Genomics vs. Proteonomics: Accessorizing Your Genes

I had the occasion this past weekend to be out with my wife doing
some shopping. Apparently, I have
been too busy of late to notice that
my wardrobe had been in some decline. My wife therefore drug me out on
grey Saturday (which follows Black Friday) to hit the local Nordstrom Rack.  I was shamed into trying on jeans formerly priced at over $200 (who pays this kind of money for a pair of jeans?), gigolo bling-bling shoes (are pointy toes really appropriate for male shoes?) , and a variety of belts and watches (how does wearing a watch “change” my outfit) required to properly “accessorize” my look. We ultimately settled out on some funky 7 Diamonds and Roar surfer shirts to match the now half priced jeans. More on the shoes later.

The experience of “accessorizing” reminded me of a recent post by Matthew Holt regarding Personal Genome Management.  Matthew reviews some of the recent buzz surrounding 23andMe, highlights longtime player DNA Direct, and then puts out some thoughts as to where the market is and can go over the next several years (there is some interesting banter within the comments section as well). He concludes with this consideration:

“The genetic test market is very small, and the
management services that these companies offer around it are going to
only be a share of the testing market itself. So the fact that Navigenics
has already raised money at a substantial valuation means that some
very astute people are thinking that genetic testing will turn from an occasional activity for a small minority of patients (usually those going into pregnancy with some type of risk factor) into
a consumer norm that most patients will have as part of the standard
testing they get done and that management of that genetic information
will be part of the new flow of consumer and clinician activity.”

Matthew hints at something that I believe most people have failed to
grasp when considering the genetic market. I actually see three
distinct limitations:

1. One and Done. Genetic profiling is
a one time event - a single snapshot that once completed really does
not need to be repeated. There is no repeat customers, no transactional
nature, and no subscription model to the DNA testing market. So while I
share the optimism in the potential market, I also recognize that it is
transactionally bound by the singular nature of a single sales event.

2. Genotype vs. Phenotype. In addition, once the
blueprint has been identified, then the analytic software goes to work
to attempt to divine the ACGT combinations in to a series of
probabilities and risk for acquisition of all types of disease states.
As anyone familiar with cellular biology knows, having the genetic
sequence of a known gene (genotype) does not equate to having the disease state (phenotype)
represented by that gene. It requires specific cellular triggers, and
specialized cellular machinary, to literally “translate” the code into
the work horse of the cellular world - proteins.

3. Cellular Accessories. Furthermore, not only do
we have still have a very limited understanding of human genes, there
are very few pathological conditions represented by single genes. Most
diseases are a confounding confluence of cascading brownian
events beyond our current scientific understanding. The inherent
complexity, and all the cellular accessory events that occur to create
these microscopic “perfect storms“, should dispel any illusion that health and disease fits neatly into discrete genetic units.

As a result, while genomics provides a methodology to understand the sheet music of life, I am much more interested in the proteomic symphony
that results by literally bringing the dead notes to life. Since the
actual performance of the sheet music can vary based on the whims of
the composer are the individual variation or free lancing of the
orchestral performers. As a consequence of these unique variations, I
am murch more interested in the potential and the promise of emerging
predictive proteomic companies like Biophysical than I am of a Kleiner Perkins backed genomic startup like Navigenic (or a Google-powered 23andMe).

Biophysical is an Austin-based spinoff that commercializes the biomarkers discovered by sister company Rules-Based Medicine.
RBM is a biomarker testing laboratory comapany that has developed
multiplexed immunoassay capabilities. This is seen in their
Multi-Analyte Profiles (MAP’s) that can measure up to hundreds of
proteins in a very small samples and sample types. The novel
utilization of MAPs provides for comprehensive and cost-effective
evaluation of the protein expression patterns critical for
understanding drug safety and efficacy, disease diagnosis, and
predictive disease modeling.

Once RBM has discovered and tested a new biomarker for clinical or
disease relevance, it is commercialized by Biophysical into a
predictive disease diagnosis/modeling tool. While the traditional Chem 20
group of analytes has provided workmanlike information, the Biophysical
250 provides exquisite detail on the biopresence of bioactive
biomarkers. In essensce, Biophysical has the capacity to detect active
pathology at the protein level long before enough cells have aggregated
to be detected by traditional diagnostic testing methods. The power of
the Biophysical model is to literally be in the accoustic chamber
actively listening to the symphonic soliloquoy.

From a business perspective, Biophysical is also a subscription
model. Essentially, they are selling “ongoing” biosurveillance. The
value of their service increases over time as more tests are run, as
baselines levels are followed over time, and as an aggregate picture of
individual biohealth emerges. This concept has recently been picked up
by Dr. Mehmet Oz, the  current leading guru of longevity and real-age health. Biophysical recently launched a modified version of the Biophysical 250 (called Biophysical YOU) with Dr. Oz on the internationally syndicated Oprah Winfrey show. The response, as you may have predicted, has been overwhelming.

So while the sheet music is absolutely essential to even contemplate
a symphonic performance, no one goes to listen to the sheet music. It
is the individual performances, the living-breathing-
audible-panoramic-splendid-view of the whole chamber coming to life
that ultimately stirs the soul. In an analogous fashion, the proteins
are the cellular accessories that literally add life to an other wise
staid genetic makeup. Infinitely more variable, infinitely more
possibilities, and as a result, infinitely more interesting.

And as far as fashion accessories, those pointy Steve Madden’s actually look pretty dang good with those funky designer jeans.

Scott Shreeve

* DISCLAIMER: I have been a consultant to Biophysical since early November 2007

2 Responses to Genomics vs. Proteonomics: Accessorizing Your Genes

  1. jd says:

    While I agree that the phenotype/genotype distinction shows a limitation of the gene identification business, I have three comments on the post above.
    First, as things stand I believe very few people are getting their entire genome sequenced, and so it isn't "one and done" for a large majority. Most people are getting a select number of genotypes read, and as science progresses they may well have to go back for more information later.
    Second, 23andMe and other companies seem to get that if they want a continual revenue stream from their customers, they are going to have to sell services. I didn't look into it, but I wouldn't be surprised if they charge a yearly subscription fee to find out how the latest advances in science apply to their genotypes.
    Third, and this is more of a philosophical/methodological point, one of the biggest limitations to genotype analysis wasn't mentioned: the impact of environment. Certainly the environment explains how phenotypes differ from genotypes, and why essentially no genotypes have a 1-1 correspondence with diseases or other complex human outcomes. Moreover, I don't think you can abstract away from non-biological environmental conditions (including social conditions) and without losing critical parts of the causal story on how diseases persist, favor some groups over others, etc. Reductionism has its limits.

  2. Gregory D. Pawelski says:

    The people wowed by genomics perhaps do not see the complexities and redundancies of human biology are beyond the ken of genomics. In cancer medicine, getting better drug development tools from advances in genomics and proteomics is one thing, but harnessing those tools for drug testing and for delivery to patients is another matter entirely.
    While genomics focuses on how genes direct activities within the cell, proteomics studies how proteins carry out those directions. Genes create the blueprints for the production of proteins within the cell. A protein is a molecule that makes a cell behave in a certain way. It does so by interacting with other proteins in a complex series of steps.
    In chemotherapy selection, gene and protein testing examine a single process within the cell or a relatively small number of processes. The aim is to tell if there is a theoretical predisposition to drug response. Whole cell "functional profiling" tests not only for the presence of genes and proteins but also for their functionality, for their interaction with other genes, proteins and processes occurring within the cell, and for their response to anti-cancer drugs.
    Genomics is only important insofar as it influences proteonomics, which is only important insofar as it influences protein function analysis (are proteins active or inactive), which is only important insofar as it influences cell function analysis (cell culture testing). There is an inverse hierachy between relevance and ease of measurement. So genomics and proteonomics are not the only potential key to genetic disease.
    Cancer is a complex disease and needs to be attacked on many fronts. The best thing to do is to combine these different tests in ways which make the most sense.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Health 2.0