rs117774819 (PDILT): UK Biobank Biomarker GWAS

Key takeaways

• rs117774819 is a variant in the PDILT gene region analyzed in large-scale UK Biobank genome-wide association studies.
• A study of 363,228 people identified 1,857 loci associated with at least one of 35 blood or urine biomarkers, covering protein-altering variants, HLA allelotypes, and copy-number variants.
• Mendelian Randomization in the primary source study revealed 51 causal relationships between biomarkers and diseases, including urate affecting gout risk and cystatin C affecting stroke risk.
• No lifestyle-specific findings are currently documented for this variant from the available study data.

What the research says A large-scale analysis of 35 blood and urine biomarkers in 363,228 UK Biobank individuals identified 1,857 associated loci containing 3,374 fine-mapped associations, and Mendelian Randomization (a method for inferring causal relationships using genetic variants as natural experiments) uncovered 51 causal relationships between biomarkers and diseases, including urate increasing gout risk and cystatin C affecting stroke risk (Sinnott-Armstrong N et al., Nature Genetics 2021). A subsequent study applied Quickdraws, a mixed-model genome-wide association method using a spike-and-slab prior on variant effects (a statistical framework that explicitly models most variants as having zero effect while a small proportion carry non-zero effects), to approximately 405,088 UK Biobank individuals across 79 quantitative and 50 binary traits, identifying 4.97% more quantitative trait associations than the REGENIE method and 22.71% more than FastGWA (Loya H et al., Nature Genetics 2025). Specific variant-level findings for rs117774819 at this locus are not detailed in the provided study excerpts.

Reported associations Specific association statistics for rs117774819 are not contained in the provided study excerpts. The following trait categories were covered in the underlying analyses, representing the scope within which this locus's data would appear:

• Blood and urine biomarkers (35-trait panel): 1,857 loci were identified at Bonferroni-corrected (adjusted downward to account for the large number of simultaneous comparisons) p below 5 x 10^-9 in a meta-analysis of 355,891 UK Biobank individuals; heritability explained by common genetic variants ranged from 0.6% for lipoprotein(a) to 23.9% for IGF-1 across continuous traits (Sinnott-Armstrong N et al., Nature Genetics 2021)
• Testosterone: One of the 35 biomarkers analyzed; covariate models explained up to 90% of phenotypic variance for this trait, the highest proportion among all 35 measures in the panel (Sinnott-Armstrong N et al., Nature Genetics 2021)
• Kidney function, liver function, and glycemic markers: Included in the 35-biomarker panel; heritability estimates using HESS (Heritability Estimator from Summary Statistics, a method that uses genome-wide summary data) ranged from 3.2% for microalbumin in urine to 57% for total bilirubin across continuous phenotypes (Sinnott-Armstrong N et al., Nature Genetics 2021)
• Quantitative and binary disease traits (79+50 panel): Analyzed in approximately 405,088 UK Biobank individuals; Quickdraws identified 4.97% more associations than REGENIE and 22.71% more than FastGWA for quantitative traits, and 3.25% and 7.07% more than those methods for binary traits (Loya H et al., Nature Genetics 2025)

Evidence quality The primary contributing studies involve large sample sizes: a meta-analysis of 355,891 UK Biobank individuals across five population groups in the biomarker study (Sinnott-Armstrong N et al., Nature Genetics 2021) and approximately 405,088 individuals in the Quickdraws analysis (Loya H et al., Nature Genetics 2025). Statistical thresholds were stringent, with Bonferroni-corrected p below 5 x 10^-9 for common imputed variants, p below 1 x 10^-6 for non-rare copy-number variants, and Benjamini-Yekutieli correction applied to HLA (human leukocyte antigen, a set of immune-related genes) alleles. Linkage disequilibrium (LD) score intercepts, a diagnostic for whether population ancestry differences are inflating association signals, ranged from 0.999 to 1.137 across the 35 biomarker phenotypes, consistent with well-controlled population structure (Sinnott-Armstrong N et al., Nature Genetics 2021). Replication of biomarker associations was assessed in FinnGen (n=135,500) and Biobank Japan, with multi-biomarker polygenic risk scores outperforming single-disease scores for chronic kidney disease, type 2 diabetes, gout, and alcoholic cirrhosis in independent datasets. Specific p-values, effect sizes, and replication status for rs117774819 are not available in the provided study excerpts, which limits the variant-specific evidence summary.

Lifestyle considerations No lifestyle considerations on file for this variant.