About bTSSfinder

Program bTSSfinder was developed in Computational Bioscience Research Center (King Abdullah University of Science and Technology, Kingdom of Saudi Arabia) and aimed to predict promoters (transcription start sites, TSSs) of 5 classes of sigma factors for Escherichia coli (sigma 70, 38, 32,28 and 24) and Cyanonacteria (sigma A, C, H, F and G).

Command line interface

bTSSfinder can run on Linux or Unix (MacOS).
- GFORTRAN "gfortran" is required.
- You must setup the environmental variable bTSSfinder_Data, for example:

	setenv bTSSfinder_Data /path/to/installation/Data

or

	export bTSSfinder_Data=/path/to/installation/Data

or to avoid doing the above everytime, add the following to your ~/.bashrc or ~/.bash_profile file

	export PATH=/path/to/installation:$PATH
	export bTSSfinder_Data=/path/to/installation/Data/:$bTSSfinder_Data

Running bTSSfinder:

	/path/to/installation/bTSSfinder -i example_Cyanob.fasta -o example_Cyanob -h 2 -r 0 -t c

Command line parameters

bTSSfinder -i  - Input fasta file  * REQUIRED
      [-p  - print (y or Y) or not print (n or N) the query sequence(s): Default: n]
      [-o  - Prefix for output file; Default: bTSSfinder]
      [-x  - a non-overlapping interval to select the highest-scoring TSS (post-processing); >=50,<=300; Default: 300]
      [-n  - Comma-separated A,C,G,T frequencies in the reference genome; Default: 0.25,0.25,0.25,0.25]
      [-a  - Neural Network Threshold for sigma70 or sigmaA promoters;  -2<=parA<=2, Default: preset in training ]
      [-b  - Neural Network Threshold for sigma38 or sigmaC promoters;  -2<=parB<=2, Default: preset in training ]
      [-d  - Neural Network Threshold for sigma32 or sigmaH promoters;  -2<=parD<=2, Default: preset in training ]
      [-e  - Neural Network Threshold for sigma28 or sigmaF promoters;  -2<=parE<=2, Default: preset in training ]
      [-f  - Neural Network Threshold for sigma24 or sigmaG promoters;  -2<=parD<=2, Default: preset in training ]
      [-h  - 1 to search on the sense strand OR 2 for both strands, Default: 1 ]
      [-t  - Taxon (e or E - for E. coli, and c or C - for Cyanobacteria, Default: E ]
      [-c  - Search Criterion; if parC =
                                 70 ... search for sigma70 promoters (E. coli)
                                 38 ... search for sigma38 promoters (E. coli)
                                 32 ... search for sigma32 promoters (E. coli)
                                 28 ... search for sigma28 promoters (E. coli)
                                 24 ... search for sigma24 promoters (E. coli)
                                 A ... search for sigmaA promoters (cyanobacteria)
                                 C ... search for sigmaC promoters (cyanobacteria)
                                 F ... search for sigmaF promoters (cyanobacteria)
                                 G ... search for sigmaG promoters (cyanobacteria)
                                 H ... search for sigmaH promoters (cyanobacteria)
                                 1 ... search for the highest ranking promoter regardless of class in the chosen Taxon (option -t)
                                 5 ... [Default], report all promoter classes in the chosen Taxon (-t)

The output file in the BED format has the following fields:

(1) The DNA segment identifier usually the fasta ID (e.g. chr1, gene1).
(2) The start position of the predicted TSS.
(3) The end position of the predicted TSS.
(4) Feature name - TSS#.
(5) Score - 0.
(6) Strand '+' or '-'.
(7) thickStart - TSS# start position.
(8) thickEnd TSS# End position, the same as (7).
(9) An RGB of the form R,G,B (set to 255,0,0).
(10) The number of main promoter blocks (in our case =2, -35 and -10 boxes).
(11) A comma-separated list of block sizes(in our case = 6,6).
(12) A comma-separated list of block start positions (calculated relative to the starting position of the query).