Credit: Unsplash/CC0 Public Domain
Research scientist Yosuke Tanigawa and Professor Manolis Kellis at MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a novel methodology in human genetics to address an often-overlooked problem: how to handle clinical measurements that fall “below the limit of quantification” (BLQ).
Published in the American Journal of Human Genetics, their new approach, “hypometric genetics,” utilizes these typically discarded measurements to enhance genetic discovery, with significant implications for personalized genomic medicine and drug development.
Imagine trying to weigh a feather on a standard bathroom scale. The scale might detect that something is there. However, it will not give an exact measurement value because the feather is too light to be weighed accurately, falling below the scale’s sensitivity.
In scientific research, measurements that fall below the calibrated range for reliable quantification of a measurement device are marked as BLQ. Traditionally, scientists often discard these data points to ensure that their analysis is not affected by unreliable information. However, the new approach shows that such quality control information can still provide valuable genetic insights.
“We often treat below-quantification-limit data points as missing or uninformative, but our approach shows that these flags hold critical information for genetic analysis,” says Kellis, senior and co-corresponding author of the study.
“By leveraging both the flags and standard quantitative traits, we not only rescue discarded data but also improve our ability to discover genetic associations, revealing key biological insights that would otherwise be missed.”
The researchers analyzed more than 220,000 de-identified participants in UK Biobank and found that genetic factors influence the presence or absence of BLQ flags for specific lipid molecules in the blood.
By integrating information from BLQ flags with conventional quantitative data, they identified genetic associations with 2.8 times as many candidate genes as the standard approach, benefiting especially from rare genetic variants with large effects. This work lays a foundation for discovering new trait-related genes and could accelerate therapeutic target discovery.
“Our approach builds on the previous work focusing on individuals with extreme trait values,” says Tanigawa, the lead and co-corresponding author of the study. “Instead of recruiting individuals with extreme trait values, we used BLQ—as an indicator of extremely low measurement values—and demonstrated that it offers valuable information for genetic discovery.”
The study has attracted substantial attention. “The work by Tanigawa resolved an overlooked but important issue of how to assess the ‘missing’ clinical values in human genomics,” says Yukinori Okada, professor of genome informatics at the Graduate School of Medicine at the University of Tokyo.
“By showing how binarized BLQ successfully empowers genotype-phenotype associations, their work should be valuable for implementing personalized genomic medicine.”
Tanigawa envisions that this technique could aid pharmaceutical companies in translating genetic findings to prioritize therapeutic targets more effectively.
“Human genetic evidence increases the success rate of therapeutic development, but genetic discovery typically requires a large number of individuals,” he explains. “Our results indicate that tapping into typically discarded information can maximize findings, making it a cost-effective strategy.”
Tanigawa also notes that he and his colleagues intend to apply the methodology to other biological data types, such as proteomics and RNA expression, potentially leading to even more genetic insights.
More information:
Yosuke Tanigawa et al, Hypometric genetics: Improved power in genetic discovery by incorporating quality control flags, The American Journal of Human Genetics (2024). DOI: 10.1016/j.ajhg.2024.09.008
Provided by
Massachusetts Institute of Technology
Citation:
Novel method goes ‘below the limit of quantification’ to enhance genetic discovery (2024, November 8)
retrieved 8 November 2024
from https://medicalxpress.com/news/2024-11-method-limit-quantification-genetic-discovery.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
