**5. Conclusions**

In vertebrate mitochondria, BLAST analyses of peptides translated from frames that are not recognized ORFs and contain stop codons align with high homology with proteins translated from regular mitochondrial ORFs in GenBank. Many such ORFs code for peptides matching usually noncoding sequences and occur in the mitogenome of *Lepidochelys olivacea* [9]. In this case, similar analyses are done for invertebrate mitochondria and a mitochondrial genome (JX566506, from *Aleurodicus dispersus*, Yu and Du, submitted in GenBank 2012, unpublished) considered as unverified are detected, probably because many protein-coding genes are undetectable with usual coding rules.

Several phenomena, and their combination, explain this situation. Alignment analyses detect the coding rules for these genes: frameshift, translation of stop codons, and depletion of stop codons in usually noncoding frames. Previous analyses detected in *Lepidochelys olivacea* CytB two stop-codon-deprived frames on the sense strand, among which one codes for the regular cytochrome B, and the other for an unknown protein. GenBank annotates erroneously the latter frame as coding for cytochrome B. A similar observation is reported here for the gene AT8 in *Aleurodicus dispersus*.

Some mitochondrial protein coding genes in *Aleurodicus dispersus* are unusual in the sense that the stop-codon-depleted frame erroneously annotated as the regular mitochondrial protein coding gene is on the strand opposite to the sense strand coding (with or without stop codons) for the actual usual protein: ND1, ND4l, ND4, ND5. The process depleting stop codons in these antisense frames is unknown and of particular interest. It is a probable combination of natural selection and enzymatically directed mutations to and from stop codons in the adequate nucleotide contexts, perhaps promoted by unknown conditions specific to *Aleurodicus dispersus*. In some genes, the protein usually coded by the regular genetic code necessitates translating stop codons (a stopless genetic code), while frames including stop codons and therefore not considered as ORFs become stop codon depleted, and hence corresponding peptides are coded by the regular invertebrate mitochondrial genetic code. This situation where peptides coded by regular and stopless genetic codes are swapped might reflect a reversal in hierarchies of needs for the expressions of the respective peptides, specific to *Aleurodicus dispersus*. The requirement for tRNAs translating stop codons would regulate these respective expressions, *de facto* swapping between regular and stopless genetic codes. I suggest that the enzymatically directed stop codon depletion is related to the process that caused directed introductions of stop codons in the coding frames of HIV proteins integrated in the nuclear genomes of "elite" HIV controller individuals [68, 69].
