We're as fascinated by genomics and "personalized medicine" as the next guy. Even Obama has the bug and touted the "Precision Medicine Initiative." Genetic testing even made it into pop science in 2007 when Google invested nearly $4M in 23andMe to introduce a DIY genetic testing kit to the masses.  It was all going pretty well until just after 23andMe did their D round of $50M and lowered their test kit price to just $99. Turns out that medical testing and associated health claims are regulated by the FDA. 23andMe is still around with one of the original three founders still on board. But their story now is about ancestry and learning more about your family - health-related reporting is no longer allowed.

Image via New Scientist.

AutoGenomics (pending: AGMX) makes it sound like there's a direct link between a genetic test and which drugs to administer when and in what combination. There's simply no such connection or any science available that proves this. In fact the leading genetic researchers acknowledge that genetics are clearly a potentially helpful input but must be combined with myriad other factors from environment and behavior. During the roadshow management notes "that many doctors are not even familiar with our genetic tests." The reason for that is that these tests are not clinically proven. These are doctors treating patients - not researchers or experimenters.

What AutoGenomics does have is a molecular diagnostics system that can be used by labs and clinics to run tests - sometimes in exchange for fees which can build an attractive business. There are indeed a few very specific genetic tests that the FDA does support like one to detect "low CYP 2C19" activity which can reduce the effectiveness of Plavix. And genetic testing on viruses is an effective way to "take the fingerprint" of the nasty thing and help determine the best way to kill it.

Besides loose clinical connections AutoGenomics faces stiff competition from dozens of other small research companies like deCODE genetics, Navigenics, 23andMe, Knome and Pathway Genomics. But the real 900lb elephants in this space is Illumina (ILMN). There are also some other very large and credible players like Roche (RHHBY) and Thermo Fischer (TMO). Both Roche and TMO have made acquisitions in this space to bolster their own internal research efforts. This market is likely to develop slowly and require large amounts of capital and staying power - it's not yet clinically or consumer driven.

Although AGMX generated $27M in revenue during 2014 with a 65% gross margin they have a spotty execution history including some competitive losses. Their business tends to be lumpy and still has customer concentration issues. They've had weak accounting and financial controls and some debt defaults in their past. We noted during the roadshow that the CFO appears to have been pulled off the external consulting bench because he has some prior experience in the banking industry. This isn't the type of market to go into with a B-team.

With a proposed price of $11-13/share and a little over 10M shares outstanding post IPO that puts the market cap of the stock a little over $120M at the mid-point and equals about 5x TTM sales. Probably too expensive for a company at this stage of development, with this management team, in this market. The company tried the IPO before in 2013 and had to pull the deal then. Even the President's "initiative" probably won't be enough to change the outcome now.

The IPO is scheduled for Feb. 11, 2015.