By John S. MacNeil
Imagine, if you will, running a biotech company that provides research services to big pharma. Sounds easy enough. Now imagine that you’re making a profit. A little bit harder to comprehend, but all right. But what if you couldn’t use that profit to reinvest in your business, much less reward yourself or your employees with higher salaries? Instead, what if you had to hand over your hard-earned rewards to a higher power that demanded cheap and impeccable research services, yet refused to accept the possibility that you might not — at the very least — break even every year?
Welcome to life as a core research facility director.
In the era of big biology, precipitated by the sequencing of the genome, the role of core facilities in academia has grown steadily in significance. More and more researchers, faced with the prospect of devoting hundreds of thousands of dollars from their own grants to purchasing high-end instrumentation for DNA sequencing, gene expression profiling, or protein identification and analysis, are turning to their universities’ core facilities to provide the data they need to help them understand their chosen biological problem. At the same time, core facility directors, under pressure from their own budget constraints, are struggling to keep up with their clients’ demands to acquire ever more sophisticated tools for biological analysis. Compounding this difficulty, core facility directors must maintain staffs of highly trained laboratory technicians, strive constantly to keep overhead low, and fight for the professional recognition and personal respect they feel they deserve. In short, core facility directors are in a pickle.
“We get it from both ends; we get it from the limitations of the technology, we get it from the people who come in here from different backgrounds and think we’re actually running a company making a profit, and therefore we’re short-order cooks,” says Alan Smith, director of the Protein and Nucleic Acid Facility at Stanford University. “You get a lot of the crap, and you get very little of the credit.”
But that’s not to imply the job is impossible. In fact, GT interviewed more than a dozen core research directors and asked them what it takes to run their labs successfully. That is, how to provide high-quality results at a price both reasonable to the investigators and that covers the costs of running the facility not picked up by the university’s tab. Although their responses vary as widely as the needs of researchers at their institutions, a few common words of wisdom emerged: widen your user base, be creative in financing new technology acquisitions, and develop communication skills to deal with headstrong clients.
First off, Guarantee Yourself Revenue
Although every core facility director will claim that good science always takes precedence over covering costs, there is still undeniable pressure to generate revenue. After all, from a university administrator’s perspective, any core facility that manages to wean itself off institutional subsidies — while maintaining superior data quality — is a significant asset. To this end, core directors have devised a number of strategies for bringing in new business.
Acquiring new technology is always a gamble, core directors say, but with forethought and a bit of market analysis, it can bring in a significant number of new users, says Ted Thannhauser, director of the BioResource Center at Cornell University.
While he most recently passed on the opportunity to integrate DNA microarray analysis into his facility — the technology went to another core facility at Cornell with more financial support — he says his gamble in the early ’90s to invest in DNA sequencing technology was nothing short of a godsend. At the time, the faculty was opposed by a ratio of eight or nine to one to acquiring a sequencer, because of the early technology’s limitations. Thannhauser purchased an instrument anyway.
“That was the easiest facility to launch in the world,” he says of the DNA analysis core. “All we had to do was get the first user” to communicate to the rest of the faculty that the service was faster, better, and cheaper than sequencing by hand. Today, Thannhauser’s DNA core serves more than 600 investigators, encompassing more than 2,000 individual users.
Building a client base is also a matter of sizing up one’s competition. Stanford’s Smith sees for-profit research service providers as competing for his clients’ business, and has developed a strategy for outwitting his private-sector rivals. Commercial services, such as Lark Technologies, may compete with Smith’s DNA sequencing lab on price per read, but he figures his advantage may lie in speed and convenience, as well as the ability to provide specialty services and educational and technical support.
Alternatively, some core directors see revenue generation in a broader sense. In addition to providing a service to the university research community, they say, a good core facility can collaborate with investigators on grant applications, helping the researcher attract funding by offering the use of the facility in exchange for a contribution from the grant toward the cost of running the lab. Mark Duncan, director of the proteomics facility at the University of Colorado Health Sciences Center in Denver, says that often when an investigator approaches him about collaborating, he’ll suggest generating some initial data for use in writing a grant. This scenario gives the investigator a leg up in applying for funding, he says, and Duncan’s lab avoids the difficult task of determining a preset fee to charge investigators for proteomics analyses that often become far more complicated than they appear initially.
To Cut Costs, Automate and Build In-house Expertise
On the other side of the coin, core facility directors are also under pressure to keep overhead low and make gains in productivity. Most core facility directors have PhD-level training in the sciences, but none in business skills, making the transition to running a cost-conscious core facility a challenge, to say the least. “It’s a set of skills that aren’t taught in any graduate program, because the business aspects you don’t get in your graduate program training, such as calculating depreciation,” says Thannhauser. Those who have learned on the job offer some advice on how to make a core facility more efficient.
First and foremost, several facility directors say, implement a sample tracking system. Smith wishes there were an off-the-shelf program flexible enough for every core facility to adopt. “Is there a need for a universal, inexpensive software for core facilities? Let me know when you start the company and I’ll be one of the investors,” he says. However, because each core facility’s requirements differ, the necessary LIMS will most likely need to be customized, he adds. At Stanford, Smith’s group invested more than $15,000 in a home-built system to track samples based on FileMaker Pro, and is still working to align the software with the university’s Oracle-based billing system.
“The biggest single difficulty is, in fact, we’re expected to run a business, but we don’t have a bookkeeping and billing component to support us like any decent company would,” Smith says.
Other core facility directors are in the same situation. At Harvard University, Bill Lane has had an integrated custom database in place since the early ’90s for storing notes from every discussion associated with a sample analyzed at his Microchemistry and Proteomics Analysis Facility. And at Vanderbilt University, Shawn Levy has implemented software for tracking samples and billing users of his DNA microarray facility.
To save money on instrument service contracts, Duncan, at the University of Colorado Health Sciences Center, employs two full-time technicians who run samples and repair his 10 mass spectrometers when they break down. In addition to an Amersham Bioscience Ettan MALDI-TOF, Duncan’s stable includes an ABI QSTAR, two ABI MALDI-TOFs, and two Thermo Finnigan ion traps. Duncan says his technicians fix about 80 percent of the problems he encounters with his instruments, which saves his lab $5,000 to $10,000 each time he does not have to call the vendor for a service visit. “We’ve invested in expertise,” he says. “Should the instruments break down, they are familiar enough with the intimate aspects of the insides to pull them apart and put them back together. That can be a substantial savings.”
Lane at Harvard has another tip for getting the most from the expertise available at an academic institution. Because software for analyzing data and managing samples is critically important to the smooth operation of a core facility, “you must have a connection to somebody who can design customized databases and data analysis software,” he says. Lane’s solution: “Tap into the undergraduates who can program.” Student programmers don’t have to be biologists, he adds, as long as they can write algorithms for handling large data sets and visualizing results. “It’s a huge help.”
How to Stretch a Nonexistent Budget for New Technology
Keeping a core facility outfitted with state-of-the-art technology presents another challenge. Facilities that serve government-supported researchers aren’t supposed to make a profit, but if they do, federal rules prohibit that money from going toward the outright purchase of new instrumentation, according to Stanford. Instead, many facility directors rely on grants to fund the purchase of new instrumentation, although this money may not cover all of a facility’s needs. Often directors resort to more creative funding mechanisms to balance their users’ demand for access to cutting-edge technology with their equally vocal demand for inexpensive service.
Beta testing new equipment may offer one solution. Although many facility directors downplay its significance as a method of acquiring new instrumentation on the cheap, Colorado’s Duncan has found that his relationships with Amersham Biosciences and other instrument suppliers have helped his lab stay current with technology.
As a reference lab for Amersham, Duncan’s facility acquired the company’s Ettan MALDI-TOF mass spectrometer (he won’t say for how much). In return, Duncan contributes feedback on the instrument’s performance and demonstrates its capabilities to new users and potential buyers. Duncan says the relationship has also paid off in terms of technical support from Amersham, and he finds value in his ability to shape the company’s product development efforts. “We don’t indiscriminately take on companies,” he says, “but by being innovative and having a good working relationship with the instrument companies, it’s possible to do more with your money.”
Cornell’s Thannhauser has another suggestion: work with vendors to acquire new technology on a “rent-to-own” basis. This strategy, he says, allows his lab to gauge investigators’ initial interest in a particular instrument, thereby minimizing the risk of investing heavily in a technology that few researchers will pay to access. In Thannhauser’s experience, a rent-to-own arrangement makes economic sense even if the lab cannot convince the vendor to count the cost of leasing instruments toward the final purchase price, because the opportunity to evaluate an instrument without commitment is worth paying for. “A lot of times companies won’t do it,” he says, “but if a company needs to place instruments, it works.”
At Cornell, Thannhauser successfully negotiated rent-to-own agreements when acquiring his Thermo Finnigan ion trap mass spectrometers in the early ’90s, an early version of Biacore’s surface plasmon resonance technology, and his original Applied Biosystems 373 DNA sequencer.
AMDeC, a consortium that represents 39 medical schools across New York state, has found that negotiating technology acquisitions for all of its participating core facilities provides the leverage to wrangle lower prices from vendors. Andrew Brooks, who directs AMDeC’s microarray resource center at the University of Rochester, says his strategy is to first poll individual core facilties to determine their technology needs, and then negotiate with the technology vendor for a discounted price based on the number of participating facilities. “The advantage is only as large as whoever wants to participate, but we get a lot of participation in our programs, because everyone respects everyone else’s considerations,” says Brooks. “We’ve done our homework, and we’ve all collectively gotten together and chosen the best product or the best piece of instrumentation for our application.” AMDeC has used this strategy to acquire the GeneTraffic server from Iobion, reagents and consumables from Clontech, Affymetrix, and PerkinElmer, and is about to announce a similar agreement with Ambion, Brooks says.
Other facility directors suggest taking innovative approaches to accounting for the cost of financing technology acquisitions. When he can’t fund the purchase of new instrumentation through NIH grants, Smith says he has resorted to borrowing from the Howard Hughes Medical Institute, which until this past fall contributed less than 15 percent of his core facility’s more than $1 million annual operating budget. Now, Smith says he’ll have to borrow from Stanford to pay for the upfront cost of acquiring new equipment, and slowly pay back the university at a fixed interest rate from his service revenue.
At Harvard, where Lane has run the microchemisry and proteomic facility since 1982, financial self-sufficiency allows him to acquire new technology without having to rely on federal grants. Adjusting his service fees to cover the cost of acquiring new capabilities, he says, enables him to make improvements to the lab’s instrumentation on average every year and a half. Ultimately, the bigger core facilities have the advantage when it comes to funding new technology, Smith says, because they are more often associated with medical schools and hospitals, which can access funds through program project grants, specialized centers, clinical trials, or even private institutional gifts. “In many ways, the larger the institution, the more ways you have for funding the purchase of new equipment,” he says.
Don’t Be a Pushover
But new technology is only as effective as those who know how to use it, and every core facility director stresses the importance of employing good people in the lab. “You can certainly do a lot more with a good person and a marginal machine than you can with the greatest machine and a marginal person,” says Thannhauser. “Getting good people is the key, and people who are good at running service facilities are few and far between — and unfortunately the skill set necessary to be a successful faculty member is almost orthogonal to the skill set necessary to run a service facility.” Meticulous laboratory skills are essential for core facility laboratory assistants, says Jeanette Papp, who directs the DNA sequencing core at the University of California, Los Angeles. But keeping them interested in routine analysis work requires “keeping them plugged into the underlying science,” adds Brooks.
And just as important, directors say, is having good people skills in general. Managing relationships with 50 to 100 different researchers requires adept interpersonal skills as well as the strength of character necessary to avoid being bullied by investigators banging the lab bench and demanding their results. “Medical schools have more egos than other institutions,” says Paul Tempst, who directs the protein analysis core at Memorial Sloan-Kettering Cancer Center. “I don’t think the university is well served if the [facility director] doesn’t speak up.”
Says Duncan, “Don’t think of us as a resource that’s sitting here — we’re not playing checkers and knitting, waiting for you to arrive to do your work. There are always many more projects on the table than there are people to work on those. So we put the burden back on the investigator to get the work done.”
Maintaining strong personal relationships with customers also tends to improve the quality of the data that emerge from the core facility, directors say. By briefing investigators on the capabilities of the core, directors can make sure their clients’ expectations are realistic, as well as learn something about the biological problem under study. “I talk to everyone,” says Lane. “Some people think of the facility as an auto shop where you bring your Mercedes. You can’t just drop off a sample. We just simply don’t allow that.”
Cores Go Regional
In the small world of core facilities, bigger often means better — and more clout for negotiating prices. Some people argue that regionalizing core facilities is one way to accomplish this.
Some groups, like New York’s AMDeC, contend that regional core facilities cash in on this benefit. The New York City-based nonprofit organization for biomedical research has established two core facilities, in bio-informatics and microarray analysis, that aim to serve its 39 member research institutions in the state.
In addition to negotiating favorable rates, Andrew Brooks, who directs the organization’s microarray initiative, says the effort to standardize technology will also allow researchers to more easily compare data between member institutions.
Ruedi Aebersold of Seattle’s Institute for Systems Biology is also hoping to create regional proteomics facilities. “What we’re talking about is cases where people really want to do proteomics experiments, where you might identify and quantify hundreds of thousands of proteins,” Aebersold says. “It requires a more involved infrastructure, including reasonably high-level computing, and that is hard to set up for” compared to core facilities that perform more run-of-the-mill proteomics analyses.
But the trend toward regional core facilities may not work for everyone. Some directors say their users appreciate their proximity and personal attention, which can lead to more fruitful scientific interactions.
AMDeC’s Brooks, however, remains committed to the idea that these meta-core facilities can provide superior data analysis at a lower price. “The advantage to building regional centers of excellence is that you can get newer technology to people faster,” he says. “And as long as the infrastructure [is in place], it can be done just as efficiently, without losing the personal touch.”