In response to Hogg’s comment, human Herpes virus 6 can be transmitted from parent to child via chromosomal integration, so yes, our genomes can enlarge. Tanaka-Taya et al 2004 http://www.ncbi.nlm.nih.gov/pubmed/15170644, and given the extensive homology between human and viral DNA, must have been doing so for millions of years. Infection, which may insert multiple gene fragments at once may be what has been, and still is, driving evolution.

The protein homology is also critical to our understanding of disease: Current viruses express proteins that are similar to our own, and upon infection will be able to act as dummy ligands, decoy receptors or enzymes, and will also interfere with protein/protein interaction networks. Such homologous viral inserts are likely to play a key role in a multitude of human diseases.

The BLAST analyses, representing over a million such alignments, can be viewed at http://www.polygenicpathways.co.uk/BLASTS.htm

and some of the consequences are discussed in a series of prepublications at NaturePrecedings

http://precedings.nature.com/users/c8e325bfe6bd1d4fc6e006ff204b5f19

I was visiting my mom and telling her about these findings, and she asked if it was possible some viruses originated in our (animal) DNA and were incorporated by other vectors and eventually broke free to spread to other organisms like a true virus. And, she wondered, since so much behaviour seems to be programmed into the brain (e.g. birds, fish migrating to areas they’ve never been before), is it possible for some types of knowledge to be transmitted in a viral-like manner?

My limited knowledge on the subject thinks ‘not likely’, but fun to think about, and maybe even inspire a fiction story.

[CZ: Matt--I'd be very doubtful of that. These insertions into the genome that get passed down through the germline are extremely rare. Over millions of years they do pile up, but on the scale of years, they almost never happen. What's more, influenza and cold viruses did not turn up in this survey.]