1) This program is written in PERL language. So, a computer with UNIX installed in it is needed to run it.

2) It is very important to enter the parameters, and in the right order (that is, first the name of the file containing the sequence, then the file containing the codon usage bias table). Likewise, it has to be checked that both the sequence and the table are written in the DNA code.

3) It is necessary to check that the sequence and table formats are the right ones. At the moment of downloading the table at the web page recommended it is advisable to follow the next steps:

                     - select the organism
                     - submit the obtained table, selecting "standard" format and "A style like Codon Frequency output in
                       GCG"
                     - click on "submit" and save the outcoming table at a .txt file
                     - run TRANSFORMER with the .txt file to get the same table in the right format to run ORFFINDER

4) Once obtained the ORFFINDER outfile, it can be filtered to obtain, in other outfiles, the ORFs that are more probable to be coding. The cut-off is fixed by the user through UNIX commands. It can be applied not to only to the score, but also to the length of the ORFs.

5) The filtered file has to be turned into GFF format to view the results graphically. It can be done using a UNIX command.

6) The GFF file transforms into PS (Postscript) file through the GFF2PS program .

7) The GHOSTVIEW program, which can be found in SHELL, is used to view the PS file at screen.

NOTE: all the necessary UNIX commands are illustrated below.

• ORFFINDER
• TRANSFORMER
• GFF2PS
• Web page that contains codon usage bias tables for many organisms

• We start from SHELL and run ORFFINDER with the following statement:

                            The name of the files is illustrative, it can be chosen.
                            The sequence (first parameter) entry in NCBI can be found here.
                            The usage codon table (second parameter) comes from the web page suggested above.

• The program will run and turn us back a file called "resultados.out" (results.out).

• We filter the results' file with the UNIX "gawk" command. In our case, we have decided to apply a double filter to remove those ORFs with a size ($4) lower than 200 bases and negative score ($5). The result of the filtration is directed to "resultados_filtro.out" (results_filter.out):

         NOTE: The parameters of the filter are changeable.

• We transform the .out file formerly generated to a GFF format file through the following UNIX command, obtaining a file called "resultados_filtro.gff" (results_filter.gff):

• Now it is necessary to turn the GFF file into a PS file, which we can view by GHOSTVIEW. In order to make, we use the program GFF2PS (you can download it from the link placed in the previous section). The command to run  GFF2PS with the default parameters is the following. We redirect the results to the file "resultados_filtro.ps" (results_filter.ps).

NOTE: To modify the format of the .ps file obtained, several GFF2PS parameters can be used. They can be found in the user's manual in the GFF2PS web page.

• To view the PS file we open it with the GHOSTVIEW program from SHELL, using the following command:

• The PS file can be turned into a JPEG one with the next command (optional):

• It is also optional to rotate the image to make easier the interpretation of the results:

ACKNOWLEDGEMENTS

To all the people that helped us (¡ya sabeis quien sois!)