Inferring protein expression changes from mRNA in Alzheimer’s dementia using deep neural networks

Identifying the molecular systems and proteins that modify the progression of Alzheimer’s disease and related dementias (ADRD) is central to drug target selection. However, discordance between mRNA and protein abundance, and the scarcity of proteomic data, has limited our ability to advance candidate targets that are mainly based on gene expression. Therefore, by using a deep neural network that predicts protein abundance from mRNA expression, here we attempt to track the early protein drivers of ADRD. Specifically, by applying the clei2block deep learning model to 1192 brain RNA-seq samples, we identify protein modules and disease-associated expression changes that were not directly observed at the mRNA level. Moreover, pseudo-temporal trajectory inference based on the predicted proteome became more closely correlated with cognitive decline and hippocampal atrophy compared to RNA-based trajectories. This suggests that the predicted changes in protein expression could provide a better molecular representation of ADRD progression. Furthermore, overlaying clinical traits on protein pseudotime trajectory identifies protein modules altered before cognitive impairment. These results demonstrate how our method can be used to identify potential early protein drivers and possible drug targets for treating and/or preventing ADRD.

Data used in these analyses, the predicted proteome data, and the estimated pseudotimes can be requested at the RADC Resource Sharing Hub at www.radc.rush.edu or downloaded from the Synapse repository (http://dx.doi.org/10.7303/syn3219045, http://dx.doi.org/10.7303/syn3159438). The deep-neural protein translation model is available at https://github.com/stasaki/clei2block and http://dx.doi.org/10.7303/syn23624037. The code for trajectory analysis is available at https://github.com/stasaki/SCORPIUS. The RNA-seq and protein data are available via the AD Knowledge Portal (https://adknowledgeportal.org). The AD Knowledge Portal is a platform for accessing data, analyses, and tools generated by the Accelerating Medicines Partnership (AMP-AD) Target Discovery Program and other National Institute on Aging (NIA)-supported programs to enable open-science practices and accelerate translational learning. The data, analyses, and tools are shared early in the research cycle without a publication embargo on secondary use. Data is available for general research use according to the following requirements for data access and data attribution (https://adknowledgeportal.org/DataAccess/Instructions).
We used all samples for which RNA-seq, proteomics, and phenotype data are available. These samples represent all subjects with available frozen brain samples at the time of data generation.
All data meeting pre-determined quality control criteria were included for analysis.
For each experiment, we have at least 196 biological replicates and all attempts at replication successfully support the conclusions of this study.
There is no randomization. ROS and MAP are community studies. We enroll everyone who doesn't have dementia at the time of enrollment and agrees to annual detailed clinical and cognitive evaluation. All participants with an available sample were used in the analyses.
The team members generating the data from the biological samples were blinded to the phenotypic data of each participant. The data analyst is required to access to the molecular and phenotypic data in order to extract information necessary for the planned analysis, execute quality control of the data and preprocessing, integrate the data with other data resources, and conduct statistical tests.
Reporting for specific materials, systems and methods We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material, system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response. The ROS study is comprised of older catholic priests, nuns, and monks throughout the USA. The MAP study recruits older lay persons from the greater Chicago area. Participants are not compensated for their participations. All visits and data collection other than optional biennial MRI are done as home visits to ensure convenience for the participant and data close to death.
The ROS and MAP studies were each approved by an Institutional Review Board (IRB) of Rush University Medical Center. Both studies enroll older persons without known dementia. All participants agree to an annual detailed clinical evaluation and organ donation at the time of death. Prior to enrollment, each participant signed an informed consent and an Anatomical Gift Act for (AGA) donation of brain, spinal cord, nerve, and muscle to the investigators for research purposes. The AGA is recognized in all 50 states and the District of Columbia. It is an advanced directive that foregoes the need to obtain consent for autopsy from a next of kin at the time of death. Participants also sign a repository consent to allow their data and biospecimens to be shared in accordance with procedures established by the relevant IRB.
Ex vivo brain MRI At autopsy brains were hemisected in accordance with standard protocol and the cerebral hemisphere with more visible pathology was immersed in 4% paraformaldehyde solution and refrigerated at 4°C. At approximately one-month postmortem the hemisphere from each decedent was imaged using a 3-Tesla MRI scanner after allowing the tissue to warm to room temperature.
This study does not contain behavioral performance measures.