Skip to first unread message
CharlesEGrant
Oct 26, 2012, 7:30:54 PM10/26/12
to
Several users have asked if there is any way they can submit more then 60kb of sequence data to the public MEME server. MEME's running time grows quadratically in the length of the input sequence data. Allowing larger sequence files to be submitted to the public web site would run the risk of creating backlogs, as the jobs with larger sequence data sets would monopolize the servers.
If you need to analyze sequence data sets larger than 60kb, we recommend installing a local copy of the MEME software. When using the command-line version of MEME, the maximum sequence data size can be set using the '-maxsize' option. If you have some UNIX background, the installation of the command-line tools is straight-forward. The software can be downloaded from
http://meme.nbcr.net/meme/meme-download.html
An input file of about 100kb will take about 20 minutes per motif on typical, commodity, hardware. If your system has installed LAM or MPICH you can take advantage of the parallelized version of MEME, which can greatly reduce the running time of your analysis.
See this post: https://groups.google.com/forum/?hl=en&fromgroups=#!topic/meme-suite/QJiJsy1QxYk for tips about running MEME on large datasets.
If you need to analyze sequence data sets larger than 60kb, we recommend installing a local copy of the MEME software. When using the command-line version of MEME, the maximum sequence data size can be set using the '-maxsize' option. If you have some UNIX background, the installation of the command-line tools is straight-forward. The software can be downloaded from
http://meme.nbcr.net/meme/meme-download.html
An input file of about 100kb will take about 20 minutes per motif on typical, commodity, hardware. If your system has installed LAM or MPICH you can take advantage of the parallelized version of MEME, which can greatly reduce the running time of your analysis.
See this post: https://groups.google.com/forum/?hl=en&fromgroups=#!topic/meme-suite/QJiJsy1QxYk for tips about running MEME on large datasets.
Reply all
Reply to author
Forward
0 new messages