BackgroundRare genetic variants from single-nucleotide polymorphisms can be a source of diseases. Analyzing their associations with metabolites can gain a better insight in their relationship, which can lead to improved diagnostics and the development of medications.

ObjectivesThe aim of this project is to evaluate associations between rare genetic variants and metabolite levels, as well as correlations between metabolite levels and clinical measurements. Additionally, the aim is to investigate how metabolite levels and rare genetic variants can be associated with diseases.

MethodologyWhole-genome sequencing, metabolomic profiles, and clinical measurements for 101 individuals involved in The Swedish SciLifeLab SCAPIS Wellness Profiling (S3WP) study were analyzed in this project. An association analysis between rare genetic variants and metabolite levels has been performed using the SKAT package in R and visualized with Manhattan plots. A correlation analysis between metabolite levels and clinical measurements was carried out; then the results from both the analyzes formed the basis for a literature study.

ResultsThe association analysis uncovered 31589 associations, of which 1082 associations under the Bonferroni threshold (2.59 ∙ 10-9), involving 71 metabolites and 500 genes. The ten genes with associations with the lowest p-values were NFU1, AGXT, GID4, MAB21L4 and BMS1P15 to glucose, RGS10, ART4 and TMEM254 to salicylic acid and STRIP2 and STK32A-AS1 to pyroglutamic acid. The five metabolites displaying associations with the lowest p–values (glucose, creatinine, pyroglutamic acid, pipecolic acid and salicylic acid) showed correlations with several clinical measurements. A few of the correlations between the metabolite levels and the clinical measurements have been reported earlier, whilst some were new or contradicting to existing literature.

ConclusionsAssociations between rare genetic variants and metabolite levels, and between metabolite levels and clinical measurements were found, but it is not certain if the metabolite levels are affected mostly by the rare genetic variants or the clinical measurements. Furthermore, several associations between genes and metabolites that can be linked to specific diseases were identified.