A Most Efficient and Convergent Principal Component Analysis (PCA) Method for Big Data

Big data usually means big sample size with many outliers, in which traditional scalable L2-norm principal component analysis (L2-PCA) will fail. Current existing L1-norm PCA (L1-PCA) methods can improve robustness over outliers, however, its scalability is usually limited in either sample size or dimension size.  The inventor proposes an online flipping method to solve L1-PCA challenges, which is not only convergent asymptotically (or with big data), but also achieves most efficiency in the sense each sample is visited only once to extract one principal component (PC).

Cannabinoid Receptor Meditating Compounds for Metabolic Disease

There is evidence that the metabolic effects of endocannabinoids are mediated by CB1 receptors in peripheral tissues. While prior attempts at generating CB1 receptor blockers have had serious neuropsychiatric side effects, inventors at NIH have discovered compounds that block CB1 receptors with reduced brain penetrance. In addition, some of these compounds also have a direct inhibitory effect on inducible nitric oxide synthase (iNOS), whereas another group of the compounds directly activates AMP kinas.

Treatment for Wolfram Syndrome and Other Endoplasmic Reticulum Stress Disorders with Endoplasmic Reticulum Calcium Modulators

This technology includes the use of JTV-519 and oxidized form of JTV-519, as a novel treatment for Wolfram syndrome and other diseases associated with endoplasmic reticulum (ER). JTV-519 can prevent the leakage of ER calcium to the cytosol and abnormal activation of a pro-apoptotic enzyme, calpain 2, in cell models of Wolfram syndrome. Further, these compounds can prevent cell death in beta cell models of these diseases.
Subscribe to NIDA