Q: What isn't Apollo?
A: Apollo will not analyze your sequence for you. It will not
directly help you find sequencing errors or SNPs. And it doesn't toast bread.
Q: Why is it called Apollo?
A: Apollo was the Greek god of the sun, music, poetry, prophecy,
agriculture, and pastoral life. Sounds good to us. Or, at any rate,
better than another possible name for our annotation tool: GOAT (GenOme
Annotation Tool).
Q: Is Apollo an applet?
A: No, Apollo is a Java application--you can download the software and install it on your
computer and run it locally.
We have recently enabled Apollo to run as a Web start application (which can be launched from a browser, much like an applet), but we have not yet made this publicly available due to limitations from the data server side.
Q: What platforms does Apollo run on?
A: Windows (98, 2000, XP, etc.), Mac OS X (please note that
Apollo is not intended for use on pre-OS X Macs), and any Unix-type
system (Solaris, Linux, etc.).
Q: Do I need to install Java on my computer in order to run Apollo?
A: Probably not. If you don't already have Java installed, you can download Apollo packaged with the appropriate Java
Virtual Machine (JVM) for your platform. However, if you are running Mac
OS X, you may need to update your JVM--see the next question.
Q: What version of Java does Apollo require?
A: JDK1.5+ is required for all platforms.
If you are
running a recent version of Mac OS X (10.2.3 or later), you should make
sure you have JDK1.5.
Please see Apple's instructions on downloading JDK1.5, using Software Updates.
Q: How much memory does Apollo need?
A: Apollo uses a lot of memory, largely due to
the sheer size of the annotation data.
In order to use Apollo, your computer will need to have at least 164Mb of
RAM. Moreover, even if you have that much memory, you will probably find that if you try
to run more than one instance of Apollo at the same time, performance
will suffer. Also note that if you load a very
large region, Apollo will use up more memory. Keep in mind that other applications running on your
computer also use memory, so you will get better performance if you shut down other applications (Photoshop, Word, etc.) before you launch Apollo.
Q: The installer is not working for me.
A: The installer software (a third-party product) has been
problematic on some varieties of Unix and Mac OS X.
If you are on Linux and are having trouble running the installer, try downloading the Linux installer
without Java VM (and make sure you have Java version 1.5+ installed
on your computer). If the installer still doesn't work, click here for an alternative way to install Apollo.
Q: I am having trouble getting Apollo to run on my 64-bit Linux system.
A: See the answer to the previous question. Also, try installing
a different Java Virtual Machine. One 64-bit Linux user reported trouble using Red Hat's Java, and was
only able to get Apollo to work when he used Sun's Java.
Q: I am having trouble getting Apollo to run on my Mac--the Apollo icon
bounces a few times, but Apollo does not come up.
A: Some security-related software updates on Panther (v10.3) interfere
with Java and prevent Apollo from launching from the shortcut. We are looking into
how to address that, but there are two workarounds:
Q: I want to see the text output from Apollo on my Mac running Tiger. I opened a Console window, as it says in the userguide, but I don't see output there.
A: Something in Tiger has stopped this from working. We are investigating this. For now, you can see the text output by launching Apollo from a Terminal window (see the previous question) rather than by clicking on the icon.
Q: Is there an Apollo mailing list?
A: Yes; you can subscribe here.
In order to reduce spam, only members can send messages to the list.
This mailing list has low traffic and is not used for any purpose other than
Apollo updates and questions. (If you receive spam that appears to be from
the apollo mailing list, it is not really--the address is forged.)
Q: What data formats does Apollo read?
A:The most robust Apollo data adapters read GAME XML (the format used for the
Drosophila annotations), ChadoXML (the version used at FlyBase Harvard) and the Ensembl version of
GFF. The Ensembl group has enabled Apollo to read data
via the Ensembl CGI and directly from Ensembl databases, with the EnsJ/Otter
adapter.
There are also two useful data adapters currently under development. The Analysis adapter lets you read raw analysis results (BLAST, sim4, GENSCAN, etc.) into Apollo. This is mostly working, although the GUI needs work. The GenBank adapter is also mostly working but needs more testing.
New in Release 1.5.0 is a Chado XML adapter. Right now, it is very FlyBase-specific and needs to be generalized, but it would be a good starting point for a group that is using a Chado database.
We recently did some preliminary work towards enabling Apollo to act as a client for DAS/2 servers. It can now read simple DAS2XML files.
Q: Is there a schema for GAME XML?
A:Yes, game.rng. If you find a
game.dtd, please disregard it--it was not a good schema description for GAME XML.
For more about the GAME schema, please see doc/schema-notes in the Apollo
distribution.
Q: Apollo is not working for me--I tried to get Drosophila
melanogaster data and got the error message "Can't connect to URL--server not responding."
A:This is not an Apollo problem--this is a problem with the
FlyBase server that supplies the D. melanogaster annotation data. The server
or the database it accesses may be down, or the traffic may be too heavy.
Another possible issue is that you might have a firewall that prevents the data from getting to your computer. If you use a proxy, be sure to set the proxy by clicking the "Proxy settings" button before trying to load data.
Q: How can I download and run the latest in-development version
of Apollo?
A:
Keep in mind that if you try to run the very latest Apollo, there are no
guarantees that it will work correctly. We are actively developing it,
and often there are many bugs that crop up between releases and then are
fixed. So on any given day, it's possible that the up-to-the-minute
Apollo won't run at all (though that is rare) or that various aspects of
it will be broken.
Occasionally, someone will check in code that doesn't compile; when this happens, it's always fixed quickly (within a day). But if you get lots and lots of errors when you try to make, there's probably something wrong with your environment (not having APOLLO_ROOT set properly is the most common mistake).
Here is the command I use for checking out the whole Apollo distribution from sourceforge:
setenv CVS_RSH ssh cvs -z3 -d:ext:myaccount@gmod.cvs.sourceforge.net:/cvsroot/gmod checkout -d apollo apolloFor "myaccount", substitute your own SourceForge account name. (It's a good idea to get a SourceForge account if you're going to be doing this--it's quick, free, and they don't send you spam.) If you don't want to get an account on SourceForge, you can do an anonymous checkout; however, anonymous checkouts give you a version that's a few days out of date. To check out an anonymous version:
cvs -d:pserver:anonymous@gmod.cvs.sourceforge.net:/cvsroot/gmod login cvs -d:pserver:anonymous@gmod.cvs.sourceforge.net:/cvsroot/gmod checkout -d apollo apollo
The "-d apollo" specifies the directory name you want the Apollo stuff to go in, so you can change that if you want to put it somewhere else.
After you've checked out the code, compiling it is straightforward:
cd apollo
setenv APOLLO_ROOT `pwd`
cd src/java
make clean [only necessary if you've already compiled and then updated]
make jar
You can also use Ant to
compile the source.To run your newly compiled apollo:
~/apollo/bin/apollo(where ~/apollo is the directory where you've installed apollo--change as appropriate).
Note that all my instructions assume that you're on a Unix-y system (which includes Macs). If you're on Windows, I have no idea what to do. I've never tried to compile on Windows, nor to run anything other than the installed version of Apollo. I gather you can use Ant to compile on Windows.
Q: Can I customize Apollo for my own use?
A: Yes. There are many user-customizable options; see the user guide for more information.
If you are a Java developer and wish to customize the Apollo code (for
example, by adding a new data adapter to read a different data format),
you can download the source from SourceForge (see
instructions above).
Q: Where can I find Apollo developer documentation?
A:
Although Apollo has extensive user documentation,
the documentation for developers is more spotty.
You can look at the javadocs,
which are a bit sparse but might still be helpful. If you are
interested in data adapters, look at the data adapter "cookbook"
(doc/html/dataadapter_cookbook.html).
Also, the Genome Biology
paper about Apollo has a diagram and discussion of the Apollo architecture.
Q: I would like to annotate the naked mole rat genome using
Apollo. What do I need to do?
A: Well, first you need to run some sequence analysis tools
(such as BLAST, GENSCAN, etc.) on your sequence data. We are currently developing
a raw analysis results output loader that can load BLAST results and some
other output formats.
The only data formats that fully support editing right now are GAME XML and Chado XML (Here is an example of annotations and results stored in GAME XML format.)
At the BDGP, we run sequences through an analysis pipeline that finishes with a program called BOP, which parses the output from the various analysis programs and does some filtering and then saves the results as GAME XML. BOP is now part of the Apollo codebase--however, BOP has not been debugged and robustified the way Apollo has, so you'll be navigating rougher terrain if you try to use it. If you've checked out the Apollo code from SourceForge, you can run BOP from the command line by using the bin/apollo shell script with the "-bop" option. BOP is also used in the Analysis Adapter, which lets you read raw analysis results (BLAST, sim4, GENSCAN, etc.) into Apollo. We are still debugging this.
Q: Can I add annotations to the human genome using Apollo?
A: At the moment, the interface to the human genome
annotations at Ensembl is
read-only: you can view the annotations but you can't edit them or add
new annotations. We hope to provide that capability soon.
You can, however, edit the Drosophila annotations (or any other annotations that are stored in GAME XML or Chado XML format, or in a Chado database). If you wish to submit corrections to our Drosophila annotations, you can mail them to flybase-updates at morgan.harvard.edu.
Q: I found a bug in Apollo.
A: We are shocked and horrified; Apollo is 100% bug-free.
Well, ok, that's completely untrue. You can report your bug using the
SourceForge Apollo bug tracker.
Please be as specific as possible when you
report bugs--tell us what dataset you were looking at, what operating system
you're on (e.g. Mac OS X), and what you did that caused the bug. Also send us the
output from the Java console (see the userguide for instructions
on accessing the console).
Be sure to include your email address so we can contact you for more information about the
bug and notify you when we fix it.
We periodically (every month or two) release a new and improved version of Apollo;
visit the download page to get the latest version.
Q: I have made some changes to Apollo that I would like to commit
to the main codebase.
A: We are happy to look at bug fixes and new additions made by
members of the community, and in fact have already incorporated several of
these. Please send your proposed changes to the Apollo mailing list (include any
classfiles or other files that have changed, as well as a description of
what you've changed) and we will review your changes and, if they seem good,
incorporate them into the codebase. (Before sending your changes, please
do a cvs update to make sure your version of apollo is completely up to date.)