Supplementary data to our manuscript on codon usage bias in flies can be found in the download area. The following is the README from this directory:
   Supplementary material to the manuscript:

   Heger A, Ponting CP. Variable strength of translational selection among 12 Drosophila species. 
   Genetics. 2007 Nov;177(3):1337-48.
   PMID: 18039870
   
   The supplementary material can be accessed from the download area.

   The following files are available:
   
   [species].indices	codon usage indices for all predicted transcripts within [species]
   [species].data	indices and additional info for all predicted transcripts within [species]
   orthologs.data	Data of predicted transcripts linked via orthology to a dmel gene. Indices
   			     are averaged over genes and in-paralogs. The ortholog assignments are	
			     based on OPTIC release 5.
   ribosomes.genes	Gene identifiers for ribosomal protein genes in D. melanogaster used in the 
			     study.
   frequencies.table	Codon frequencies in ribosomal protein genes for each species. Frequencies
                        are normalized for each codon group coding for the same amino acid. 
   weights.table	Relative adaptiveness values used in the CAI computation.

   All these files are tab-separated tables with column headers. Please ignore lines starting with "#'.

   See below for legends. 

   The species names are of the format _vs_dmel and include the following twelve species 
   (all release CAF1):

   dana_vs_dmel6     D. ananassae
   dere_vs_dmel6     D. erecta
   dgri_vs_dmel5     D. grimshawi
   dmel_vs_dmel4     D. melanogaster
   dmoj_vs_dmel6     D. mojavensis
   dper_vs_dmel3     D. persimilis
   dpse_vs_dmel8     D. pseudoobscura
   dsec_vs_dmel3     D. sechellia
   dsim_vs_dmel6     D. similis
   dvir_vs_dmel6     D. virilis
   dwil_vs_dmel1     D. willistoni
   dyak_vs_dmel7     D. yakuba

   Transcript identifiers follow the convention:

   species|transcript_id|gene_id|quality_code

   Legend for file [species].indices:
   
   GENENAME			transcript (;=) name
   is_selected			transcript is high quality transcript and best quality transcript per gene
   is_dominant			transcript is part of dominant set (see ACAI method)
   CAI				CAI using codon usage bias of the dominant set (ACAI method)
   CAIREF			CAI using the traditional reference weights for D. melanogaster
   CAIBIAS			CAI using ribosomal codon requencies
   ENC				Effective number of codons    
   length			number of nucleotides in transcript 
   ncodons			number of codons in transcript
   nA-nZ			amino acid counts
   pA-pZ			amino acid frequencies
   nstops			number of stop codons
   nsites[x]d-nsites[x]d	number of sites with degeneracy x
   ngc				G+C counts
   ngc3				G+C counts at third codon position
   ndgc3			G+C counts at degenerate third codon positions
   n[x]dgc3			G+C counts at [x]-fold degenerate third codon positions 
   pgc				G+C frequency
   pgc3				G+C frequency at third codon position
   pdgc3			G+C frequency at degenerate third codon positions
   p[x]dgc3			G+C frequency at [x]-fold degenerate third codon positions 
   entropy			sequence codon entropy: -sum pi log pi.
   ml[x]			Message length using codon frequency table n.				
   relml[x]			Message length using codon frequency table n divided by number of codons.
   relentropy[x]		-sum pi log qi
   kl[x]			Kullback-Leibler divergence between codon usage in transcript and codon frequency table
						 sum qi * log( qi / pi )

   The codon frequency tables provided are:
   
   0:	dominant.frequencies	= codon usage table based on dominant set (see ACAI method)
   1:	nondominant.frequencies = codon usage table of non-dominant set (see ACAI method)
   2:	selected.frequencies	= codon usage table of high quality transcripts
   3:	uniform			= uniform codon frequencies
   4:	biased.frequencies	= codon usage table based on ribosomal usage
   
   with pi: codon frequency in transcript, and
	qi: codon frequency in codon frequency table.

   The index "relml4" corresponds to the relative message length per codon in the manuscript.

   Legend for file [species].data:

   Note that some fields are redundant with [species].indices

   id				transcript name
   length			number of nucleotides in transcript 
   ncodons			number of codons in transcript
   nA-nZ			amino acid counts
   pA-pZ			amino acid frequencies
   nstops			number of stop codons
   nsites[x]d-nsites[x]d	number of sites with degeneracy[x]
   nsitesd3			number of degenerate sites at third codon position
   ngc				G+C counts
   ngc3				G+C counts at third codon position
   ndgc3			G+C counts at degenerate third codon positions
   n[x]dgc3			G+C counts at [x]-fold degenerate third codon positions 
   pgc				G+C frequency
   pgc3				G+C frequency at third codon position
   pdgc3			G+C frequency at degenerate third codon positions
   p[x]dgc3			G+C frequency at [x]-fold degenerate third codon positions 
   entropy			sequence codon entropy: -sum pi log pi.
   ml[x]			Message length using codon frequency table n.				
   relml[x]			Message length using codon frequency table n divided by number of codons.
   relentropy[x]		-sum pi log qi
   kl[x]			Kullback-Leibler divergence between codon usage in transcript and codon frequency table
						 sum qi * log( qi / pi )
   nAAA-nTTT			codon counts
   rAAA-rTTT			codon frequencies

   The codon frequency tables provided are:
   
   0:	dominant.frequencies	= codon usage table based on dominant set (see ACAI method)
   1:	nondominant.frequencies = codon usage table of non-dominant set (see ACAI method)
   2:	selected.frequencies	= codon usage table of high quality transcripts
   3:	uniform			= uniform codon frequencies
   4:	biased.frequencies	= codon usage table based on ribosomal usage
   
   with pi: codon frequency in transcript, and
	qi: codon frequency in codon frequency table.

   Legend for orthologs.data:
   
   The indices for orthologs.data are identical to [species].indices and [species].data, except that they
   are prefixed by the species name, thus:

   dere_vs_dmel6_p2dgc3 will give you the average number of two-fold degenerate sites at the third codon 
			position in orthologs in D. erecta to a particular gene in D. melanogaster.

   These data were prepared for download at 29/11/2007.