Firstly, we have submitted the genomic sequences of type 1 and 4 herpesvirus to FindOrf 2003 version 1.0 (see conditions). As a result, we have predicted 76 orf for EHV1, of which 88 in the complementary chain; and 143 for EHV4, of which 73 in the complementary chain (EHV1 predicted orfs , EHV4 predicted orfs).
In second place we have compared the predicted orfs with the ones annotated in GeneBank and Telford et al. 1998, to be able to identify the predicted orfs already described. The results of this comparison can be seen in Table 1:
Once the described orfs have been identifyed we have submitted all the other orfs to an individualised search using PSH- and PSI- Blast, with the objective to look for possible alignments with other sequences that may be informative of biological function for our non-described orfs. After that search we have obtained alignments with E value superior to the threshold for 12 orfs of EHV1 and 12 of EHV4. Results are shown in Tables 2 (EHV1) and 3 (EHV4).
For each one of these significant alignments obtained by Blast we have aligned conserved the fragments of the analised orf and of the resultant sequences using Clustalw. However, the multiple alignment has shown that in most cases, specially for EHV4, blast results have not been really significative (data not shown) because, although having a significant E value, this is due to higly repeated regions of the viral genome that, analised using blast, are coincident with:
A more profund analysis of this non-significative orf shows that they owe to regions described as repetitive, in lots of cases to TR and IR (see description of EHV genomes). Orf that shows results that suggest further analysis or that ones for which we have obtained alignments that have given significative results are represented in Table 4.
Furthemore, while running blast we have identifyed two putative conserved domains for an EHV1 orf, >gi|M86664|herpesvirus equí 1| 357-818, of which only one is significative, the other belonging exclusively to bacterias taxa. This orf has also shown matches with other proteins (see table 2), but these ones are not related to the domains observed in EHV1 orf. The significative domain contains 4 turns, each containing three imperfect tandem repeats of a hexapeptide repeat motif (X-[STAV]-X-[LIV]-[GAED]-X) (see alignment). This hexapeptide repeats often correspond to enzymes showing acyl-transferase activity, of which many are trimeric in their active form (see 3D structure).
