|
|
|
|
||
From two years ago: Emerald Logic, King's College London Collaborate on Identifying Alzheimer's Markers....https://www.genomeweb.com/informatics/emerald-logic-kings-college-london-collaborate-identifying-alzheimers-markers Emerald Logic, King's College London Collaborate on Identifying Alzheimer's Markers Feb 05, 2013 | a GenomeWeb staff reporter . NEW YORK (GenomeWeb News) – Emerald Logic today said that it is collaborating with King's College London on the discovery of biomarkers for Alzheimer's disease and other neurological conditions. Aliso Viejo, Calif.-based Emerald Logic has developed a software product called FACET (Fast Collective Evolution Technology), which automatically generates quantitative models that classify or predict outcomes or disease states and is being used by researchers at King's College to identify the biomarkers. According to the firm, in just six weeks the researchers were able to produce a list of 14 discriminative biomarkers from a set of more than 11,000 markers from 245 study subjects, which were then validated on an additional 82 subjects. Using the markers, plus APOE genetic information and demographics, the researchers were able to produce a mathematical classifier of 94 percent accuracy in distinguishing Alzheimer's study subjects from controls or those with mild cognitive impairment, Emerald Logic said in a statement. The firm is collaborating with Simon Lovestone and Richard Dobson of King's College on the project, which is intended to have a translational focus, developing tools that will be useful for detection and potential treatment of Alzheimer's and other diseases associated with aging. "Emerald Logic's software evolved Alzheimer's disease classifiers from our blood markers with the best accuracy we've seen to date, while simultaneously identifying the most useful markers from a vast dataset including a whole genome transcript assay," Lovestone said in the statement. According to Emerald Logic CEO Patrick Lilley, the firm intends to provide quantitative biomarker discovery and quantitative diagnostic modeling services using its FACET software. He told GenomeWeb Daily News in an email that the firm can do this regardless of the source of the assay data. "We can also integrate assay data with other modalities, including genetic information, demographics, imaging, cognitive tests, vitals, etc.," he said. "Part of the reason we so highly value our collaboration with King's College London is that they take an integrative approach, looking at Alzheimer's disease from many biometric perspectives. In our experience, combining modalities tends to produce major improvements in accuracy and insight." http://www.ddn-news.com/index.php?newsarticle=7135 It’s only logical March 2013 by Jim Cirigliano LONDON—King's College London and Emerald Logic have announced an ongoing partnership in identifying blood markers for Alzheimer's disease and other neurological conditions, and creating innovative screening methods by integrating biomarkers with other diagnostic models. The collaboration will involve King's College's large team of bioinformaticians attempting to discover multiomic blood markers of Alzheimer's disease, with a specific focus on early preclinical markers, markers of progression, rate of decline and diagnosis. Emerald Logic will assist in this search with the computational power of its FAst Collective Evolution Technology, or FACET, software, which can identify which subsets of data in complex datasets show relevant interactions by using non-linear, evolutionary computing to quickly identify factors of interest. The partnership has three successive goals. First is the identification of factors (especially biomarkers) that are significant in Alzheimer's disease, and also the exploration of which factors or sets of factors are interactive in the data. Second is the development of novel quantitative models for screening that yield significant improvements over existing diagnostic models for healthcare providers and patients. Third is the development of a predictive and discriminative model for Alzheimer's disease that will enable treatment to be given quickly to the correct patients as early as possible in the progression of the disease, possibly even before symptoms appear. "Early—ideally, preclinical—identification of Alzheimer's disease pathology using biomarkers is a critically important research goal and could enable more effective therapeutic intervention and improved disease management," says Dr. Richard Dobson, lecturer in bioinformatics at the Institute of Psychiatry at King's College London. "The use of biomarkers to identify individuals with Alzheimer's disease prior to the appearance of clinical symptoms—the so-called predementia phase of the disease—will be essential to the development of drugs for early intervention. Additionally, if sufficiently powered, some biomarkers could be used as part of a screening program for at-risk elderly people." "The goal of the collaboration is to produce a quantitative model that is both human-readable and usable by a computer to serve as an accurate classifier, to discriminate between which patients are going to develop Alzheimer's as opposed to mild cognitive impairment," says Patrick Lilley, CEO of Emerald Logic. Kevin Horgan, an Emerald Logic advisory board member formerly of GE and Merck, has a personal acquaintance with the key collaborators from King's College, and facilitated the two organizations coming together. Their collaborative efforts have already yielded exciting outcomes, having conducted an initial pilot using 6,000 blood gene expression measures from an Alzheimer's disease biomarker dataset, which confirmed that working together could produce eminently useful results. The pilot effort identified 14 discriminative blood markers on 245 study subjects; these markers were validated on an additional 82 study subjects in two separate holdout samples. The team was able to use these markers in combination with APOE genetic information and demographics to produce a mathematical classifier that distinguished Alzheimer's study subjects from controls or with those with mild cognitive impairment with 94-percent accuracy. "Emerald Logic's software evolved Alzheimer's disease classifiers from our blood markers with the best accuracy we've seen to date, while simultaneously identifying the most useful markers from a vast dataset including a whole genome transcript assay," Dr. Simon Lovestone, professor of old age psychiatry at King's College's Institute of Psychiatry, said in a January media release. "Some identified biomarkers confirm some of our existing research, while others do not appear to have been implicated in Alzheimer's disease previously," says Dobson. Today, there are more than 36 million people diagnosed with Alzheimer's disease worldwide; it is estimated that Alzheimer's affects 1 in 8 people over the age of 65. King's College London is one of the top 30 universities in the world, and the fourth oldest in England. It enrolls more than 24,000 students—including more than 10,000 graduate students—from nearly 140 countries, and employs more than 6,100. It is in the top seven U.K. universities for research earnings and had an overall annual income of about $790 million in 2011. Emerald Logic is a pioneer in quantitative personalized medicine and biomarker discovery. The company's FACET software combines mathematics and principles from biology, engineering and particle physics to answer intractable questions with high human and economic impact. The company is based in California. "The prospective partnership with Emerald Logic will represent an important addition to our NIHR Mental Health Biomedical Research Centre and Dementia Unit portfolio of collaborative studies on biomarkers for dementia," says Dobson. "As we move forward, Emerald Logic could play an important role in these collaborations as well as other projects, including the recently funded IMI European Medical Informatics Framework, a large public-private consortium which aims to merge clinical, molecular and neuroimaging datasets from across Europe for studies on Alzheimer's disease." Lilley expresses enthusiasm for working alongside Lovestone and Dobson. "Taking a broad-ranging look at multiple datasets across many types of research—that integrative approach is very powerful," he says. "They're not limiting themselves to looking at one narrow area of research: They're rare in that regard." |
return to message board, top of board |