######################################################################
######################################################################

This directory contains all datasets needed for the reproduction of
the figures presented in the paper :

Palmeira, L.,Guéguen, L. & Lobry, J.R., UV-targeted dinucleotides are
not depleted in light-exposed Prokaryotic genomes. (in prep.)

#######################################################################

For any complementary information, namely on how these datasets were
obtained, please contact : palmeira at biomserv dot univ-lyon1 dot fr

#######################################################################
#######################################################################

1. Systematic study:

"zcodonCDS" and "zinterg" files correspond to the mean z-scores on
coding (zcodonCDS) and non coding sequences (zinterg) as computed on
all available genomes retrieved from the Genome Reviews database on
2005-06-16.

The two files have the same structure: 16 columns and 242 rows. Each
of the columns corresponds to one of the 16 dinucleotides, each of the
rows corresponds to one chromosome of a specific bacteria. For
example, if a bacteria contains three chromosomes, it will appear on
three different rows, as "Genre_species_A", "Genre_species_B" and
"Genre_species_C". The values contained in the table are the mean
z-scores as computed on all coding sequences (zcodonCDS) and on all
non coding sequences (zinterg) for the specified chromosome and
dinucleotide.

2. Prochlorococcus marinus as a model organism:

"zrhotheo_cds_AE017126", "zrhotheo_cds_BX548174" and
"zrhotheo_cds_BX548175" files correspond to the z-score statistics
computed on all coding sequences obtained from the three strains of
Prochlorococcus marinus mentioned in the paper, which are
respectively: SS120 strain (AE017126), MIT 9313 (BX548174) strain and
MED4 (BX548175) strain.

The three files have the same structure: 16 columns -- each
corresponding to a dinucleotide-- and a number of rows equal to the
number of retrieved coding sequences (1627 for MED4 strain; 1396 for
SS120 strain; 1340 for MIT 9313 strain). The values contained in the
table are the zscore statistics computed on each coding sequence, and
for each dinucleotide.