ProteomicsOmics dataKnight-ADRCparietal cortex
Knight ADRC brain proteomics
Get the data
Controlled access. Request data access for Knight-ADRC on the AD Knowledge Portal; you need a free account there. See How to get the data.
| Where | ID | Access | What you get |
|---|---|---|---|
| AD Knowledge Portal | syn69670592 | Controlled | Files described below |
pip install synapseclient
synapse login
synapse get -r syn69670592QTL results computed from this data, and their files on the AD Knowledge Portal, are listed on Knight ADRC brain pQTL.
Contact
Zining Qi
- Contact Email: zq2209@cumc.columbia.edu
- Contact Affiliation : Columbia University
- Contact Role : Zining Qi performed QTL analysis by using Knight proteomics
Study Overview
- PI : Carlos Cruchaga (Washington University in St. Louis)
- Grant number : R01AG044546, P01AG003991, RF1AG053303, R01AG058501, U01AG058922, RF1AG058501, R01AG057777, P50AG05681, P01AG03991, P01AG026276
- Publication : PMID: 38997326, PMID: 34239129
- Genetics Publication : PMID: 38997326
- Acknowledgement : This work was supported by grants from the National Institutes of Health (R01AG044546, P01AG003991, RF1AG053303, R01AG058501, U01AG058922, RF1AG058501 and R01AG057777). The recruitment and clinical characterization of research participants at Washington University were supported by NIH P50 AG05681, P01 AG03991, and P01 AG026276. 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.
- Study name : ROSMAP DLPFC proteomics analysis
- Study Description : Performed analysis on 442 Knight proteomics samples via the ADSP FGC xQTL pipeline. The output format is prepared for pQTL analysis.
- Disease : Alzheimer’s Disease
- Data Citation : Omics data: ng00102; Genetics data: ng00127
Dataset Description
Raw data
All brain samples were collected from parietal lobe. Brain tissues (~500mg) were collected from the fresh frozen human parietal lobes. SomaScan 1.3k platform were applied to measure the brain protein. The protein levels were reported as relative units of intensity (RFU or Relative Fluorescence Unit).
Molecular phenotype matrices
Quality Control: Data released had some initial quality control:
- Analyst with SF >= 0.5 and CV >= 0.15 removed
- Use IQR to detect outliers and replace outliers with NA
Normalization: In this notebook, we applied more quality control steps:
- Log2 transformed with median centered near zero.
- This normalization make data have approximate normal distribution and more comparable with ROSMAP proteomics data.
Quantitative proteomics:
The phenotype raw data contains 1296 genes(rows) from 442 samples(columns). The data were processed after sampling, quantification, and quality control with NAs.
Other key data files
Annotation:
It contains gene and protein information of each sample.
Other reference files created via Reference_data_notebook:
Age at death, sex and pmi covariates will be extract from ROSMAP raw data:
Links to omics data analysis notebooks
See notebooks in:
- https://github.com/cumc/fungen-xqtl-analysis/blob/main/analysis/Wang_Columbia/knight/pqtl/ lab access only
The notebooks in this folder contain the commands and data wrangling codes for analysis of the proteomics data in Knight. (data wrangling exist because not all data are processed using the xqtl-pipeline from the begnning and need to be reformated to fit one intermediate step of the pipeline).
Phenotype Preprocessing
Phenotype data were processed via the following workflow: phenotype_preprocessing lab access only
Genotype Preprocessing
genotype_preprocessing.ipynb lab access only provides information about the steps.
Association - TensorQTL
- cis_pqtl_association.ipynb lab access only provides information and result of cis-QTL analysis.
- trans_pqtl_association.ipynb lab access only provides information and result of AD genes trans-QTL analysis.
Other procedures / files
- proteomics_pheno_whole_genome.proteomics_cov.ROSMAP_NIA_WGS.leftnorm.filtered.filtered.prune.pca.resid.PEER.diag.pdf lab access only view of PCA result of the pQTL analysis (steps included in
genotype_preprocessing.ipynb) - AD gene list includes 76 AD genes for trans-QTL analysis.
QTL Analysis
QTL analysis for this dataset is documented in ../../qtl/pQTL/Knight_ADRC_brain_proteomics_qtl.md.
Flagship paper analyses:
- Fine-mapping (SuSiE-RSS): syn69670592
- TWAS weight models: syn70090845 (this dataset’s folder; all models are in syn69670600)
- Quantile TWAS (qTWAS) models: not available yet
- Multi-context colocalization (ColocBoost): syn69670597
- AD GWAS–xQTL colocalization results: syn69865816
- AD GWAS–xQTL colocalization models: syn69670630