OpenGE 0.1.1 released

[leecbaker]

We're pleased to announce that version 0.1.1 of OpenGE is available for download. This release contains some performance work, minor new features, and bug fixes.

New in this release:
* Mark duplicates performance has increased. We now split by chromosome in some cases, and process some chromosomes in parallel. This can be disabled with a --nosplit option.
* Output file format is detected from file extensions. For example, any filename ending in "SAM" or "sam" will be written in SAM format.
* The convert command has been removed, as its functionality was present elsewhere. See the command documentation for more details.
* The statistics command can now print out the distribution of read lengths. See the command documentation for details.
* Several small bug fixes.

OpenGE can be downloaded from the Github downloads page:

https://github.com/adaptivegenome/openge/downloads

Bug reports can be put on the Github issues page:

https://github.com/adaptivegenome/openge/issues

We would appreciate any help testing the code or comparing the speed to related tools. Let us know of any improvements or features you would like to see in future versions to make the software more useful to you.

[leecbaker]

As we continue to work on more new features, we want to let you know that we have added FASTQ as a supported output format, in addition to the SAM and BAM support that we currently have.

Support for CRAM is pending (we are waiting for the final specs from EBI).

[Richard Finney]

I think I grabbed a github (then) current(?) snapshot ..

-bash-3.00$ ./openge version
Open Genomics Engine 0.2-dev
Built on Jul 26 2012 at 10:52:24
(c) 2012 Virginia Bioinformatics Institute

Is this newer or later, should I grab another, better version ?

Also, I don't get how to control processor usage. Is there a parameter to specify number of threads to use?

BTW .. primary goal is to use the mergesort ... I need a faster mergesort to fix terabyte scale CGI whole genomic data.

Also, does the mergesort build an index at the same time ?

[adaptivegenome]

I think I grabbed a github (then) current(?) snapshot ..

-bash-3.00$ ./openge version
Open Genomics Engine 0.2-dev
Built on Jul 26 2012 at 10:52:24
(c) 2012 Virginia Bioinformatics Institute

Is this newer or later, should I grab another, better version ?

Also, I don't get how to control processor usage. Is there a parameter to specify number of threads to use?

BTW .. primary goal is to use the mergesort ... I need a faster mergesort to fix terabyte scale CGI whole genomic data.

Also, does the mergesort build an index at the same time ?

The current stable release (binaries) is 0.1.2 and yes you can use the source which is 0.2. The new features are not related to mergesort.

You can control how many threads to spawn using the '-t' option. When using mergesort you would also want to specify '-M' so it marks duplicates at the same time.

If you look at the PDF doc for mergesort there are some other things you can do too like:

* choose to compress or not compress temp files
* specify how many temp files to make (by default we make temp files with 500k reads each)
* specifying compression level (default is 6 like samtools)

Particularly you might play with the number of temp files to make. If you increase the 500k default value then more reads go into a file, less files are made and it generally goes faster, but the cost is you need lots of memory.

Please let me know if you have any other questions or need help with anything else.

Final thing --> right now mergesort does not produce an index so you will need to index the file afterwards like you would with samtools. When 0.2.0 is finished we will feature automatic indexing.

[Richard Finney]

PDF file did not come with with snapshot. Just tex files.
I'll look at them, though.

[adaptivegenome]

PDF:

http://cloud.github.com/downloads/adaptivegenome/openge/openge_commands-0.1.2.pdf

EDIT: Link to PDF is now also included in README.MD