Abdallah M. Eteleeb, Brenna C. Novotny, Carolina Soriano Tarraga, Christopher Sohn, Eliza Dhungel, Logan Brase, Aasritha Nallapu, Jared Buss, Fabiana Farias, Kristy Bergmann, Joseph Bradley, Joanne Norton, Jen Gentsch, Fengxian Wang, Albert A. Davis, John C. Morris, Celeste M. Karch, Richard J. Perrin, Bruno A. Benitez, Oscar Harari.
Abstract
Unbiased data-driven omic approaches are revealing the molecular heterogeneity of Alzheimer disease. Here, we used machine learning approaches to integrate high-throughput transcriptomic, proteomic, metabolomic, and lipidomic profiles with clinical and neuropathological data from multiple human AD cohorts. We discovered 4 unique multimodal molecular profiles, one of them showing signs of poor cognitive function, a faster pace of disease progression, shorter survival with the disease, severe neurodegeneration and astrogliosis, and reduced levels of metabolomic profiles.
Introduction
Alzheimer disease (AD) is a heterogeneous multifactorial neurodegenerative disorder pathologically characterized by amyloid (Aβ) plaques, neurofibrillary tangles (NFTs), neuroinflammation, and synaptic and neuronal loss. Recently, distinct spatiotemporal trajectories of tau pathology, brain atrophy, postmortem brain transcriptomics profiles, or cerebrospinal fluid proteomics have been associated with multiple clinical and pathological AD features.
Methods:
Frozen postmortem parietal lobe tissue samples from the Knight Alzheimer Disease Research Center (Knight ADRC) participants were provided by the Knight ADRC Neuropathology Core. Written informed consent for research use was obtained from all participants or their family members. The informed consent was approved by the Institutional Review Boards of Washington University School of Medicine in St. Louis, and research was carried out in according to the approved protocols and to the principles of the Helsinki Declaration.
Discussion
In the current study, we leveraged machine learning approaches to integrate high-throughput cross-omics data from controls and AD cases in multiple cohorts and brain regions. Our results highlight the utility of integrating cross-omics data and indicate that cross-omics signatures can better capture differences and discriminate molecular variations in complex and heterogeneous diseases such as AD. In the discovery stage, we used data from the parietal cortex, an understudied brain region affected in later stages of AD.
Acknowledgments
We thank all the participants, their families, the many involved institutions, and their staff, whose help and participation made this work possible. This work was supported by access to equipment made possible by the Hope Center for Neurological Disorders and the Departments of Neurology and Psychiatry at Washington University School of Medicine.
Citation: Eteleeb AM, Novotny BC, Tarraga CS, Sohn C, Dhungel E, Brase L, et al. (2024) Brain high-throughput multi-omics data reveal molecular heterogeneity in Alzheimer’s disease. PLoS Biol 22(4): e3002607. https://doi.org/10.1371/journal.pbio.3002607
Editor: Yejin Kim, The University of Texas Health Science Center at Houston, UNITED STATES
Received: September 6, 2023; Accepted: March 28, 2024; Published: April 30, 2024.
Copyright: © 2024 Eteleeb et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Omics raw data from the Knight ADRC participants are available by request at the NIAGADS Knight ADRC collection (https://www.niagads.org/knight-adrc-collection) with accession numbers NG00083 (https://www.niagads.org/datasets/ng00083), NG00102 (https://www.niagads.org/datasets/ng00102), NG00113 (https://dss.niagads.org/datasets/ng00113/), and NG00108 (https://dss.niagads.org/datasets/ng00108/) for transcriptomics (bulk), proteomics, metabolomics, and single-nuclei RNA-seq respectively. Access to all individual-level data requires an approved NIAGADS application. Documents required for NIAGADS data request can be found here (https://www.niagads.org/resources/documents-and-guidelines) and instructions on how to submit a data access request are available here (https://www.niagads.org/data/request/data-request-instructions). MSBB transcriptomics and proteomics raw data are publicly available at Synapse under Synapse IDs syn3157743 (https://www.synapse.org/#!Synapse:syn3157743) and syn25006650 (https://www.synapse.org/#!Synapse:syn25006650) respectively. ROSMAP transcriptomics and metabolomics raw data are publicly available at Synapse under Synapse IDs syn17008934 (https://www.synapse.org/#!Synapse:syn17008934) and syn25878459 (https://www.synapse.org/#!Synapse:syn25878459) respectively. Information on how to access controlled data (e.g., individual-level human data) can be found at the AD Knowledge Portal (https://adknowledgeportal.synapse.org/Data%20Access). Uncontrolled data (e.g. processed data) can be downloaded with only a Synapse account. Code availability: The analysis codes used to integrate cross-omics data for the three cohorts and all downstream analyses were deposited in Zenodo under the DOI: 10.5281/zenodo.10729969.
Funding: Research reported in this work was supported by the Knight Alzheimer Disease Research Center at Washington University School of Medicine through the National Institute on Aging (NIA: grant no. P30 AG066444 - JCM), Healthy Aging and Senile Dementia (HASD: grant no. P01AG003991 - JCM), and Antecedent Biomarkers for Alzheimer Disease: The Adult Children Study (ACS: grant no. P01AG026276 - JCM). This work was supported by grants from the National Institutes of Health: R01AG057777 (OH), R01AG074012 (OH), U01AG072464 (OH), K01AG046374 (CMK), R56AG067764 (OH), R01NS118146 (BAB), R21NS127211 (BAB), and K25AG083057 (AME) and by the Chan Zuckerberg Initiative (CMK) and the BIDMC 2023 Translational Research Hub Spark Grant Award (BAB). O.H. is an Archer Foundation Research Scientist. A.M.E. is a scholar recipient of the Knight ADRC Research Education Component (NIA: grant no. P30 AG066444). The funders of the study had no role in the collection, analysis, or interpretation of data, in the writing of the report, or in the decision to submit the paper for publication.
Competing interests: The authors have declared that no competing interests exist.