DRIM (Discovery of Rank Imbalanced Motifs) is a tool for discovering short motifs in a list of nucleic acid sequences.
DRIM was originally developed for DNA sequences and successfully applied on ChIP-chip and CpG methylation data.
The current version has enhanced functionality and can be applied for both DNA and RNA.
This new version was used to predict UTR motifs and Splicing Factor binding motifs based on RIP-Chip or CLIP data.
From a mathematical point of view, DRIM identifies subsequences that tend to appear at the top of the list more often than in the rest of the list. The definition of TOP in this context is flexible and driven by the data. Explicitly - DRIM identifies a threshold at which the statistical difference between top and rest is maximized. An exact p-value for the optimized event is also provided.
As an input, DRIM receives a list of DNA or RNA sequences that are ranked according to some criterion (for example TF binding signals as measured by Chip-chip or 3'UTR sequences ranked according to differential expression levels).
The new version of DRIM allows for various degrees of accounting for multiplicities in computing motif occurrences. This feature is particularly useful in the RNA context, where motifs can be rather redundant and occur in multiple copies.
The output of DRIM consists of the enriched sequence motifs and of an indication of the statistical significance. PSSMs are also provided.
DRIM is jointly maintained by the Yakhini Lab and the Mandel-Gutfreund Lab, at the Technion.
If you use DRIM please cite:
E. Eden, D. Lipson, S. Yogev & Z. Yakhini. Discovering Motifs in Ranked Lists of DNA Sequences, PLoS Computational Biology, 2007.
NEW - Also visit our new tool - DRIMUST
For more information please contact: email@example.com