Abstract
Based on increasing evidence suggesting that MS pathology involves alterations in bioactive lipid metabolism, the present analysis was aimed at generating a complex serum lipid-biomarker. Using unsupervised machine-learning, implemented as emergent self-organizing maps of neuronal networks, swarm intelligence and Minimum Curvilinear Embedding, a cluster structure was found in the input data space comprising serum concentrations of d = 43 different lipid-markers of various classes. The structure coincided largely with the clinical diagnosis, indicating that the data provide a basis for the creation of a biomarker (classifier). This was subsequently assessed using supervised machine-learning, implemented as random forests and computed ABC analysis-based feature selection. Bayesian statistics-based biomarker creation was used to map the diagnostic classes of either MS patients (n = 102) or healthy subjects (n = 301). Eight lipid-markers passed the feature selection and comprised GluCerC16, LPA20:4, HETE15S, LacCerC24:1, C16Sphinganine, biopterin and the endocannabinoids PEA and OEA. A complex classifier or biomarker was developed that predicted MS at a sensitivity, specificity and accuracy of approximately 95% in training and test data sets, respectively. The present successful application of serum lipid marker concentrations to MS data is encouraging for further efforts to establish an MS biomarker based on serum lipidomics.
Highlights
Multiple sclerosis (MS) is regarded as a chronic inflammatory, demyelinating and neurodegenerative autoimmune disease that affects the central nervous system[1]
Unsupervised machine learning, applied to identify structures in the data space D = {xi, i = 1, ... , 403} ⊂ d, provided an emergent self-organizing feature map (ESOM), in which large U-heights indicated a large gap in the data space, whereas low U-heights indicated that the points are close to each other in the data space, indicating structure in the data set (Fig. 3A)
It was concluded that the set of lipid markers included information suitable to separate multiple sclerosis (MS) from controls, which had been the aim of this data analysis step
Summary
Multiple sclerosis (MS) is regarded as a chronic inflammatory, demyelinating and neurodegenerative autoimmune disease that affects the central nervous system[1]. The disease course is mostly characterized by a worsening of non-remitting clinical symptoms with each additional relapse[2]. Lipid metabolism has been suggested, among others[1], to be a major pathophysiological mechanism of multiple sclerosis (MS)[16], even that MS is a disease due to disturbed lipid metabolism[17]. Recent research addressing ceramides in MS show that these lipids modify the course of experimental MS models[22,24]. The benefits of cannabinoids for symptomatic control of MS-associated pain and muscle spasms[25,26,27] and experimentally proven anti-neuro-inflammatory effects of cannabinoids[28,29] further suggest a contribution of bioactive lipids to symptom control, resolution of inflammation and possibly remyelination[17]
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call