Using machine intelligence to uncover Alzheimer’s disease progression heterogeneity

Abstract

Aim: Research suggests that Alzheimer’s disease (AD) is heterogeneous with numerous subtypes. Through a proprietary interactive ML system, several underlying biological mechanisms associated with AD pathology were uncovered. This paper is an introduction to emerging analytic efforts that can more precisely elucidate the heterogeneity of AD. Methods: A public AD data set (GSE84422) consisting of transcriptomic data of postmortem brain samples from healthy controls (n = 121) and AD (n = 380) subjects was analyzed. Data were processed by an artificial intelligence platform designed to discover potential drug repurposing candidates, followed by an interactive augmented intelligence program. Results: Using perspective analytics, six perspective classes were identified: Class I is defined by TUBB1, ASB4, and PDE5A; Class II by NRG2 and ZNF3; Class III by IGF1, ASB4, and GTSE1; Class IV is defined by cDNA FLJ39269, ITGA1, and CPM; Class V is defined by PDE5A, PSEN1, and NDUFS8; and Class VI is defined by DCAF17, cDNA FLJ75819, and SLC33A1. It is hypothesized that these classes represent biological mechanisms that may act alone or in any combination to manifest an Alzheimer’s pathology. Conclusions: Using a limited transcriptomic public database, six different classes that drive AD were uncovered, supporting the premise that AD is a heterogeneously complex disorder. The perspective classes highlighted genetic pathways associated with vasculogenesis, cellular signaling and differentiation, metabolic function, mitochondrial function, nitric oxide, and metal ion metabolism. The interplay among these genetic factors reveals a more profound underlying complexity of AD that may be responsible for the confluence of several biological factors. These results are not exhaustive; instead, they demonstrate that even within a relatively small study sample, next-generation machine intelligence can uncover multiple genetically driven subtypes. The models and the underlying hypotheses generated using novel analytic methods may translate into potential treatment pathways.

Small Patient Datasets Reveal Genetic Drivers of Non-Small Cell Lung Cancer Subtypes Using a Novel Machine Learning Approach

“…This organizational technique was used to extract insights from models that could then be compared with statistical methods suitable for small data. An interactive hypothesis-generating interface was used such that human interaction could facilitate the analysis of different models [44,45]. This methodology allows the user to explore hypotheses generated by the unsupervised clustering methods of the system.…”

“…This organizational technique was used to extract insights from models that could then be compared with statistical methods suitable for small data. The only proprietary method used for these results are the techniques referred to as a feature selection tool [20, 21], in order to help us reduce the size of the data set to 16 dimensions. More specifically, we used these methods to create several new 16-dimensional data sets.…”

“…2) At this point we exercised two options: a) we used methods [20] to arrive at 16 variable data sets (in order to test this system), and b) we allowed step 1 above to be our sole variable selection method. For replication purposes, one may run step 1 alone.…”

“…Clustering was performed via principal components, t-SNE, HDBSCAN and UMAP and these were the basis of the maps found in this paper. Some proprietary algorithms were used to organize the resulting clustering models, in addition to the random forest models, such that we were able to explore the models interactively to derive a deeper understanding of the driving genes behind the sub-clusters [20]. The NetraAI system goes beyond these capabilities, but we did not utilize these proprietary methods to maintain academic standards.…”

Shattering cancer with quantum machine learning: A preview

“…As an example, in Qorri et al, 4 the authors used a novel methodology that encompasses these characteristics in order to extract the driving etiologies behind subpopulations of Alzheimer's patients. This, and the article previewed here, makes it clear that not only is it worth exploring novel machine learning approaches but that quantum computing can potentially provide major opportunities to process datasets like never before.…”