6 February 2015. Computer scientists designed and tested decision-making models using artificial intelligence for robots to plan and perform unstructured and ill-defined tasks, from doing household laundry to conducting search-and-rescue operations. The team led by Siddharth Srivastava, now with United Technologies in Berkeley, California and Shlomo Zilberstein from University of Massachusetts in Amherst described their work in a paper presented to a meeting of the Association for the Advancement of Artificial Intelligence at the end of January in Austin, Texas.
Srivastava, then a postdoctoral researcher at University of California in Berkeley, and Zilberstein, with colleagues from the two schools, developed solutions to address tasks that are relatively simple, even boring, for humans to organize and carry out, but prove difficult for robots because of their undefined nature. For autonomous systems to conduct these tasks required up to now extensive algorithms that became complex and difficult to generalize to new, unstructured, or ill-defined problems.
The Berkeley-Amherst team, funded by a National Science Foundation grant, applied insights from the way humans approach the tasks, with generalized plans based on repetitive simple decisions and actions that made assumptions about the uncertainty. “These are plans that don’t just work in a particular situation that is very well defined and gets you to a particular goal that is also well defined,” says Zilberstein in an NSF statement, “but rather ones that work on a whole range of situations and you may not even know certain things about it.”
The researchers used laundry as an example of a human task with a minimum of structure — e.g., fixed locations of closet and washing machine — yet still undefined quantities and specifics of items to be washed, such as weight and colors of fabrics. The team wrote computer models, reflected in algorithms based on different types of planning and decision-making models, then programmed those algorithms into routines with loops and branches to direct the robots to first plan, then carry out the work. The models can also adapt to uncertain or unstructured situations by learning from examples, giving them more capability when confronted with previously undefined tasks.
The team tested their models on a PR2 robot, originally made by Willow Garage, a company no longer in business, but that runs on the open-source ROS operating system, and is still commercially supported. The robot was tasked with taking a stack of dirty laundry from a closet and depositing the items in a washing machine. The robot successfully gathered the laundry, placed the items in a basket, took the basket to a laundry room, placed the basket on a table, opened the washer, removed the items from the basket, placed them in the washer, then closed the washer. The researchers say determining the solution plan took the robot less than one second.
“Using this approach, solutions to high-level planning can be generated automatically,” says Srivastava. While laundry provided a convenient and familiar problem for testing the process, the researchers say it can be applied to a more consequential, tedious, and high-risk operations such as manufacturing and search-and-rescue missions.
The following brief video demonstrates the robot performing the laundry task.
5 February 2015. A new report from Brookings Institution says the United States is losing ground to overseas competitors in critical advanced industries that hold the key to the country’s long-term economic future. The study by the Washington, D.C. think tank was discussed in a forum today with six CEOs of U.S. companies and the governor of the nearby state of Virginia.
The Brookings team led by Mark Muro, director of its Metropolitan Policy Program, identified 50 industries as advanced industries, defined as those where R&D spending per worker is in the 80 percentile or higher among all industries, and whose share of workers in occupations requiring a high degree of scientific or technical knowledge is above the national average of 21 percent. The 50 industries include 35 from manufacturing, 12 from services, and 3 in energy sector.
Advanced industries include the most disruptive, market-making technologies, such as additive manufacturing, robotics, genomics, advanced materials, big data analytics, and connecting smart devices into the Internet of things. Companies in these industries tend to cluster in large metropolitan areas, with 70 percent of all advanced industry jobs found in the 100 largest metro regions. San Jose, California is the most concentrated hub with 30 percent of its workforce in advanced industries, followed by Seattle (16%), Wichita (15.5%, mainly aircraft), Detroit, (15%, largely automotive), and San Francisco (14%).
Small numbers of people that pack a big economic punch
As of 2013, say the authors, these 50 industries employed 12.3 million people in the U.S., or about 9 percent of all workers. Yet despite its relatively small size, advanced industries produced some $2.7 trillion in value, or about 17 percent of total U.S. gross domestic product. Advanced industries accounted for 80 percent of engineers employed, 90 percent of all private sector R&D spending, and 85 percent of all patents awarded in the U.S. Moreover, economic activity of the 12.3 million workers in advanced industries support another 14.3 million workers in support or supply-chain jobs.
In addition, people working in advanced industries earn better pay, with pay rates rising faster than the rest of the economy. Advanced industry workers earned about $90,000 a year in total compensation in 2013, compared to $46,000 for the average worker in other sectors.Between 1975 and 2013, average earnings of workers in advanced industries grew by 63 percent when adjusted for inflation, compared to 17 percent for other workers.
The higher pay rates extended across education levels, with people earning an average of $53,000 a year as of 2013 in total compensation for some college but no degree, compared to $38,000 a year elsewhere. Workers with an associates degree in advanced industries earned $58,000 a year in 2013, compared to an average of $55,000 a year for a bachelors degree in other industries.
4 February 2015. Revolution Medicines Inc., a spin-off enterprise based on research in protein chemistry at University of Illinois, is starting up with $45 million in first-round venture funds. The company is founded by Illinois biochemistry professor Martin Burke, and initially financed by Third Rock Ventures, a San Francisco venture capital firm.
Burke’s lab in Urbana, Ilinois studies the ability of low molecular weight chemicals to supplement missing or dysfunctional proteins in the body, particularly the synthesis of chemicals that perform functions similar to proteins. Studies by Burke and colleagues led to a simpler process for synthesizing these complex compounds for therapeutics with a common set of biochemical building blocks.
Among the discoveries in Burke’s lab is a compound that kills yeast cells causing deadly infections, yet apparently without the same toxic effects on healthy cells of amphotericin-B, an anti-fungal drug reserved for life-threatening infections because of those adverse effects. In a paper published in May 2013, Burke and colleagues describe a synthetic derivative of amphotericin that in lab cultures binds to organic molecules found in yeast cells, but avoids similar molecules in human cells. A subsequent paper describes the biochemical activity behind the discovery.
Revolution Medicines, based in Redwood City, California, aims to commercialize the process of synthesizing these chemicals with protein-like properties, offering what it says is a more rapid, rational, and standardized technique of drug design and discovery, rather than traditional one-off methods used up to now. The company holds an exclusive license to Burke’s technology from University of Illinois, and plans to make a synthetic and less toxic alternative to amphotericin-B as its first product.
Burke will serve as chair of Revolution Medicines’ scientific advisory board. The company itself is led by Martin Goldsmith, an academic scientist and senior executive in the biotechnology industry, and a partner at Third Rock Ventures, a venture capital company specializing in life sciences.
Burke tells more about his lab’s technology in the following video.
4 February 2015. A biologic therapy for rheumatoid arthritis was shown to work as well as an approved name-brand biologic treatment in a late-stage clinical trial, according to findings released by biotechnology company Amgen. The study found Amgen’s biologic code-named ABP501 performed as well as adalimumab, marketed as Humira by AbbVie, in relieving symptoms of patients with moderate to severe rheumatoid arthritis.
Rheumatoid arthritis is an autoimmune disease, where the immune system is tricked into attacking healthy cells, that leads to inflammation of joints (wrists, fingers, feet, and ankles) and surrounding tissue. In the U.S., rheumatoid arthritis is estimated to affect1.3 million people, making it the most prevalent autoimmune disease. While the rate of occurrence is declining, the disorder occurs in twice as many women as men.
Adalimumab is an engineered antibody designed to relieve pain and inflammation caused by rheumatoid arthritis by limiting the effects of tumor necrosis factors, proteins in the body known to cause inflammation. The drug, given as an injection under the skin, can also treat other autoimmune disorders including Crohn’s disease, ulcerative colitis, psoriatic arthritis, and chronic plaque psoriasis. ABP501 is also an engineered antibody designed to block the actions of tumor necrosis factors, with a dosage similar to adalimumab.
The clinical trial enrolled 526 patients with moderate to severe rheumatoid arthritis at 111 sites in North America and Europe. Patients were randomly assigned to receive either an injection of adalimumab or ABP501 every 2 weeks for 22 weeks. Researchers looked primarily at improvement in a standard index of symptom occurrence devised by American College of Rheumatology, using a gain of 20 percent or more after 24 weeks as the main measure of effectiveness. The study also tallied patients’ scores on other effectiveness scales over the study period, as well as safety indicators and ability of the drugs to generate an immune response.
Amgen, in Thousand Oaks, California, says patients given ABP501 scored within a specified margin on the American College of Rheumatology symptom index after 24 weeks as patients given adalimumab, indicating clinical equivalence. In addition, patients receiving ABP501 had similar adverse effects as patients receiving adalimumab, as well as similar immune responses.
In October 2014, Amgen released results of a similar test of ABP501 with patients having moderate-to-severe plaque psoriasis. That trial showed ABP501 relieved plaque psoriasis symptoms — red, itchy, scaly skin — among patients as well as adalimumab, with similar safety and immune response results.
3 February 2015. A new initiative, called iConquerMS, is recruiting 20,000 patients with multiple sclerosis in the U.S. to offer their health data and research ideas to find a cure for the disease. iConquerMS is an undertaking of the Accelerated Cure Project for MS, with the research network segment funded by Patient-Centered Outcomes Research Institute or PCORI.
The initiative aims to harness big data tools to help advance research in finding treatments for multiple sclerosis. The disease is known as an auto-immune disorder, where the body’s immune system is tricked into attacking the protective myelin sheath that covers nerve cells. Damage to the myelin interferes with signals between the brain and the rest of the body, causing a wide variety and severity of of symptoms, but can lead to disability. There is no cure for multiple sclerosis and current treatments are designed to help recover from attacks and manage symptoms.
iConquerMS is seeking widespread participation by patients to collect a large enough pool of data on multiple sclerosis to enable more comprehensive and detailed analysis of patterns in the disease that would not otherwise be apparent. The initiative’s organizers believe the data can help answer questions about causes of multiple sclerosis, identify candidates for various therapies, and find new or better treatments for the disease.
Data for iConquerMS are expected to come from patients’ health records, with identifying details removed. Participants will also be asked to complete occasional questionnaires, including suggestions for new studies, and feedback on the project.
The Accelerated Cure Project for MS sponsoring iConquerMS and based in Waltham, Massachusetts, promotes faster progress on treatments for multiple sclerosis, and has underway a repository of bio-specimen samples from more than 3,000 subjects, a large-scale collaborative research program, and an online community for researchers. Feinstein Kean Healthcare, a life sciences communications firm, wrote and designed the iConquerMS Web site content. Complex Adaptive Systems, a part of Arizona State University, working with with Life Data Systems Inc., provides the information technology behind the system.
In addition, iConquerMS is part of PCORNet, a national health care data network for researchers under the Patient-Centered Outcomes Research Institute or PCORI, a government-chartered organization that funds comparative effectiveness research for making better-informed health care decisions. PCORI is funding the national network portion of the project, one of its patient-powered research networks as the organization calls them. According to PCORNet’s Web site, funding for the network is about $989,000 and extends for 1.5 years.
3 February 2015. Lysosomal Therapeutics Inc., a biotechnology company developing treatments for inherited neurodegenerative disorders, raised $20 million in its first venture funding round. The Cambridge, Massachusetts enterprise, begun last year, is backed by several venture capital funds and pharmaceutical companies.
Lysosomal Therapeutics is based on the research ofDimitri Krainc, a neurologist now at Northwestern University, but earlier at Harvard Medical School and Massachusetts General Hospital in Boston when he and colleagues founded the company. Krainc’s research investigates accumulation of mutated proteins that occur in several neurodegenerative diseases, such as Parkinson’s disease, as well as processes that degrade the functioning of lysosomes, the parts of cells that digest and clear these aberrant proteins.
Parkinson’s diseaseis a progressive disease of the central nervous system and affects more than 1 million in North America and some 4 million worldwide. Symptoms often begin as tremors or shaking in limbs, but can advance into psychiatric conditions, including depression and hallucinations.
Studies done by Krainc and colleagues trace lysosomal storage disorders to mutations in the gene forglucocerebrosidase, also known as GBA or GCase, which are linked to Parkinson’s and Gaucher disease. Their findings indicate increased activity of GCase in neurons of Parkinson’s disease patients can help normalize their lysosomal storage functions. Lysosomal Therapeutics is developing a molecular therapy that stimulates GCase, as a treatment for Parkinson’s disease. The company is also investigating other lysosomal enzyme deficiencies as potential therapy targets.
First-round investors in Lysosomal Therapeutics are most of the same seed fund providers in May 2014: Atlas Venture, Hatteras Venture Partners, Lilly Ventures, Sanofi-Genzyme BioVentures, Roche Venture Fund, Partners Innovation Fund and several original angel investors. Atlas Venture led the round, raising $20 million.
In addition to gaining nearly $5 million in seed funds, Lysosomal Therapeutics also received a $230,000 Michael J. Fox Foundation grant in December 2014 to study GCase as a target for Parkinson’s disease therapies, as well as a biomarker for diagnostics. The biomarker can also help predict Parkinson’s disease patients most likely to respond to GCase treatments.
Proceeds from the new financing are expected to advance Lysosomal Therapeutics Inc.’s (LTI’s) development of a Parkinson’s disease candidate, but also related conditions. Bruce Booth, a partner at Atlas Venture noted in a company statement that investors “are not only supporting further development of LTI’s breakthrough therapeutic mechanism for the treatment of Parkinson’s disease, but we are also enabling LTI to expand beyond Parkinson’s disease and build out a platform around additional novel targets implicated in underserved orphan and neurodegenerative disorders.”
Sherry Pagoto (Univ. of Massachusetts Medical School)
2 February 2015. A team from University of Massachusetts Medical School and Worcester Polytechnic Institute are creating a smartphone app combined with a cloud-based data store to help people who overeat due to stress control their eating. The system known as the RELAX Application Suite is funded by a 3-year, $2 million grant from National Heart, Lung, and Blood Institute, part of National Institutes of Health.
The project includes research and development of technology for obese patients and their physicians needing a lifestyle intervention to reduce overeating and weight gain from stress. For some patients, stress is associated with out-of-control eating, more junk food consumption, and less consumption of healthier food items such as produce and whole grains.
The research team, headed by UMass medical school psychologist Sherry Pagoto and Worcester Tech business/technology professor Bengisu Tulu, aim to design a system that fits into interventions planned by clinicians to help patients control their weight. The system is expected to allow obese patients to track their daily activities with their smartphones, and make the data available to their clinicians who can monitor the results and make any needed adjustments in treatment.
Bengisu Tulu (Worcester Polytechnic Institute)
The researchers say the new system will be an improvement over most mobile solutions now available. “Most commercial apps available today focus on tracking diet and exercise, but do not help the user understand why they are eating so much and/or exercising so little,” notes Pagoto in a statement by the institutions. “Our clinical and research experience suggests that stress is a very common trigger for overeating and it is a barrier to exercise.”
The data collected will also offer a more complete picture of the patients’ lifestyle than most apps available today. The smartphone app is expected to track daily activities and events, including stress-inducing events, along with detailed eating patterns. The app will collect its data with bar code scanning, GPS coordinates, and text inputs, then provide feedback to the patient about times of day producing the most stress, itemized lists of food consumed, and the relationship between stress and their food intake. The RELAX suite will store the data in an online Microsoft Health Vault database that can be accessed by clinicians to track their patients’ progress and make changes in counseling or treatments as required.
Pagoto, Tulu, and colleagues plan to pilot test the RELAX system against a generic weight-loss app, and an intervention with no technology. The pilot test will be preceded by a usability study with patients and clinicians. Depending on the pilot test results, the researchers would then proceed with a more comprehensive clinical trial of the RELAX system combined with a brief lifestyle intervention compared to the lifestyle intervention alone.
The goal of the overall system is to devise an intervention strategy that reduces the patients’ weight and stress in half of the visits needed by traditional methods. Part of that goal is to identify the real factors causing weight gain and provide coaching or counseling on the spot to patients. “Imagine a person driving into the parking lot of a fast-food restaurant, at a certain time of day,” says Tulu, “and getting prompted with a message asking them to think about what they are feeling and whether or not it is the right time to eat.”
2 February 2015. An experimental engineered antibody for treating lung cancer received a breakthrough designation from the U.S. Food and Drug Administration. Genentech, a biotechnology subsidiary of drug maker Roche, says breakthrough status was granted for its cancer immunotherapy code-named MPDL3280A to treat programmed death-ligand 1 or PD-L1 positive non-small cell lung cancer whose disease has progressed during or after platinum-based chemotherapy.
Non-small cell lung cancer is the most common form of lung cancer, accounting for about 85 percent of all cases. American Cancer Society estimates more than 221,000 new cases of lung cancer will be diagnosed in the U.S. during 2015, affecting somewhat more men than women, leading to 158,000 deaths.
PD-L1 is a protein associated with a number of solid tumors, including this common type of lung cancer. MPDL3280A counteracts the PD-L1 protein’s ability to bind to receptors on T-cells, white blood cells in the body’s immune system. In addition, the company believes MPDL3280A helps activate T-cells, thus restoring their roles in the immune system of finding and attacking tumor cells.
Genentech is testing MPDL3280A with patients of skin (melanoma) and bladder as well as lung cancer, in early stage clinical trials, combined with currently approved cancer drugs bevacizumab, marketed as Avastin, or vemurafenib, brand-named Zelboraf. Intermediate-stage trials of MPDL3280A with lung cancer patients are in progress. In May 2014, FDA granted MPDL3280A breakthrough status for patients with metastatic bladder cancer, marked as PD-L1 positive.
The new breakthrough designation applies as well to MPDL3280A in patients with non-small cell lung cancer, characterized as PD-L1 positive, but also who whose disease progressed during or since treatments with platinum-based chemotherapy. Platinum increases the ability of chemotherapy drugs to bind to cellular DNA, and is often used to treat these kinds of cancer, but tumors can sometimes develop a resistance to it.
FDA assigns a breakthrough designation to therapies that address a serious condition and demonstrate through clinical evidence that the treatment is a substantial improvement over current methods. Once designated as a breakthrough therapy, a drug or biologic can receive early and frequent communications with FDA staff, quick resolution of questions, intensive guidance on drug development, and an organizational commitment from top managers.
Oral polio vaccine being given in Democratic Republic of the Congo (USAID.gov)
30 January 2015. Researchers at Stanford University in California are beginning an interdisciplinary project to improve the way vaccines harness the immune system for protecting against disease. The initiative, which will establish a Human Systems Immunology Center at Stanford, is funded by a 10-year, $50 million grant from the Bill and Melinda Gates Foundation.
Immunologist and medical school professor Mark Davis is heading the project, who with colleagues aim to better understand the human immune system to develop more effective vaccines against infectious diseases. The effort will reside at Stanford’s Human Immune Monitoring Center, which Davis started in 2007. The lab provides standardized immune monitoring assessments at basic cellular and protein levels, and develops new technologies for immune monitoring.
The project plans to assess some of today’s ideas and approaches behind vaccines. Researchers will investigate reasons why some people build a resistance to infectious disease, while others fall victim to the same pathogens. Likewise, the team plans to look into factors in preclinical studies using animals for testing new vaccines that do not translate into positive outcomes when tested with humans. In addition, the Stanford researchers expect to investigate obstacles encountered in testing vaccines with conventional large-scale clinical trials that can take years to conduct and produce definitive results.
“What we need is a new generation of vaccines and new approaches to vaccination,” says Davis in a university statement. “This will require a better understanding of the human immune response and clearer predictions about vaccine efficacy for particular diseases.”
The new Human Systems Immunology Center plans to combine medical science with expertise well outside biology to generate new solutions, including Stanford’s engineering school. Expertise will be recruited from both the Stanford campus and outside, and from fields reaching into genetics, health policy, and photonics.
Vaccines are a key priority for the Gates Foundation, which aims to reduce the cost and extend the reach of vaccines, particularly in low-resource regions where an estimated 1.5 million children die each year from preventable diseases such as diarrhea and pneumonia. The foundation supports efforts to boost countries’ immunization systems, develop new diagnostic tools assess immunity to disease in a population, and strengthen supply chains that transport, store, and distribute vaccines to recipients.
Earlier this week, Bill Gates told a vaccine conference in Berlin, “Over the last 15 years, we have learned a lot about what it takes to produce and introduce new vaccines. We’ve also learned a lot about the challenges of reaching all children. Increasing coverage and vaccine equity will require us to double down on what we know works and harness innovation to overcome obstacles that have capped our success for years. The good news is that we have a strong base to build on.”
30 January 2015. Spark Therapeutics Inc., a biotechnology company in Philadelphia, is raising $161 million in its initial public stock offering. The company, developing genetic therapies for inherited diseases, priced 7 million shares of its common stock at $23.00 a share. The stock trades on the NASDAQ exchange under the symbol ONCE, and as of 12 noon on 30 January, was trading at $42.28.
Spark Therapeutics develops one-time gene therapiesfor inherited diseases, based on research conducted at Children’s Hospital of Philadelphia, a teaching hospital of University of Pennsylvania. The company’s technology uses engineeredadeno-associated viruses, benign microbes designed to deliver genetic material into the cells of patients with defective genes. The company began in October 2013, founded byfour scientistsfrom Children’s Hospital, including hematologist Katherine High, who now serves as Spark’s president and chief scientist.
Spark’s lead product, code-namedSPK-RPE65, is an experimental treatment for the inherited retinal disorders Leber’s congenital amaurosis and retinitis pigmentosa now in late-stage clinical trials. The therapy is designed to fix a mutation in the RPE65 gene expressed in the retina. The gene encodes a protein that helps convert light entering the eye into electrical signals transmitted to the brain, making sight possible. Without this protein functioning properly, visual cycles are disrupted and impaired, leading to blindness.
In addition, Spark has a gene therapy for choroideremia, another rare inherited retinal disease, code-named SPK-CHM in early-stage clinical trials. Choroideremia is associated with mutations in the CHM gene that block production of a protein needed to transport other key proteins to various cells in the body, including the retina.
Another Spark program develops gene therapies to treat hemophilia B, an inherited disorder that results from the absence of a clotting protein in blood known as factor IX or FIX. People with hemophilia B bleed longer than people with FIX, with some 60 percent of cases considered severe, meaning people with the condition experience bleeding after an injury, including spontaneous bleeding into muscles and joints.
A Spark-developed therapy, code-namedSPK-FIX, is designed to deliver genetic material with an adeno-associated virus to the liver, where coagulation factors including FIX are produced. The company says proof-of-concept and early-stage clinical trials show SPK-FIX expresses a therapeutic gene in the liver that stimulates FIX at sufficient levels for patients to discontinue preventive or replacement treatments for some periods of time.
In December, Spark began a licensing and collaboration agreement with drug maker Pfizer to further develop and commercialize SPK-FIX. Under the deal, Spark can gain as much as $280 million in return for sharing worldwide rights to SPK-FIX with Pfizer.
Science and Enterprise offers researchers, administrators, business people, government officials, and interested observers news and insights into the value chain connecting scientific discovery with the marketplace.