cis-pQTL analysis in the FastBio study

• Link to preprint in BioRxiv: https://www.biorxiv.org/content/10.1101/2025.07.21.665884v1
• Link to summary statistics produced in this study: https://figshare.com/s/4f0d95978fdeb55f1ca5

Graphical abstract

Contents

Files:

• 01_olink.1536.qc.R script containing all the necessary steps for processing and normalizing the raw proteomic data (NPX values) from Olink 1536 Explore.
• 02_genotype.pc.estimation.sh shell script used to estimate the genotype-derived principal components using plinkv2. These were used as covariates in the main cis-pQTL analysis.
• 03_covariate.selection.R script used to select the covariates to adjust for in the main state-state cis-pQTL analysis. These include both fixed covariates, such as age, sex, BMI and medication_use, as well as latent factors derived from PC analysis of protein and genotype data.
• 04_tensor.cis.qtl.mapping.sh shell script used to perform the following tasks using TensorQTL: (a) Perform cis-pQTL mapping with permutations and default parameters, (b) full summary statistics per protein tested, (c) independent cis-pQTL mapping using default parameters.
• 05_vep.annotation.indep.cis.qtls.R script to explore the annotations of all independent cis-pQTLs across contexts. The annotations were retrieved using the VEP online tool (https://www.ensembl.org/Tools/VEP)
• 06_top.cis.qtl.replication.R script used to explore the replication of top cis-pQTLs against the cis-summary statistics of the latest UKBB-PPP publication. Data were freely accessed using the following link https://www.synapse.org/Synapse:syn51365303.
• 07_mashr.cis.qtl.mapping.R script used to jointly analyze the results of steady-state cis-pQTLs from all contexts and decipher which SNP-protein pairs exhibit differential effects between contexts.
• 08_colocalization.analysis.R script used to (a) perform phewas queries of differential cis-pQTLs against OpenGWAS catalog and (b) based on the associated traits run colocalization per context.
• regional.plot.function.R this function was used to produce stacked regional plots based on the colocalization results across contexts.
• dosage.effect.plot.function.R this function was used to produce dosage effect plots for all the differential cis-pQTL for each context.

Software and dependencies:

R version 4.4.1

• ieugwasr_1.0.1
• coloc_5.2.3
• flashier_1.0.53
• magrittr_2.0.3
• ebnm_1.1-27
• mashr_0.2.79
• ashr_2.2-63
• patchwork_1.3.0
• ggpubr_0.6.0
• locuszoomr_0.3.5
• lubridate_1.9.3
• forcats_1.0.0
• stringr_1.5.1
• dplyr_1.1.4
• purrr_1.0.2
• readr_2.1.5
• tidyr_1.3.1
• tibble_3.2.1
• ggplot2_3.5.1
• tidyverse_2.0.0
• data.table_1.16.2

External software

• PLINK v2.00a3LM
• Python 3.8.10
• tensorqtl 1.0.8