EntropicProfiler

Homepage:

http://sels.tecnico.ulisboa.pt/ep/

Description:

Entropic profiles of DNA sequences are local information plots of the relative over and under-expression of motifs per position. They are calculated based on Chaos Game Representation (CGR) using a recently proposed fractal kernel and Parzen's window density estimation method. They allow the visualization of motif densities for two different parameters: resolution L and smoothing parameter Φ. This method detects biological significant regions of DNA. An important simplification allows its calculation using segment counts, explored in this application through suffix trees.

References:

Vinga, S. and Almeida, J.S.: Local Renyi entropic profiles of DNA sequences. BMC Bioinformatics 2007, 8:393. Fernandes F., Vinga S., Freitas A.T., Almeida J.S.: Entropic Profiler - detection of conservation in genomes using information theory. BMC Research Notes 2009, 2:72.

Manual:

Usage:

ep -t<type> -f<file> -l<l> -p<phi> -i<position> -m<findmax> -w<window>

Options:

• t : sequence type, should always be 'f', which stands for "file"
• f : fasta file to load, should be less than 2 GB, only the 1st sequence inside is loaded, 'N's are converted to 'A's
• l : L value (resolution length), should be higher or equal to 3 and less or equal to 10
• p : Phi value (smoothing parameter), should be less or equal to 10
• i : position to study
• m : automatically find the value of L that maximizes the function in that position 'i', should be '0' for false or '1' for true
• w : window length to study around that position 'i'
• x : (optional parameter) load previously saved project (".tree" file), should be '0' for false or '1' for true
• y : (optional parameter) study by motif instead of position, should be a string of DNA characters
• z : (optional parameter) get EP values for entire sequence, does not create plots, should be '0' for false or '1' for true

Example:

ep -tf -fExample1.fasta -l8 -p10 -i35840 -m0 -w100