Personalized Genetics/Genomics: Blogterview with Steven Murphy, MD May 20, 2007

I’ve recently decided to deepen my knowledge on the field of personalized genetics/genomics as it has an exceptional future in the realm of medicine (and business). And who is the right person to answer my geek questions? Of course, Steven Murphy, MD, the blogger of the Gene Sherpa. He is the Clinical Genetics Fellow at Yale University and is also the founder of a Personalized Medicine practice.

• We’ve heard a lot about personalized medicine, but please tell us more about personalized genetics.

Sure. There are some fundamental differences here. When I think about personalized genetics (Which is different than personalized genomics) I think about modifier genes involved in single gene disorders such as Cystic Fibrosis. A few months ago, I diagnosed a 70 year old woman with CF. She had been treated as if she had emphysema, had never smoked, and no Alpha 1 Antitrypsin deficiency. Something didn’t sit right with me. We had her get sweat tested and sequenced. Guess what? Compound heterozygote with one Delta 508 mutation. How could this happen? Modifier genes. There is a nice review of modifier genes in CF several months back in the New England Journal of Medicine. That is personalized genetics in my opinion…

Should we treat you aggresively or not? This woman clearly did ok without Creon (pancreatic enzymes), aggressve pulmonary toilet, or inhaled antibiotics. Now with the newborn screen we detect so many more patients with Cystic Fibrosis. Who should we treat? How should we treat? Personalized Genetics is like personalized medicine for those with single gene disorders. I remind you that “No gene is an Island” so we need to take it in context of the rest.

• Personalized genetics or genomics? Is there any important difference?

Personalized Genomics is a totaly different animal. Here we deal with what everyone affectionately calls the “Personal Genome”. This is the dream of everyone gets a genome sequenced at birth, we assess risk, create prevention plans, identify idosyncratic drug reactions prior to medication therapy. The fear is obvious…”When is GATACA coming?” I think that we need to put protections in place to prevent discrimination from more than just employers and insurers. What about that University you want to get in to? In addition there are several problems with whole genome screening aside from its multimillion dollar pricetag (which is dropping quickly). That is the problem in medicine known as the incidentaloma.

Quite often when ordering a CT scan, or chest xray, or what ever radiologic test we find tumors/cysts/masses in a completely asymptomatic patient. Does this mean we identify a cancer or other life threatening thing before it can cause damage? Sometimes, but more often than not we end up spending thousands of dollars evaluating something that turns out to be nothing. Just an incidental finding in an otherwise asymptomatic patient. An article entitled The incidentalome: a threat to genomic medicine.” was published in JAMA in July of 2006. Mathemeticians modeled sequencing the whole genome. As they get up to sequencing 10.000 people they find that the fraction of the population with a false positive result skyrockets up to 60%. What does this mean? Well, we have to carefully select who we test. Or better yet we need an immense database of “Normal Variants”. At a minimum we will need 1000s of “sequence specialists” or “computer sequence analysis programs” to evaluate and decide if the “work up” is indicated or not. Personal Genomics is very complex, even more than personalized medicine.

• What about the big companies focusing on personalized genomics/genetics?

We have key players including Illumina, 454 who has now been eaten by Roche, Affymetrix, Ventner, I could go on and on. The newest one to watch for is from the brainchild of 454 Jonathan Rothberg. He is launching a company called RainDance technologies. RainDance is already collaborating with Bayer Pharmaceuticals on high-throughput screening assays, noting the vastly superior statistics and reagent costs. What this means is a whole new means to sequence. If you add that into the nanopore sequencing mix at Harvard, then you have a robust field for development. I am sure I have missed a few, but these seem to be the key and future players to me.

As for personalized genetics, I know that the old stalwards like Genzyme, Genentech, BioMarin are all playing a role in defining the right infusion/pill for the right person with the right monogenic disease. Also you cannot forget about TGen who is building a presonalized medicine medical school in Arizona. My gosh I could just ramble about all of these things, but I will spare you all the details.

• As bioinformatics plays an important role in this rising field of medicine, how can web (especially web 2.0 ) help personalized genetics?

Web 2.0 and 3.0 can be best harnessed by networks of researchers sharing findings in open source forum. We need to give up the “Prize” for publication. Instead we need to nurture inter-institutional collaboration. In fact I would say we should prize how many universities were involved in every study. The web infrastructure can allow all of these things. In fact imagine coming up with a question in EST and shipping it to Mumbai for analysis during EST night-time. This is already happening in business. True, research takes more than overnight, but what if we were just talking about design. We could literally be working 24/7 to solve problems!

My dream is just that. The spirit of innovation, collaboration, and revolution all moving to solve the greatest code ever devised…DNA

I’m very thankful to Steven Murphy for the kind answers! Follow his blog for the most interesting news and explanations of personalized genetics.