In the rapidly evolving landscape of human genetics, the "one variant, one gene" hypothesis has long served as the bedrock of regulatory interpretation. For years, researchers have mapped genetic variants associated with complex diseases to specific regulatory regions, then engaged in rigorous debates over which single gene was the "true" causal driver. However, a groundbreaking study published in The American Journal of Human Genetics (AJHG) by Kate (Kathryn) Lawrence, a PhD candidate at Stanford University, suggests that this fundamental paradigm may be obscuring our understanding of human biology.
In her paper, titled "Focus on single-gene effects limits discovery and interpretation of complex-trait-associated variants," Lawrence argues that by limiting our analytical scope, we are systematically missing the regulatory signals that govern complex traits. By shifting the focus toward multi-gene regulatory mechanisms, Lawrence introduces a new conceptual framework that promises to redefine how we interpret non-coding genetic variation.
The Core Thesis: Challenging the Reductionist View
The primary motivation behind Lawrence’s research stems from a recurring frustration in the field of genome-wide association studies (GWAS). When researchers "fine-map" a variant—narrowing down the search for the specific genetic trigger of a trait—they often find themselves staring at a regulatory element situated between multiple genes.
"We kept running into GWAS loci where the whole ‘one variant, one gene’ story felt incomplete," Lawrence explains. "You’d fine-map to a regulatory region sitting between multiple genes, and everyone would argue about which one was ‘the’ causal gene—but what if that was the wrong question?"
Lawrence notes that existing CRISPR interference (CRISPRi) data has long hinted that regulatory elements often influence several nearby genes simultaneously. Despite this, standard quantitative trait loci (QTL) methods remained tethered to a single-gene testing approach. This disconnect prompted the development of what Lawrence calls pcQTLs (polygenic/poly-cis QTLs). By testing genes as a collective unit rather than in isolation, the research team discovered that many regulatory signals previously overlooked by traditional methods were hiding in plain sight.
Chronology of Discovery: From Observation to Framework
The evolution of this project was a multi-stage endeavor that bridged the gap between computational observation and biological validation.
Phase 1: The Disconnect
The project began as an inquiry into the limitations of existing GWAS interpretations. Lawrence and her colleagues observed that as genomic datasets grew in size and resolution, the "missing heritability" and the ambiguity of non-coding variant functions remained persistent problems. The team hypothesized that the analytical bottleneck was not the data itself, but the statistical framework being applied to it.
Phase 2: Developing the pcQTL Model
Recognizing that regulatory elements often function as hubs, the team moved away from the linear gene-centric model. They began developing statistical frameworks that could account for the regulatory impact of a single variant on a cluster of genes. This transition required moving from simple univariate analysis to a more complex, multivariate approach that acknowledges the interconnected nature of the genome.
Phase 3: Conceptualizing "Allelic Proxitropy"
Perhaps the most significant contribution of the study is the introduction of a new term: "Allelic Proxitropy." This concept provides a formal language for describing variants that influence the regulation of multiple nearby genes. It moves the discussion away from "which gene is causal" toward "what is the aggregate regulatory outcome of this variant."
Supporting Data: Why the Paradigm Shift Matters
The implications of Lawrence’s work are supported by the limitations of traditional colocalization studies. In conventional genetics, when a variant is associated with a disease, researchers look for the gene that is most highly expressed in the relevant tissue. However, this method frequently fails when a single genetic switch controls multiple downstream genes, each contributing a fraction to the overall trait.
The Problem of "The One"
In the current paradigm, if a variant affects Gene A, Gene B, and Gene C, but only Gene A shows a clear, statistically significant signal in a traditional test, the contributions of Genes B and C are discarded as "background noise." Lawrence’s study suggests that this noise is actually the signal. By treating the regulatory unit as a whole, the team was able to map variants that were previously labeled as "functional, but ambiguous."
Statistical Precision
The pcQTL approach allows for a higher degree of statistical precision when dealing with non-coding regions. By accounting for the regulatory landscape’s topology—where enhancers loop and interact with promoters—the researchers were able to capture regulatory signals that had been effectively "averaged out" by older, less sensitive methodologies.
Official Perspective and Academic Implications
The editorial board of The American Journal of Human Genetics selected Lawrence’s paper for a deep-dive interview, underscoring the significance of the findings for the broader scientific community.
A Call for Nuance
The primary implication for the human genetics community is a mandate for increased nuance. Lawrence advocates for the adoption of multi-gene frameworks, suggesting that the "one variant, one gene" approach is a simplification that no longer serves the complexity of modern genomic data.
"We hope that the use of pcQTLs or other multi-gene frameworks in the human genetics community will lead to a more nuanced and complete view of the impact of genetic regulation," Lawrence stated during her interview.
Redefining Causal Mechanisms
By shifting the focus, the community can move toward a more "systems biology" approach to genetics. Instead of viewing the genome as a list of independent switches, researchers are encouraged to view it as a network of regulatory nodes. This has direct consequences for drug discovery; if a disease is driven by the dysregulation of three genes rather than one, targeting a single protein product may be insufficient. Understanding the regulatory network allows for more sophisticated therapeutic interventions.
Reflections on the Scientific Life
Beyond the data, Lawrence’s work serves as a testament to the importance of academic independence and intellectual flexibility. When asked for advice for trainees and young scientists, she emphasized the importance of forging one’s own path.
"There are as many ways to be a success at science as there are successful scientists," she noted. This sentiment reflects her own journey—from a high-level athlete racing in the Enduro World Series to a leading voice in human genetics at Stanford.
Lawrence’s transition from elite mountain biking to high-stakes genomic research is perhaps a metaphor for her work: both require the ability to navigate complex, changing landscapes while maintaining focus on the ultimate goal. Her dedication to questioning established "molds" has not only advanced the field of genetics but has also provided a blueprint for how young researchers can challenge the status quo to reveal deeper truths about the human blueprint.
Future Outlook
As the field of human genetics moves into the era of large-scale biobanks and single-cell multi-omics, the volume of data will only increase. The "one variant, one gene" model will likely become an increasingly fragile bridge for this data. Lawrence’s work provides a necessary expansion of the theoretical toolbox.
In the coming years, we can expect the adoption of pcQTLs and similar multi-gene regulatory models to become standard practice in GWAS interpretation. As these methods mature, the ability to predict the downstream effects of non-coding variants will improve, potentially unlocking new insights into the genetic architecture of everything from autoimmune disorders to neurodegenerative diseases.
By daring to ask "what if we are testing the wrong unit?", Kate Lawrence has not only provided a new statistical tool—she has challenged the scientific community to see the genome not as a series of isolated parts, but as a deeply connected, multifaceted system. The journey to decode the human genome is far from over, but thanks to this new framework, we are finally looking at the right map.
About the Author
