Analyse sequences >60000

[paolo lorenzini]

Hi

I have a set of FASTA sequences in which I would like to do some motifs analysis, however the total size of the sequences is more then 60000 nucleotide which is the maximum allowed on the online MEME website.

Should I install MEME in linux and if yes how to run the analysis?

regards

Paolo

[CharlesEGrant]

Yes, we have to limit the size of jobs submitted to the MEME web application to 60kb because larger jobs would end up monopolizing the server. One solution is to install a local copy of MEME. However, be aware that running MEME on sequence databases larger than 1MB may not be practical.

This posting on the MEME Q&A site may be helpful:

MEME: Dataset size and run time issues

The source for the MEME Suite can be downloaded here.

The installation guide can be found here.

The documentation for the command line version of MEME can be found here.

Alternatively, if your sequences are highly redundant, and the motifs centrally located, (typical of ChIP-Seq data), you may be able to use MEME-ChIP rather than MEME. The MEME-ChIP web application will accept sequence files up to 50MB. Internally it will sample 600 sequences at random from the full dataset, and then trim the sampled sequences to their central 100bp before analyzing them with MEME. A protocol for using MEME-ChIP can be found here.