necessary to log out of orac and re-log in to your molbiol acount).
See "Installing and configuring eXceed" for further details.
top | back
How do I overcome these errors:
- "X Error of failed request: BadAtom (invalid Atom parameter)"?
- "X Error of failed request: BadAccess (attempt to access private resource denied)"
Either of these errors may occur when attempting to use any X-application software (Staden, Stars, nedit etc.) if you have logged
on to orac from a mac or a linux-based machine. The problem can be cured by logging out and logging back in again using the following command
(replacing username for your username on molbiol):
ssh -Y username@orac.molbiol.ox.ac.uk
The error is caused by a default setting in OpenSSH 3.8 or later, that is included with Mac OS X Tiger or has been installed on your
Linux/UNIX machine.
top | back
Why do I get, "command not found" when I type, "staden"?
On enterprise, in order to use any of the Staden programs ("trev" or "gap4" etc), you had to
first type staden before the name of the program. On orac, this is not necessary and you can simply type the name
of the program you want to use directly. For example, to use the Staden program, "pregap4", just type:
pregap4
top | back
What does the STARS error message, "Programme in use, please try again later" mean?
This message normally indicates STARS is running and is "locked" to prevent data being overwritten.
If you are sure that you or no-one else is running the program, delete the file called LOCK in the project's data directory (~/PROJECTROOT/project-name/data/LOCK).
Please note, your permissions may not allow you to delete the LOCK file created by someone else.
You may also see the related message - "Please finish your current session first". If you are sure you are not in a session,
delete the file session.log (in the ~/staden/session/ directory) and start again. Again under normal circumstances this file is used to prevent
data being overwritten.
To prevent these messages occuring in the future, please always shut down STARS using the quit button within the application (rather than the
"X" on the window bar).
See also:
top | back
Bioinformatics:
SEE ALSO: bioinformatics faq
How can I make a multi-fasta file of my sequences on molbiol?
You can use the UNIX command cat ( concatenate ) to make a fasta file containing more than one sequence. This
example assumes that you have a number of query sequences in separate files ("human.seq", "horse.seq",
"mouse.seq") and that you would like to construct a single file called "all_seqs.fasta" containing all of the sequences:
cat *.seq > all_seqs.fasta
top | back
How can I make my own BLAST database?
A custom BLAST database can be made using your own data in two stages. Firstly, all sequences should be assembled in one place.
Apart from downloading this data from elsewhere, the most convenient way to do this is to construct a multi-fasta file containing
all of the sequences you would like in the database ( see FAQ: How do I make a multi-fasta file....? ).
This file is then used to construct the indexes for the BLAST database using "formatdb".
formatdb -p F -i all_seqs.fasta -n customBLASTdb
In the above example, formatdb is used to construct a BLAST database called "customBLASTdb" from a fasta file called "all_seqs.fasta"
containing multiple nucleotide sequences. More...
top | back
How do I run BLAST on molbiol?
We have NCBI's blastall installed on molbiol. In this blastall example command line, the nucleotide BLAST database called
"ensembl_mouse_cdna" is searched using a nucleotide query sequence contained in a file called "human.seq". The results are
output as a file called "human_search.blast":
blastall -p blastn -d ensembl_mouse_cdna -i human.seq -o human_search.blast
There are many modifications possible to this basic command line. More...
In the above example, the BLAST search used one of the BLAST databases that are
available to all users on orac. It is also possible to produce a custom BLAST database using your own sequence data (see above
FAQ: "How can I make my own BLAST database?").
top | back
How do I download sequences into my account?
Database list - The CBRG maintain up-to-date copies of the major sequence databases. To see the databases
that are currently available (and the abbreviations for each one - used when downloading sequences), type:
showdb
The database list produced is also available when you first start GCG on orac.
seqret - You can use the EMBOSS program, "seqret" to retrieve sequences from these databases.
The sequences will then be available for further analysis as separate files in your account. For example, to download the human thymidine
kinase DNA sequence from the RefSeq database, type:
seqret rf:NM_003258
After following the on-screen prompts, this will create a fasta-format sequence file in your account. You can modify the output sequence
format of the resulting file. For example, to download the human adenosine kinase protein sequence from the
SwissProt database and to save the resulting file in GCG format, type:
seqret sw:ADK_HUMAN -osformat gcg
fetch - For people accustomed to using the GCG program, "fetch", we have created a fetch-like
program that works in a similar way. To use this program, type (once you have started GCG) "fetch dbname:accession" and follow
the prompts on screen. Fetch will download the sequence from the database and create a gcg format file in your account. For
example, if you wish to download the human hexokinase sequence from the EMBL database as a GCG format file, type:
fetch embl:AF016357
|
