Breast cancer will strike more than 200,000 women in the United States this year, and 40,000 will die. X-ray mammography is the best way to detect early tumors, but the technique misses one in five cases, and women find the test uncomfortable. Susan Hagness and collaborators have invented a better breast-imaging technique. A woman lies on her back so that her breasts flatten naturally, and an instrument Hagness is developing scans the breast tissue with very-low-power microwaves, which are safer than x-rays. Hagness’s preliminary measurements on breast biopsy specimens indicate that microwave imaging makes malignant tumors stand out better than x-rays do. The energetic Hagness developed sophisticated computer algorithms—which process data collected by the imaging instrument—to enhance the detection and discrimination capabilities of microwave imaging. So far, her computational studies indicate that her approach should detect tumors just a couple of millimeters across, an improvement on the five-millimeter limit of x-ray mammography. The first version of Hagness’s instrument will be used for research.
Think of the human genome as Tolstoy’s War and Peace in the original Russian—immense, exciting, but for most, indecipherable. British bioinformatician Ewan Birney wants to make the genome’s information accessible to all. His Ensembl software and data allow researchers to find information on the Web on any known or predicted gene and to automatically match pieces of genes they have sequenced with other genes—without tediously combing through endless raw-sequence data. Cofounded by Birney, the Ensembl project has become one of the most popular resources for genome research, and its software is freely available to use and modify. With his related work adapting programming languages such as Perl and Java to biological projects, Birney has become a force in the bioinformatics open-source community. “In bioinformatics” he says, “the software is actually not that important. What’s much more important is the data.” Birney’s ambitious tools will help researchers deliver on the promise of new drugs and treatments derived from the Human Genome Project.
Stephen Boppart grew up in Illinois farm country, where he acquired a get-things-done attitude. His master’s-degree advisor says, “The speed with which he can conceptualize, test and implement is remarkable. ”While simultaneously completing a PhD in medical and electrical engineering at MIT and an MD at Harvard, he published 44 peer-reviewed papers and book chapters. During those seven years, Boppart helped dramatically improve the resolution of optical-coherence tomography, an imaging technique that sends near-infrared laser light into a person’s tissues and then interprets its reflection from structures within. Boppart also converted the hardware into a handheld probe that looks like a laser pointer. Surgeons at Brigham and Women’s Hospital in Boston are using it to see through a patient’s skin before making an incision. Recently Boppart received funding from the Whitaker Foundation, the National Cancer Institute and NASA to determine how to use optical- coherence tomography in cancer diagnosis. He is now developing contrast agents, such as carbon and melanin, that will increase a tumor’s resolution when seen using this technique.
Pathogens often exploit our cells to thrive. Fiona Brinkman therefore hypothesizes that some of their genes are similar to human genes. By identifying such genes computationally, Brinkman is trying to understand how drugs could stop pathogens from storming the body’s fortress. Brinkman coordinates this interdisciplinary work through an online “pathogenomics” project she runs from Simon Fraser University in British Columbia. It uses a free program she developed called PhyloBlast to evaluate relationships between genes by comparing their sequences and the proteins they code for. Previously, Brinkman organized the first Internet-based effort to refine and annotate bacterial genome data, focusing on P. aeruginosa, a widely drug- resistant bacterium that causes fatal infections in cystic-fibrosis patients. The group gained critical insight into how the bacterium works, which had eluded researchers. “Once we have the parts list for how the bug is functioning,” Brinkman says, “we can figure out new approaches to drugs.” Her collaborative electronic approach is now being used by other genome researchers.
Although he’s a doctor, Stephen Brossette thinks he can save more lives by using technology than by seeing patients. As a University of Alabama at Birmingham graduate student, Brossette developed a mathematical technique for finding subtle, otherwise unnoticed patterns in medical data. His approach can reveal an impending outbreak of a hospital- borne infection by identifying patients who share similar demographic backgrounds, frequent the same hospital rooms and have the same odd microorganisms in their lab cultures. In 2000,Brossette formed MedMined in Birmingham,AL,to market his innovations. The company currently has eight employees serving as many hospitals. They sift through masses of data provided by the hospitals to uncover patterns of nascent antibiotic resistance and outbreaks of infectious pneumonia, diarrhea and other diseases. Brossette says early detection can improve intervention—vital because two million patients acquire infections in U.S. hospitals each year, and 90,000 of them die. The amateur photographer and Cajun/Creole chef is also analyzing how best to apply his tools to bioterrorism and contaminated foods.
Chris Burge admits it’s been hard to choose a research focus. In high school he won math contests but in college majored in biology. He traveled to Nicaragua to see if medicine was his calling but wound up teaching people there about computers. He finally settled on the interface between biology and mathematics and returned for graduate study in math at Stanford University, his alma mater. The sophisticated computer program he developed there, called Genscan, predicts the locations of genes in the human genome and what proteins they produce. Released in 1997,Genscan is the most popular program of its type. Geneticists and molecular biologists are using it to identify human disease genes and potential drug targets and are applying it to agriculture. It is available free on the Web for nonprofit use and has been licensed to dozens of companies. Now at MIT’s biology department, Burge has developed a companion program, GenomeScan, that compares known proteins to increase the accuracy of Genscan predictions. He hopes to answer fundamental questions that could shed light on how human genes are expressed.
Chemical biologist Benjamin Cravatt is developing tools to illuminate the roles of proteins and enzymes in humans and animals. Cravatt and colleagues have synthesized dozens of fluorescent probes that chemically bind to enzymes in laboratory samples of healthy and diseased tissues, then light up when excited by a laser scanner. The technique can show which enzymes are more or less active in cancerous cells, which could herald a breakthrough for proteomics—the attempt to identify the structures and functions of human proteins. Cravatt’s protein-activity- based approach represents an advance over methods that merely infer protein function by comparing the abundance of proteins in samples. His technology also forms the basis of ActivX Biosciences in La Jolla, CA, which he cofounded in 2000 and now employs more than 40 people. Applications for the chemical probes include improving medical diagnostics, identifying new drug targets and facilitating drug tests. “Helping with the development of a single drug would be huge, ”the hyperkinetic researcher says, “but we hope to do this many times over.”
Leave it to a structural biologist who thought about becoming a pastry chef to write an industrial-scale recipe for accelerating drug discovery. In 1999,with a doctorate from the University of California, Berkeley, Nathaniel David cofounded Syrrx, the world’s first automated factory devoted to analyzing proteins and their interactions with drugs using structural biology. The three-dimensional shape of a protein determines how well a particular drug will bind to it, but the structures of many critical human proteins remain unexplored. It can take researchers in a traditional lab months to produce,purify and crystallize a single protein and confirm its shape. Under David, San Diego-based Syrrx adapted robot arms from auto- motive assembly lines to crystallize proteins with far greater speed: the company can reveal 11 to 15 structures a month. Some scientists doubted the feasibility of automating an intricate lab process but David—who claims his best quality as an innovator is stubbornness—prevailed. Assole employee for 16 months, he raised $25 million. Syrrx now has 131 employees, $100 million in capital and has analyzed 90 potential drug targets for three pharmaceutical makers.
Jennifer Elisseeff is shining light on better ways to repair human tissue. While getting her doctorate in medical engineering from the Harvard University-MIT Division of Health Sciences and Technology, Elisseeff designed a liquid polymer that can keep cartilage cells alive. In patients, the polymer hardens into a hydrogel—a scaffold on which the cells can develop into new tissue. Normally, surgeons have to cut open a patient to insert such a polymer, and shine light on it to induce it to harden. Elisseeff wondered if she could devise a polymer that hardened under minimal light. That way, surgeons could simply inject the compound and shine a light through the skin to trigger solidification, obviating the need for surgery. Her experiments with mice and rats succeeded. Now, Advanced Tissue Sciences of La Jolla, CA, is investigating the polymer as a way to repair everything from ruined knees to facial damage. Meanwhile, Elisseeff is impregnating hydrogels with stem cells—which can mature into different human cells—to try to create a new form of cartilage replacement. “So little is known about stem cells,” she says. “It’s very exciting.”
Derek Hansford’s unobtrusive bearing is just what you’d expect from someone who designs ways to sneak drugs past the immune system. Hansford has been fabricating tiny polymer particles that can hold drugs and be injected into a patient’s bloodstream. Once there, they could hunt down tumors and release their drugs, without affecting healthy cells. Along the way, the particles would shield the drugs from degrading enzymes and would not elicit attacks from the immune system—a common problem for cancer drugs—because they do not attract immune cells. Although other bioengineers are making polymer drug-delivery devices, none has made large numbers of uniform particles small enough to travel in the blood-stream; each of Hansford’s particles is about the size of a red blood cell. The scientist has adapted a technique called soft lithography to make the particles, casting hundreds of millions of them in varied shapes out of reusable molds. Startup company iMedd plans to license his technology. Hansford will now try to make particles for inhalable drugs—an alternative to injections.
Growing up in the shadow of Amgen in Thousand Oaks, CA, and then working in the company’s labs during college, John Harrington saw what it takes to succeed in biotech’s upper echelon. He kept that in mind when he founded Cleveland-based Athersys in 1994, then worked 18 hours a day to build it while still a postdoc. During that time Harrington coinvented the first man-made human chromosome; gene therapists are now investigating how to use such artificial chromosomes to penetrate cells and repair disease-causing genes. In his academic work, Harrington discovered fen1,a DNA-cutting enzyme that can accelerate the spread of cancers. Athersys is pursuing drugs that inhibit fen1 because they have the potential to treat cancer, alone or with chemotherapy. Athersys’s 130 employees are also commercializing Harrington’s most recent invention, a process called “random activation of gene expression,” which unveils the functions of proteins and could be an important tool in medical research and therapy. Athersys investors seem happy with Harrington’s creations and his ability to go all out: since 2000 they have chipped in $90 million.
Suzie Hwang Pun
Think of Suzie Hwang Pun as a traffic cop for genes. The chemical engineer uses polymers to carry injected genes through the bloodstream. With a system of molecular tags, she can direct a gene—say, one that blocks cancer progression—to just the right spot—like the nuclei of cells in a tumor. It’s a trick that could solve a huge problem in gene therapy research: a new gene does no good if it doesn’t make it to the right place. While viruses are the typical delivery vehicles in gene therapy, they’re hard to manufacture and can be intercepted by the immune system. Pun’s materials avoid those problems and open the possibility of delivering drugs, as well as genes, with exquisite precision. “This is the tip of the iceberg,” says Caltech chemical engineer Mark Davis. He was so excited by Pun’s accomplishments as a graduate student in his lab that he founded Insert Therapeutics in Pasadena, CA, primarily to commercialize her work. The clear-spoken Pun jumped at the chance to be a senior scientist and employee number one. If all goes well, her technology could enter human trials within a few years.
J. Joseph Kim
Viruses learned how to better infect people over millions of years of evolution; chemical engineer and MBA J. Joseph Kim is using their knowledge to fight other diseases. Kim figures touse the viruses’ strategies as the basis for new drugs for cancer and inflammatory illnesses. In 2000,after several years at Merck, Kim founded Viral Genomix in Philadelphia, for which he has raised $1 million. “Since I was in high school I wanted to start my own biotech company,” he says. His company may soon have its first drug: Kim has tinkered with a protein called vpr, which helps the HIV virus replicate, and has coaxed it to trigger cell death in more than 50 different cancer cell lines. So far, the protein has worked in laboratory cultures and in mice and macaques; Kim intends to begin testing it in human cancer patients within a year and a half. To that end, Viral Genomix has established a partnership with the University of Pennsylvania. Kim is also developing a drug based on vpr that can limit the proliferation of the immune system cells that cause rheumatoid arthritis and psoriasis.
Deadly genetic defects often involve single- nucleotide polymorphisms—single changes in the base pairs that make up DNA. As a graduate student at Johns Hopkins University, Steven Laken discovered that such a change occurs in six percent of Ashkenazi Jews and correlates with a 20 to 30 percent risk of colon cancer. With 20 million Ashkenazi Jews potentially at risk, Laken was not satisfied with simply finding the defect; he wanted to devise a rapid test for it. He created a lab procedure that separates DNA into fragments and then uses mass spectrometry to quickly search the fragments for the polymorphism. Doctors are now using the technique to screen patients with Ashkenazi backgrounds for colon cancer. After completing his graduate work, Laken joined Maynard, MA-based Exact Sciences, where he is now adapting the innovation for broader genetic tests, including one for nonpolyposis colon cancer, the most common inherited form of the disease. Laken believes his methods could spot virtually any illness with a genetic component, from asthma to heart disease.
Corinna E. Lathan
While involved in biomedical studies funded by NASA, Cori Lathan realized that astronauts in orbit encounter physical challenges much like those faced by people with disabilities. An astronaut, for example, must learn to move in an awkward space suit much the way a spinal-cord injury victim may have to relearn to walk. The experience guided Lathan in her search for better assistive tools as founder and CEO of College Park, MD- based AnthroTronix. An expert in human-performance engineering, Lathan devised interfaces that allow children to communicate with a half-meter-tall robot via body movements. Wireless sensors are placed on the child’s body, and Lathan’s playful, furry JesterBot solicits and mimics the movements and facial responses of its human buddy. The interaction can help a child with cerebral palsy get through painful physical therapy. Lathan is applying similar ideas to army research. Gestural interface technology can keep a night patrol leader in wordless contact with soldiers equipped with goggles that display his gestures as small icons. “I never thought about what I wanted to be,” she says.“I always just looked for cool things to do.”
Mad-cow disease occurs when an unruly protein called a prion causes healthy proteins in cattle brains to misfold. The same is true for the human versions of mad cow—“variant” Creutzfeldt-Jakob disease, which is contracted from beef, and the naturally occurring “sporadic” form. But until Kelvin Lee unleashed a new style of protein analysis, diagnosing these maladies required a postmortem brain biopsy—obviously, too late for patients. During postdoctoral work at Caltech in 1996,Lee identified a marker protein for sporadic Creutzfeldt-Jakob disease, yielding the first premortem test for the ailment. Lee went beyond the traditional method of studying a few proteins at a time; instead, he simultaneously analyzed the 2,000 proteins in human spinal fluid to pick out the telltale compound. In 1997 he confirmed that the disruptive protein also appears in mad-cow-afflicted cattle. People are now being tested for the protein in the U.S. and Europe. No one has confirmed whether the same marker characterizes variant Creutzfeldt-Jakob disease, but Lee’s team recently identified other protein indicators that may prove fruitful. Lee is also working on a similar test for Alzheimer’s.
If laboratories are ever to become factories that can produce human organs, scientists must find ways to grow cells faster and in a more controllable way. Achemical engineer who trained at the intensively competitive Indian Institute of Technology, Bombay, Surya Mallapragada is closing in on that goal. Mallapragada has designed biodegradable polymer scaffolding to guide the growth of individual cells in the same way that wooden supports guide the tendrils of a grapevine. In experiments, she implanted her scaffolds in rats, tied the ends of torn nerve cells to them and showed that the cells could relink by growing along fine grooves on the polymer surface. Carving the grooves was key;t he usual technique of bombarding the scaffold with ions degrades the polymer, so Mallapragada used alternatives like laser etching and atomic-force microscopy that minimize degradation.To entice tissue to grow quickly, Mallapragada lined the grooves with special cells that ooze growth-inducing proteins. When she’s not busy teaching nerve cells to grow, Mallapragada spends a little time learning tae kwon do.
Sean J. Morrison
Stem cells have become icons of medical hope, and Sean J. Morrison has made fundamental discoveries that explain their workings. As a post- doc at Caltech, Morrison devised a way to harvest neural stem cells from fresh tissue rather than from tissue cultured in the lab, where stem cells might have been created as an artifact of the culturing process. In sodoing, Morrison clarified many of the cells’ properties. “He is one of the most talented stem cell researchers,” says David J. Anderson, Morrison’s advisor at Caltech. While studying similarities between stem cells and cancer cells, Morrison and a collaborator made the surprising discovery that tumor growth may be driven by rare “cancer stem cells.” Now an assistant professor at the University of Michigan, Morrison recently cofounded Cancer Stem Cell Genomics to investigate the possibility that that discovery could lead to better ways of developing cancer-killing drugs. The nascent company is Morrison’s second business foray. In college he developed an inexpensive process for mass-producing fungi, but his company lacked capital. Given stem cells’ medical applications, his new business is not likely to have similar trouble.
Marrying biological and electronic systems could yield advances in drug discovery, bioweapon detection—even computing. But the chemistry must be just right for living cells and electronics to talk. Milan Mrksich is the perfect matchmaker. The University of Chicago chemist coats the surfaces of electronic devices with organic molecules that can convert a chemical signal into an electrical one, and vice versa—creating a means of communication unlike anything developed by the handful of other researchers working on hybrid devices. Cells in a bioweapon detector, for example, could produce an enzyme when infected by a virus. The enzyme would interact with molecules on the surface of a microchip, triggering an electrical signal that would set off an alarm.With funding from the U.S. Department of Defense, Mrksich aims to build a prototype detector within five years. He also envisions computing devices that would exploit the different ways living and electronic systems handle information. But for now, Mrksich is just excited that he’s sparked a conversation between cells and circuits.
Stephen O’Connor is equal parts scientist, engineer and salesman. Armed with a PhD in chemistry from Caltech and more than a dozen jointly held patents, he helped start four companies, raising most of the seed money himself. His first venture made ultrafast optical sensors. His second, Clinical Microsensors, made DNA detection instruments he designed to quickly read the genetic makeup of plant and animal tissue; Motorola bought the company for $300 million. Money and experience gave O’Connor the confidence to found Nanostream in Pasadena, CA, in 1999.It makes custom chips that analyze microscopic amounts of blood or other fluids,some of the first commercial products in the rapidly growing field of microfluidics. With $11 million in funding, Nanostream markets the chips topharmaceutical companies for drug discovery tests. Also in 1999,O’Connor founded CO2,a profitable incubator company that has invested in 11 local scientific startups by outfitting their labs.O’Connor’s Caltech advisor, John Baldeschwieler, says his ex-student is “playing an increasingly important role in the economic development of Pasadena.”
Soon after Watson and Crick found that DNA is made up of four subunits, including one called cytosine ,scientists discovered a so-called fifth subunit: methylated cytosine. Experiments in the 1990s showed that methylated cytosine acts as a switch that can turn a gene on or off. But researchers had trouble distinguishing it from ordinary cytosine. Alexander Olek found an easy way to make it stand out, exposing relationships between the switch and disease. Olek also developed lab techniques for quickly scouring large volumes of DNA for the switch. His work made him a pioneer in “epigenetics,” which explores how environmental factors alter DNA. Olek, who dreams of helicoptering to a mountaintop to ski virgin snow, brings an adventuresome attitude to his work. At 19 he started his first enterprise, which looked for genetic features of diseases common in South America. While he was finishing his doctorate, Olek started Epigenomics in Berlin to advance his methylation work. With $35 million of investment capital, Epigenomics plans to market cancer detection tests that sense tumors’ methylation signals.
If scientists understood how the body’s proteins folded, they could better battle diseases like Alzheimer’s. But analyzing a protein’s trillions of possible folding steps is daunting, even for a supercomputer. In 1999 Vijay Pande, a professor of chemistry and structural biology at Stanford University, wrote algorithms that enable thousands of isolated computers to calculate tiny portions of a folding sequence and combine their solutions. The pragmatic Pande then sought advice from distributed-computing entrepreneur Adam Beberg (a former TR100 honoree) on how to integrate his code into a screen saver that PC users could download. Dubbed Folding@home, the software makes calculations any time the PC’s screen saver is running and reports the results to Pande’s computer. Since the project’s October 2000 debut, some 75,000 volunteers worldwide have helped simulate, for the first time, the complete folding behavior of five important proteins. Born in Trinidad to Indian parents, Pande is now using distributed computing to map the final folded structures of the proteins. On any given day, 35,000 PCs are providing the computing power.
The biotech industry dreams of automating millions of biological experiments on mass-produced chips. Colleagues say Steve Quake has the creativity, intellect and ambition to make it happen. When Quake was a Stanford University postdoc, he investigated the behavior of biological polymers. After becoming a Caltech professor he developed his first micro-fabricated tools, which use electric fields to sort cells and manipulate DNA molecules. Soon thereafter, Quake used soft lithography to build the first set of microvalves and pumps practical enough to be mass-produced, a key step toward developing the hotly anticipated chips. In 1999,Quake and some former college buddies founded San Francisco-based Fluidigm to supply his patented equipment and intellectual property to life science and pharmaceutical companies. Fluidigm has received $50 million in capital and recently signed a deal to supply GlaxoSmithKline.Now a tenured professor, Quake divides his time between researching the structure and function of proteins and devising more-sophisticated microfluidic tools.
“Right now we have no way of saying, ‘Give me a drug candidate, then give me a list of every protein in the human body it interacts with,’” says David Sabatini, whose mellow demeanor is more characteristic of a jazz guitarist than a molecular biologist. “But my technology can do that.” The payoff, he says, could be better drug design .His technology is a glass chip, essentially a microscope slide, spotted with several thousand mammalian cells. Each spot of cells makes a different protein; researchers can wash a potential drug over the slide to see how it interacts with thousands of proteins at once. In the past, testing all those proteins might have taken months. Today a patent on the chip is pending, and Sabatini has raked in $6.5 million in capital for his Cambridge, MA, startup, Akceli. But drug screening is only one application: Sabatini aims to make a cell-based chip that will allow researchers to study every protein encoded in the human genome at once. He says his chip could allow researchers to identify the mutated genes that lead to disease.
John Santini knows all about managing chronic illnesses; he was diagnosed with lupus at age 12 and has been taking daily medication since. Small wonder he chose to pursue drug delivery technology. Today Santini is chief scientific officer of MicroChips, which he cofounded in 1999 to make pills and injections obsolete. The Cambridge, MA, company is developing an implantable chip that stores drugs and releases them at a programmed rate. Santini devised the technology as an MIT grad student. A dime-sized, surgically replaceable chip can hold several hundred single-dose drug reservoirs. Patients could control the chip’s microprocessor remotely—a benefit for, say, patients taking pain medications. MicroChips recently began testing the chip with an undisclosed drug. Santini’s technology could be ideal for delivering new protein drugs. Most proteins must be injected into the bloodstream because they are too fragile to survive the digestive system. But an implanted chip could replace such injections. And with the sequencing of the human genome, Santini says, “There’s going to be an explosion in protein compounds in the next five to 10 years.”
David Schaffer spends most of his time making things grow:I n his garden, orchids; in his lab, stem cells. The biomedical engineer is trying to coax stem cells that lie nearly dormant in the brain to multiply at a much quicker rate than they ordinarily do, which could help regenerate damaged nerve tissue in patients with Alzheimer’s or Parkinson’s diseases. Last year Schaffer and his colleagues discovered a protein that causes stem cells to grow, and he showed that the protein’s action could trigger the repair of nerve cells in mice. By determining how the protein works, Schaffer may be able to get neural stem cells in human patients to replace damaged neurons. To carry the protein to stem cells, Schaffer is using inactivated viruses as delivery vans and is now tinkering with their molecular properties to help them find their targets precisely. Schaffer’s background equipped him well for his work: he grew up in a family of doctors, was interested in mathematics and majored in engineering. Prodding stem cells to grow is harder than cultivating orchids, but the potential rewards are richer, too.
The human immune system defends against foreign objects with vigilance, but Kevin Shakesheff wants to create lasting peace between synthetic surfaces and the biological world. He is building polymer scaffolds, on which living cells can grow, to form the backbones of what will one day be transplant-ready organs, as well as drug delivery vehicles that can steer themselves to target sites. That work began when the pharmaceutical sciences graduate spent a year in the lab of MIT bioengineering pioneer Robert Langer. He returned to the University of Nottingham in his native England to start his own lab. There, Shakesheff figured out how to incorporate stem cells as well as support cells that he calls the “unsung heroes” of tissue regeneration into biodegradable polymer structures for organs. Shakesheff is now using the technique to develop small polymer capsules that can deliver human cells to injury sites. Last year, the hard-working Shakesheff founded Regentec in Nottingham to commercialize his work. He’s forging agreements with pharmaceutical companies to mass- produce miniature tissue and organ samples for drug testing.
Vivek Subramanian is an inventor’s inventor. His credits include a novel memory chip that led him to start Santa Clara, CA-based Matrix Semiconductor; a tiny, award-winning transistor; and his current project, ultracheap, flexible displays for note-taking gadgets. But his greatest ambition is to put small amounts of computing power into everyday items. Subramanian has devised radio frequency sensors that can be printed onto the plastic and paper that wrap fresh foods and packaged goods in stores. He’s confident his University of California, Berkeley, group can produce the circuits for less than one cent each—compared with the current manufacturing cost of one dollar for a conventional radio frequency tag. Such tags on grocery items could give shoppers price and content information, even on-the-spot discounts. A sensor in a carton of milk could measure lactic-acid levels and signal when it’s time for a fresh container. “I’m not looking to make the best and fastest electronic devices,” Subramanian says. “I’m just making them good and fast enough so they can be placed everywhere in everything.”
Christoph Westphal invents startups. He restlessly searches for scientific advances he can transform into practical technologies. After grabbing an MD and PhD at Harvard in a mere six years, Westphal did a two-year stint at consultancy McKinsey, where he designed business development strategies for high-tech firms. He jumped to Waltham, MA-based Polaris Venture Partners in 2000 and is now a general partner advising five startups, one of which he cofounded. Westphal brings more than cash to the table: just ask the people at MIT spinoff Mimeon, in Cambridge, MA. Robert Langer, a prolific MIT inventor, says Mimeon was launched in 2001 after Westphal asked him “penetrating questions” about the underexploited potential of carbohydrate therapeutics. Langer introduced Westphal to MIT bioengineer Ram Sasisekharan and his technology for sequencing complex carbohydrates. Within months, Westphal brought in other scientific experts as well as $2 million in seed money. Mimeon is now zeroing in on its first target: an improved version of the blood thinner heparin—a substance derived from hog intestines that generates $2 billion in sales annually.