Novel Stroke biosignature and machine learning model identified for greatest diagnostic accuracy through blood samples

Posted on October 1, 2019

In a close collaboration with a prestigious Austin, TX medical center, InSyBio applied its proprietary technology to identify targeted biomarkers to provide an unbiased, non-interventive diagnostic method for Stroke. This is incredibly important when making clinical decisions in such a time-sensitive situation as stroke, where blood biomarkers could aid in faster diagnosis and treatment. Classifying true stroke from stroke-mimic usually requires multiple tests and imaging whereas blood biomarkers’ use could considerably expedite this process and save not only precious time, but possible brain function due to faster action.

To further elaborate on the importance, Dr. Max Shpak (Center for Systems and Synthetic Biology, University of Texas at Austin) noted: "The paper by K. Theofilatos et al developed as a collaboration between the staff scientists at InSyBio and the research group at this medical center. At the time, a primary focus of the medical center’s research arm was on the clinical and biomedical aspects of stroke and other neurological conditions. InSyBio’s network-based approach to characterizing gene expression patterns provides a novel means of understanding some of the physiological sequelae associated with acute ischemic stroke events, in particular the genetic pathways in nervous, immune, and circulatory system cells that are activated or suppressed in response to stroke. Network models provide researchers with a potentially powerful tool for identifying the most functionally significant genes in these pathways. By focusing on patterns of gene co-expression and interactions among sets of genes, such network-based methods provide a more intuitive and informative means of analysis than standard methods that only consider the changes in expression levels of individual genes. It is hoped that this research will eventually provide the foundation for identifying molecular biomarkers for quantifying the magnitude of stroke effects and patient prognoses, as well as for helping to determine possible treatment options."


Through the use of predictive analytics and network modeling, InSyBio uncovered a signature of 6 genes (ID3: inhibitor of DNA binding 3, HLH protein, MBTPS1: Membrane-bound transcription factor site-1 protease, NOG: Noggin, SFXN2: sideroflexin 2, BMX: BMX non-receptor tyrosine kinase, and SLC22A1: solute carrier family 22 member 1) that had significantly higher accuracy (89.6%) in differentiating between stroke and stroke-mimic patients and of which advanced our understanding of the physiological changes the body undergoes following a stroke. In addition to being the smallest revealed gene set to date, this biosignature is also capable of the highest diagnostic accuracy compared to the current solutions which were tested in the study. This important research is titled: "Discovery of stroke-related blood biomarkers from gene expression network models" and has now been published (link below) in one of the top genomic journals, BMC Medical Genomics.

InSyBio’s CSO, Dr. Seferina Mavroudi commented on this significance saying: "This discovery marks an important milestone in the improved diagnosis of stroke. Not only does the identified biosignature significantly increase diagnostic accuracy but it is also a small, highly targeted gene set which is incredibly important for clinical relevancy as well as the feasibility of creating a diagnostic test to be used in practice."

About InSyBio: InSyBio is an international pioneer bioinformatics company that is revolutionizing the medical field, clinical trials and food/nutrition industries through targeted biomarker discovery and highly-accurate predictive analytics.

Its solutions provide the R&D departments of Pharma, Research Institutes and Food/Nutrition companies, with the means both analytically and computationally to meet their challenging research and innovation goals (such as robust biomarker discovery and accurate predictive models). This is done through a sophisticated, software-as-a-service, machine learning platform which comprehensively integrates multi-omics data reducing the number of samples and additional validation experiments needed. Since 2013, InSyBio has worked with hundreds of academics, large and small companies, hospitals and research centers allowing them to succeed in their biomarker discovery tasks with an increase of at least 10% in predictive accuracy and decrease of at least 80% in overall time and cost.


Visit to see how InSyBio can help you reach your R&D goals in the era of Precision Medicine and Nutrition.

For any further detail contact our Business Development Manager, Mackenzie Hastings, at .

For more News visit our InSyBio Blog.