NEW YORK – A team from the Weizmann Institute of Science and other centers in Israel has untangled genetic contributors to complex phenotypes with the help of detailed physiological and molecular data for thousands of participants in a longitudinal population study in Israel.
"Our results showed novel genotype-phenotype relationships that may lead to improved understanding of human health and disease," senior and corresponding author Eran Segal, a computer scientist and applied mathematician at the Weizmann Institute, and his colleagues wrote in the journal Med on Thursday, noting that the findings have been shared online "for download and interactive exploration."
With multiomic data for 8,706 healthy adult participants in the longitudinal Human Phenotype Project, the researchers performed genome-wide association studies focused on 729 clinical features and more than 4,043 molecular profiles, together encompassing a wide range of phenotypes related to everything from body composition or sleep quality to blood serum metabolites or gut microbiome measurements.
The Human Phenotype Project "is centered around the cohort based on voluntary self-assignment and a screening survey and aims to identify novel diagnostic biomarkers and targets for disease," the authors explained. "Over the course of five years, the study has collected a wide range of clinical data and molecular data, and it aims to uncover markers for disease and treatment targets."
Based on more than 8.9 million SNPs profiled in the participants, they flagged nearly 1,200 SNPs tracking with the clinical traits studied, while highlighting 16,047 ties to clinical traits in subsequent polygenic risk score (PRS)-informed phenome-wide association studies (PheWAS).
While PheWAS in general search for "associations between many phenotypes and a single genetic variant or signal," the authors explained, "the proportion of [human trait] variance explained by individual genetic variants is small."
Consequently, they turned to a PRS-PheWAS strategy to search for phenotypes with ties to sets of genetic risk variants found in PRS established for various human traits or conditions using data from the UK Biobank.
With their combination of GWAS and PRS-PheWAS analyses, the investigators unearthed genetic risk variants that tracked with those identified in prior studies, including many genetic contributors to serum metabolomic features.
The investigators noted that the standard deviation in mean glucose levels measured at a given time point each day — a measurement used to assess individuals' within-day glycemic variance — was particularly prone to PRS associations in their PRS-PheWAS, for example.
While the team's analyses revealed ties between diabetes-related genetic risk variants and glycemic variance from one day to the next, within-day glycemic variance appeared to be inversely related to genetic variants implicated in obesity.
Such results prompted the authors to suggest that the glycemic variance associations "could serve as a foundation for further studies on the role played by glycemic control mechanisms in human health."
Beyond that, the team started digging into potential relationships between the complex traits they considered in their study and the microbial composition and output of individuals' gut microbiomes.
Although findings from those analyses were "largely negative," the authors explained, "the role of host genetics in determining gut microbiome composition is currently poorly understood." In particular, they speculated that person-to-person variability in gut microbiome composition, coupled with a lack of statistical power, may account for the lack of robust reproducibility in gut microbiome-genetic findings.
The investigators also called for additional replication studies in larger cohorts and in individuals from other populations to shore up the GWAS and PRS-PheWAS findings reported, since the HPP group mainly contains individuals of Ashkenazi Jewish ancestry. Still, they suggested that the current findings are expected to "facilitate further investigations into the genetic determinants of human health and disease."