Major Sunday NY Times article on PEB:
July 23, 2000
The Microsoft (and Gates) of the
Genome Industry
By ANDREW POLLACK
OSTER CITY, Calif. --
When President Clinton
announced last month at a
White House ceremony that
scientists had effectively
mapped the human genetic
code, the man who was
arguably most responsible for
the achievement was not in
the room. The man, Michael
W. Hunkapiller, was home
with chicken pox.
He didn't mind too much,
though. He has never sought
the limelight.
Yet it was his company, PE
Biosystems, that designed the
high-speed DNA sequencers
used to unravel the human
genome. And it was Dr.
Hunkapiller, along with his
boss, Tony L. White of the PE
Corporation, the parent
company, who thought to set
up Celera Genomics and
have it use 300 of the new
machines to sequence the
genome years ahead of the
publicly financed Human
Genome Project.
The public project responded
by stepping up its own pace
and buying about 200 of
PE's machines. Dr.
Hunkapiller, in short, had
set off an arms race in which
he was the arms merchant.
And while Celera and its
outspoken president, J. Craig Venter, got the attention
and a spot on the dais with President Clinton, PE
Biosystems and the quiet Dr. Hunkapiller made the
money.
Much as Cisco Systems did with the Internet, PE
Biosystems -- which is changing its name back to
Applied Biosystems -- has emerged as by far the leading
supplier of equipment to the exploding genomics
industry.
The company -- like Celera, a subsidiary of the PE
Corporation with its own tracking stock -- controls at
least two-thirds of the market for gene sequencers. It is
also the leading supplier of equipment used for one of
the fundamental processes in biotechnology, the
amplification and detection of DNA by polymerase chain
reaction, or P.C.R. And it is in the early lead for
machines used to analyze proteins, a field known as
proteomics. That field is expected to be the next big thing
after genomics, because it will help researchers better
understand what all those genes, which guide the
body's production of proteins, actually do.
"There used to be a saying in the computer business
that no one ever got fired for buying I.B.M.," said
Lawrence S. Schmid, president of Strategic Directions
International, a market research company in Los
Angeles that tracks the analytical instrument industry.
"The same is true with PE."
ndeed, customers and competitors often refer to PE as
the Microsoft of genomics equipment, not only
because of its commanding position but also for what
they see as its aggressive tactics. And if PE is the
Microsoft, then Dr. Hunkapiller, whose doctorate is in
chemical biology from the California Institute of
Technology, is the Bill Gates, a man most comfortable
discussing the intricacies of technology but able to
transform himself into a respected, shrewd and tough
businessman.
Customers say, only half-jokingly, that when the
company operated as Applied Biosystems Inc., the
initials A.B.I. stamped on its machines stood for
"arrogance beyond imagination." Rivals say the
company uses its powerful patent portfolio and frequent
lawsuits as cudgels to stave off competition or to force
others to license technology it needs.
And some critics say that until it introduced its powerful
Model 3700 sequencers two years ago, the company,
facing little competition, had kept its prices high and
improved its machines only slowly.
"I think the genome project could have gone ahead faster
if they had made the capabilities of the system available
sooner," said Clark Tibbetts, a scientist at Virginia
Tech. Several years ago, while at Vanderbilt University,
Dr. Tibbetts took matters into his own hands by
modifying a sequencer to quadruple its speed.
Any company being compared to Microsoft has to be
aware of antitrust concerns.
PE recently dropped plans to acquire a smaller
company, Third Wave Technologies, saying the Justice
Department's antitrust review was taking too long. In
1998, PE had to divest some patent rights to clear
another acquisition.
The Justice Department would not comment. And after
PE sued two of its largest competitors, contending patent
infringement, those rivals countersued, contending
antitrust violations.
The Department of Health and Human Services,
meanwhile, is investigating whether federal funds were
used in the invention of the first automated DNA
sequencer at Caltech in the 1980's.
If federal funds were used, PE might have to give the
government a discount on gene sequencers or share the
patents more widely. Federal officials are also looking at
whether someone other than those listed on the patent
might have invented the machine. If so, the patent could
be invalidated.
Dr. Hunkapiller, 51, one of the inventors at Caltech, said
the invention issues had already been examined by
Caltech and by various patent offices. He said the
Justice Department had told his company that it was
not under investigation. And both he and Caltech said
that while the university received some federal
financing, it was not until the sequencer had already
been invented.
In any case, he said, PE already gives the government a
discount and licenses its technology to others.
"Despite that, we have very strong market shares," he
said. "We don't apologize for that."
He said the federal investigations had been stirred up by
people who resented Celera's competing with the public
genome project or by companies sued by PE. "There are
people out there who don't like paying licenses for
technology they do not own," he said.
ecurities analysts dismiss the Caltech issue.
Virtually all those who follow PE recommend its
shares, agreeing with company projections that it will be
able to maintain the 20 percent annual revenue growth
it has recorded since 1993. Profits are also expected to
grow at that pace.
"PE Bio is the best of breed," said Winton Gibbons, an
analyst at William Blair & Company in Chicago. The
company's stock is up 176 percent over the last 12
months, though it is down by almost half from its peak
in March, to $81.
Still, the going could get tougher, now that the mapping
of the genome is nearly finished. PE sold 1,100 new
Model 3700 sequencers at $300,000 apiece in a single
year, adding about $300 million to revenues. Now the
arms race seems likely to subside.
"The real game is not more sequencing," said Michael
McKenna, vice president for operations at the CuraGen
Corporation, a genomics-based drug development
company. "The real game is seeing which genes are
drug targets. It's a completely different set of tools."
PE officials say sequencers will still be needed for the
genomes of other organisms, for DNA fingerprinting in
forensics and for finding differences among human
genes that can help diagnose diseases. Moreover, each
Model 3700 automated sequencer uses as much as
$100,000 of chemicals a year, providing a continuing
income stream. While PE will not be able to keep selling
1,000 high-end machines a year, it expects its overall
sequencing business to keep growing.
Analysts say sequencers account for 40 to 45 percent of
PE's revenue, which is expected to total about $1.4
billion for the fiscal year that ended in June. Parts of the
polymerase chain reaction business are growing even
faster, said Eric Schmidt, an analyst with SG Cowen.
And Celera, which raised $1 billion in a March stock
offering, plans to spend a big chunk of it on proteomics
equipment being developed by PE Biosystems.
Still, PE is not likely to be as dominant in proteomics.
And technology is evolving rapidly: start-up companies
have developed chips as small as postage stamps that
can analyze thousands of genes at a time or perform
complex chemical analyses. Such chips could
undermine PE's bigger instruments, just as personal
computers undermined I.B.M.'s mainframes and the
Internet threatens Microsoft's dominance.
So PE, which is behind in developing such chips, has
entered into more than 20 alliances with companies
ranging from 3M to start-ups like Illumina and Aclara
BioSciences.
"It's critical for our growth that the start-ups, the new
ideas, come to us first," said Michael Albin, vice
president for science and technology.
Indeed, PE has long been adept at licensing technology
from others. Even the powerful new sequencers used in
the genome project were developed in good measure by
Hitachi. PE's sales force and financial strength make
the company an attractive partner, and it excels in the
combination of chemistry, biology, engineering and
software needed to turn raw ideas into usable products.
Applied Biosystems was founded by venture capitalists
in 1981 to commercialize technology from the laboratory
of Leroy Hood, then a Caltech professor and a pioneer in
the automation of genomics. Two Hewlett-Packard
executives, Sam H. Eletr and Andre Marion, were
recruited to run the company, and they in turn recruited
Dr. Hunkapiller, Dr. Hood's right-hand man. Starting
in research and development, Dr. Hunkapiller rose to
head of the company in 1995. "Mike has really given the
vision to the company," said Mr. Marion, who retired in
1995.
Dr. Hunkapiller is a man of few words, rarely shooting
from the hip or showing his emotions. But while
outsiders can be put off by his bluntness, his engineers
seem to idolize him for his lack of airs and the creative
atmosphere he has fostered at the company's sprawling
campus here, next to San Francisco Bay.
"Mike is in a cubicle at the end of the hall," Dr. Albin
said. "If you want to walk up to Mike, you walk up to
Mike."
Despite his terseness, Dr. Hunkapiller was a star
debater in his high school in Seminole, Okla., having
been coached by another debater from the school, David
L. Boren, who later became a United States senator and
governor. And that skill shows up in negotiations.
"If he believes in something, he'll argue it, argue it," said
John H. Richards, a close friend who was Dr.
Hunkapiller's thesis adviser at Caltech.
"He doesn't give up."
Not that Dr. Hunkapiller is all work and no play. In fact,
he plays soccer at lunchtime nearly every day. He is a
"Star Trek" fanatic who owns videos of every television
episode and movie. And he is a fervent Dallas Cowboys
fan who was thrilled when his company arranged to
have the former quarterback Roger Staubach, whom he
had never met, deliver a videotaped greeting for his 50th
birthday.
After graduating from Oklahoma Baptist University, Dr.
Hunkapiller went on to get his doctorate at Caltech and
then moved to Dr. Hood's lab.
It was there that Dr. Hunkapiller and others, including
his younger brother, Tim, made the key invention of
attaching different colors of fluorescent dyes to the four
different chemical units in the genetic code, allowing the
sequence to be read by machine.
From the start, Applied Biosystems outran or outlasted
bigger competitors like DuPont, EG&G and Beckman
Coulter.
The company has been helped by its control of the key
Caltech patents. Normally, a company would not be
obligated to license patents to anyone. But the
agreement with Caltech requires PE to license the DNA
sequencer patents to others on reasonable terms. Dr.
Hunkapiller said the company has abided by this but
added, "reasonable, from our perspective, doesn't mean
free."
Rivals say PE drives a hard bargain and plays its
patent hand brilliantly. Hitachi, the Japanese
electronics giant, could not come to terms on a license.
"There were many attempts to license it and
cross-license it throughout the years," said Mark
McDonald, president of Hitachi Instruments.
But by 1997 or so, a huge threat was developing. A
Silicon Valley start-up called Molecular Dynamics was
developing a high-capacity sequencer called the
MegaBace that could run rings around anything PE
offered at the time. Molecular Dynamics found PE's
asking price for a patent license prohibitive. But
Molecular linked up with -- and was eventually
acquired by -- a company called Amersham Pharmacia
Biotech that had wrestled a patent license from PE in
exchange for giving PE rights to a key enzyme used in
the sequencers.
Faced with the MegaBace threat, PE turned to Hitachi
for the technology to build its own high-capacity
machine, the 3700.
But the MegaBace beat the Model 3700 to market by
months. Amersham's first big customer was Incyte
Pharmaceuticals, a genomics company that had grown
impatient with what it saw as PE's slowness in
introducing new technology. Some customers say the
MegaBace, which cost less, is more flexible and reliable
than the 3700.
"We have struggled extensively with the reliability" of the
3700, said Elaine R. Mardis, assistant director of the
genome sequencing center at Washington University in
St. Louis, which has machines from both makers. Dr.
Mardis said she had to fly to California with records of
machine's downtime to get PE to provide more service
technicians. "It's not 'The customer is always right,' "
she said of PE's attitude. "The customer is always
wrong."
Susan Eddins, product manager for the Model 3700,
said early service problems arose because the company
had been overwhelmed by orders. "When this thing first
came out, it was really insane," she said. "We took our
service guys to the limit." Service, she said, has
improved since then.
Ms. Eddins also said the company's reputation for
arrogance was unjustified and arose because customers
expected customized machines, which PE could not
reasonably provide.
"They don't understand that we put out 5,000 boxes a
week," she said.
espite Amersham's lead in getting the MegaBace to
market, PE's 3700 could work with less pure
samples of DNA and was more automated, able to run
all day and night without human intervention.
It thus became the machine of choice for the
24-hour-a-day DNA crunching at Washington
University and at three of the other four centers that did
most of the public Human Genome Project, even though
many of its scientists would have preferred not to buy
from the company that financed Celera.
Still, Amersham Pharmacia, a joint venture of Nycomed
Amersham of Britain and Pharmacia, the drug
company, has provided the most significant competition
to PE in years. It sold 288 MegaBace machines last
year, giving it about 20 percent of the high-end market
-- more if Celera's 300 machines were excluded from
PE's sales of 1,100 machines. One of the five centers of
the Human Genome Project, as well as companies like
CuraGen and Incyte, now called Incyte Genomics, uses
Amersham's machines exclusively.
Amersham and PE are locked in a nasty patent lawsuit
over technology, although in this case it was prompted
by Amersham withholding a license from PE.
PE is also in a battle with MJ Research of Waltham,
Mass., the leading competitor in equipment used to
prepare DNA for the sequencing machines, which is
often done by using the polymerase process. PE wants
all makers of machines to buy a P.C.R. license. But MJ
has refused, saying its machines can be used for other
processes. It is countersuing on antitrust grounds.
Just this month, PE filed a patent infringement suit
against Affymetrix, the leader in gene chips.
While Dr. Hunkapiller has been with Applied
Biosystems nearly from the start, the company's
transformation to a powerhouse was completed by Mr.
White, the head of the parent company.
In 1993, Applied Biosystems was acquired by
Perkin-Elmer, the old-line scientific instrument
company in Norwalk, Conn., that had rights to the
polymerase technology. Mr. White took over as chief
executive at Perkin-Elmer in 1995 and focused the
company on life sciences. Last year, he sold off the
mainstay but stagnant analytical chemistry business to
EG&G, which then renamed itself PerkinElmer Inc.
Mr. White's company, meanwhile, became the PE
Corporation, with no stock of its own but two divisions
with tracking stocks -- PE Biosystems and Celera. Now,
because of confusion with PerkinElmer, PE Biosystems
is changing its name back to Applied Biosystems,
pending shareholder approval this fall.
The PE Corporation is contemplating a name change,
too.
Mr. White said that one reason for setting up Celera,
which sells genomic information, was concern that the
instrument business might be "marginalized" in the
long run, with the real money being made in using the
instruments to study genes or to make drugs.
But the move is risky in that PE Biosystems, through
Celera, is now competing with its own customers.
Company officials say Celera was not favored in
shipments or service.
Still, Mr. White talks about Celera, which is based in
Rockville, Md., having "first mover advantage," because
it gets to look at the machines while they are still in
development. PE expects to benefit as well, using genetic
information from Celera to design genetic test kits.
So far, though, rather than hurting sales to other
customers, the formation of Celera has spurred them. "It
woke up the industry," Mr. White said.
Dr. Hunkapiller said he did not necessarily think that
Celera's formation would provoke the public genome
project into buying so many PE machines. "We certainly
didn't get into it with that intention," he said. "It just
happened."
ome people at PE Biosystems say the company in
general, and Dr. Hunkapiller in particular, should
have received more recognition at the White House
ceremony. Dr. Hunkapiller said he felt bad for his
employees but not for himself.
The positive side to his being ill was that his daughter,
Kathryn, a biologist who works for Celera, got to go in
his stead.
"He just doesn't care," his brother Tim said. "Maybe
deep down there's some little twinge. But he doesn't
express it." |
