Update - AME & RAMEN now in MEME Suite:
AME is now part of the MEME Suite, and the latest maintained version can be found there. It can either be used as part of MEME-ChIP on the web, or downloaded as part of the source package.
This page is maintained for historical reasons, so that the tool used in the paper remains available. Unless you have a specific reason to use this version of AME, we highly recommend using the MEME Suite version. In particular, this version of AME cannot handle some of the larger motif libraries available today.
A major goal of molecular biology is to determine the mechanisms that control the transcription of genes. Recently, a number of different approaches for determining motifs enriched in a set of regulatory regions have been developed. Most of these approaches require on the biologist to select a set of regulated genes, based on an expression microarray fluorescence level threshold.
We have produced a tool that implements several different algorithms and methods for motif enrichment analysis. This tool, of course, can not only be used for comparing methods (as we did in the paper), but also for investigating potential regulators and co-regulators of gene sets from experiments such as ChIP-chip and microarrays.
More information about this tool, AME, can be found in the paper in which it was initially published. A slight variant to AME, RAMEN (Regression Analysis of Motif ENrichment) is included in the package with more specific options for linear-regression-based analyses.
Several different algorithms which are used in a variety of different motif discovery and motif enrichment tools have been developed and packaged in the tool available here, AME. These methods include:
- Fisher Exact Test
- Multiple-hypergeometric Test
- Rank-sum Test
- Linear Regression
- Spearman Rank Correlation Coefficient
In the study that we performed, we also tested two methods implemented by other researchers:
The tool is currently available for download as a binary for both Mac OS X (compiled under 10.5) and Linux x86 from the downloads page. A slight variant to AME, RAMEN (Regression Analysis of Motif ENrichment) is included in the package with more specific options for linear-regression-based analyses.