Clustering of Alzheimer’s and Parkinson’s disease based on genetic burden of shared molecular mechanisms

Abstract

One of the visions of precision medicine has been to re-define disease taxonomies based on molecular characteristics rather than on phenotypic evidence. However, achieving this goal is highly challenging, specifically in neurology. Our contribution is a machine-learning based joint molecular subtyping of Alzheimer’s (AD) and Parkinson’s Disease (PD), based on the genetic burden of 15 molecular mechanisms comprising 27 proteins (e.g. APOE) that have been described in both diseases. We demonstrate that our joint AD/PD clustering using a combination of sparse autoencoders and sparse non-negative matrix factorization is reproducible and can be associated with significant differences of AD and PD patient subgroups on a clinical, pathophysiological and molecular level. Hence, clusters are disease-associated. To our knowledge this work is the first demonstration of a mechanism based stratification in the field of neurodegenerative diseases. Overall, we thus see this work as an important step towards a molecular mechanism-based taxonomy of neurological disorders, which could help in developing better targeted therapies in the future by going beyond classical phenotype based disease definitions.

Caution, “normal” BMI: health risks associated with potentially masked individual underweight—EPMA Position Paper 2021

“…markers measured in biofluids, imaging modalities and various omics data), AI models can be built that learn from complex signals hidden in the data to make personalised predictions based on the biomarker signatures of individuals. Such models can, for example, predict clinical disease onset [ 11 , 15 – 17 , 71 , 142 ], stratify patients into distinct sub-groups [ 10 , 143 – 145 ], help understanding disease progression [ 71 , 146 ] and aid as clinical decision support tools [ 147 , 148 ].…”

Leveraging the Potential of Digital Technology for Better Individualized Treatment of Parkinson's Disease

“…Recent noteworthy examples from the area of PD include a machine learning approach to predict the risk of an individual patient to receive the clinical diagnosis PD using routinely collected data from electronic health records about 5 years in advance ( 79 ). A further example is a machine learning approach to cluster AD and PD patients into 4 different subgroups based on the genetic burden on 15 molecular mechanisms ( 80 ). The authors in ( 81 ) developed a machine learning approach to predict the progression of PD using a signature of 27 inflammatory cytokines measured in blood serum.…”

Animal models in the study of Alzheimer's disease and Parkinson's disease: A historical perspective

“…Mounting evidence suggests a clinical and pathological overlap between AD and PD [7][8][9] (Figure 1) and transgenic models of synucleinopathy that simultaneously reflect AD-reminiscent cognitive impairment 10,11 have even prompted genome-wide association studies (GWAS) into establishing a possible genetic link between the two diseases. 12,13 However, being ultimately unique to human beings, modelling AD and PD in lesser animals is a steep learning curve that is the subject of this review. Through the discussion that follows, we aim to highlight not only how animal models have been instrumental in the explication of disease pathology but also how modelling practices have evolved over the years.…”