Thursday, November 29, 2007

A little Knome fact!

Well, after fighting to keep the name Helix Health of Connecticut. I have realized that just because you have a catchy name does not mean someone else doesn't have it too.

Thus is the case with Knome. I could ramble on about how whole genome sequencing is superior to the limited SNP analyses currently available. This superiority is very important for personalized medicine, but not quite important for a cocktail party. I could have posted some of the press release, but Hsien, Berci and the rest of the DNANetwork do a great job of covering this sort-of new entry to corporate genomics.

Instead, I wanted to focus on the fact that another Knome exists. In fact several Knomes exist. Some even have vice-presidents. Well it turns out Knome has been around a little while before this. In fact a company called Cambridge genomics has filed an application for trademark of the name Knome in August of this year. Smart, unfortunately I did the same thing and have met challenges. Well, they should be ok, so long as their aren't any other software apps out their with knome in their knome...oooops

The Sherpa Says: With the decreasing gene count, clearly the secrets in the genome lie in other places. These include copy number variation, methylation, and other epigenetic changes. You see, the genome is constantly in flux and change. This is why a genome without a solid knowledge set behind it is as useless as a map without a key. But it's just about as much as 2 Ferrari F430s. Living in Greenwich CT, believe me, I have seen my share of his and her Ferraris. Why not his and her genomes.....What a great Christmas present for the person who has everything else, including 17th Century Samurai Swords...p.s. Why hasn't anyone mentioned SmartGenetics yet?


Keith Robison said...

Copy number variation is an interesting point. If you really could get end-to-end sequencing then Knome's product would tell you all the CNV, because it would show up as such.

With shotgun data, you might be more likely to just see a bunch of funny breakpoint sequences & perhaps in some cases note that the frequency of some fragments is outside of the expected range -- and from those together you would have varying confidence estimates of the copy number in various places.

Steve Murphy MD said...

Funny breakpoint sequences sounds like a concept similar to MSI.....It is high in approximately 10% of non-mismatch repair colorectal cancers...
A rough but decent estimate