Dataset Information

Molecular subtyping of Alzheimer's disease with consensus non-negative matrix factorization.

ABSTRACT: Alzheimer's disease (AD) is a heterogeneous disease and exhibits diverse clinical presentations and disease progression. Some pathological and anatomical subtypes have been proposed. However, these subtypes provide a limited mechanistic understanding for AD. Leveraging gene expression data of 222 AD patients from The Religious Orders Study and Memory and Aging Project (ROSMAP) Study, we identified two AD molecular subtypes (synaptic type and inflammatory type) using consensus non-negative matrix factorization (NMF). Synaptic type is characterized by disrupted synaptic vesicle priming and recycling and synaptic plasticity. Inflammatory type is characterized by disrupted IL2, interferon alpha and gamma pathways. The two AD molecular subtypes were validated using independent data from Gene Expression Omnibus. We further demonstrated that the two molecular subtypes are associated with APOE genotypes, with synaptic type more prevalent in AD patients with E3E4 genotype and inflammatory type more prevalent in AD patients with E3E3 genotype (p = 0.031). In addition, two molecular subtypes are differentially represented in male and female AD, with synaptic type more prevalent in male and inflammatory type in female patients (p = 0.051). Identification of AD molecular subtypes has potential in facilitating disease mechanism understanding, clinical trial design, drug discovery, and precision medicine for AD.

SUBMITTER: Zheng C

PROVIDER: S-EPMC8136734 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background:Matrix factorization is a well established pattern discovery tool that has seen numerous applications in biomedical data analytics, such as gene expression co-clustering, patient stratification, and gene-disease association mining. Matrix factorization learns a latent data model that takes a data matrix and transforms it into a latent feature space enabling generalization, noise removal and feature discovery. However, factorization algorithms are numerically intensive, and hence there is a pressing challenge to scale current algorithms to work with large datasets. Our focus in this paper is matrix tri-factorization, a popular method that is not limited by the assumption of standard matrix factorization about data residing in one latent space. Matrix tri-factorization solves this by inferring a separate latent space for each dimension in a data matrix, and a latent mapping of interactions between the inferred spaces, making the approach particularly suitable for biomedical data mining. Results:We developed a block-wise approach for latent factor learning in matrix tri-factorization. The approach partitions a data matrix into disjoint submatrices that are treated independently and fed into a parallel factorization system. An appealing property of the proposed approach is its mathematical equivalence with serial matrix tri-factorization. In a study on large biomedical datasets we show that our approach scales well on multi-processor and multi-GPU architectures. On a four-GPU system we demonstrate that our approach can be more than 100-times faster than its single-processor counterpart. Conclusions:A general approach for scaling non-negative matrix tri-factorization is proposed. The approach is especially useful parallel matrix factorization implemented in a multi-GPU environment. We expect the new approach will be useful in emerging procedures for latent factor analysis, notably for data integration, where many large data matrices need to be collectively factorized.

Dataset Information

Molecular subtyping of Alzheimer's disease with consensus non-negative matrix factorization.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets