Bioinformatic analysis suggests that the Orbivirus VP6 cistron encodes an overlapping gene

Table 2 ORFX MLOGD statistics. MLOGD statistics for ORFX in different Orbivirus alignments. These statistics were derived using MLOGD in the 'Test Query CDS' mode (Figure 3) – specifically testing the coding potential of the whole ORFX – rather than the 'Sliding Window' mode used for Figure 2.

Species	Reference¹	N_seqs	Length	ln(LR)²	var/nt³	ln(LR)/nt⁴	$N_{var}^{5}$	${div}_{\max}^{6}$
BTV	NC_006008	48	234 nt	101.8	0.77	0.44	180	0.21
AHSV	NC_006019	3	429 nt	15.8	0.06	0.04	26	0.05
PALV	NC_005992	11	180⁷ nt	29.7	0.23	0.16	41	0.12
PHSV/YUOV	NC_007753	2	336 nt	33.0	0.56	0.10	189	0.56

1. GenBank reference sequence used for MLOGD.
2. Total MLOGD log likelihood score – positive values indicate that ORFX is likely to be coding. Formally, exp(ln(LR)) gives $\frac{P (alignment | ORFX coding)}{P (alignment | ORFX noncoding)}$ , which may be equated to $\frac{P (ORFX coding)}{P (ORFX noncoding)}$ if equal Bayesian priors are assumed. These probabilities are, however, subject to the assumptions of the MLOGD sequence evolution model [15]. Nonetheless, extensive tests with known single-coding and double-coding sequences indicate that 'N_var ≥ 20' and 'ln(LR)/nt ≥ $\frac{1}{6}$ × var/nt' signals robust detection of an overlapping same-strand CDS [16] (and unpublished data).
3. Alignment divergence per nucleotide – i.e. mean number of independent base variations per alignment column in the ORFX region.
4. Log likelihood score per alignment column.
5. Approximate total number of independent base variations in ORFX region.
6. Maximum pairwise divergence from the chosen reference sequence.
7. Alignment of PALV partial sequences – does not cover the entire ORFX region.

ISSN: 1743-422X