A small but meaningful group of celiac patients test negative for the genes doctors normally look for — yet they still have the disease. New research published in the journal HLA helps explain why: computer modeling shows that two lesser-studied genetic variants can trigger the same immune attack on the gut through pathways that current screening largely overlooks.
For celiac families, this matters. My son has celiac disease, and we were fortunate — his diagnosis followed a relatively direct path. But plenty of celiac families hit dead ends, and some of those dead ends come down to genetics that standard tests weren’t built to catch.
What This Means for You
Most people with celiac disease carry one of two genetic variants: HLA-DQ2.5 or HLA-DQ8. These are what standard celiac genetic tests screen for. But a clinically meaningful subset of celiac patients — people who meet every diagnostic criterion — carry neither variant. For them, and for their doctors, this has long been a puzzle.
This study suggests the answer lies with two other variants, HLA-DQ5 and HLA-DQ9. Using computer simulations, researchers found that both can bind to the same gluten protein fragments that drive celiac disease — through their own distinct mechanisms. The immune machinery needed to attack the gut when gluten is present can be assembled from more than one set of genetic parts.
The practical implications are still down the road — this is computational research, not a clinical trial. But the results chip away at a real problem: patients who fit every symptom profile for celiac but are told their genetics “don’t match.” Understanding which other variants might be involved could eventually improve testing and reduce missed diagnoses.
For celiac families right now, the takeaway is this: a negative genetic test for DQ2.5 and DQ8 does not rule out celiac disease. If symptoms persist and antibody tests or biopsy results still point toward celiac, genetic results should inform the conversation with a gastroenterologist — not end it.
Key Takeaways
- Standard celiac genetic tests screen for HLA-DQ2.5 and HLA-DQ8, but those aren’t the only genes that can drive the disease.
- Computer modeling shows HLA-DQ9 can trigger celiac through a pathway that overlaps with DQ8.
- HLA-DQ5 works differently — it presents a distinct set of gluten fragments that the classic celiac genes largely miss.
- Gamma-gliadin, a component of gluten, emerged as the strongest immune trigger across all four genetic variants studied.
- A negative result on standard celiac genetic testing does not rule out celiac disease — clinical judgment remains essential.
The Science
Want to understand how this actually works? We’ll walk you through the technical details below and define every term. No medical degree required.
What HLA Genes Do
HLA (Human Leukocyte Antigen) genes code for proteins on the surface of immune cells. Think of these proteins as molecular trays — they grab fragments of foreign material, including pieces of gluten, and hold them up for inspection by T-cells, the immune system’s frontline responders.
In celiac disease, trays built from HLA-DQ2.5 or HLA-DQ8 happen to fit gluten fragments particularly well. When they present those fragments to T-cells, the T-cells flag them as threats and launch an immune response that damages the small intestinal lining. This is why DQ2.5 and DQ8 are the classic celiac risk genes — people who carry them have trays that are, unfortunately, well-suited to picking up gluten.
The Non-Canonical Gap
What about celiac patients who don’t have DQ2.5 or DQ8? Their immune cells carry different trays — built from other HLA variants including DQ5 or DQ9. These non-canonical (outside the standard model) variants have been poorly understood because most celiac research has centered on the classic risk genes.
An Italian research team at Fondazione IRCCS San Gerardo dei Tintori tackled this gap by building a computational model of the entire process — from gluten entering the gut to fragments being presented to T-cells.
How the Study Was Built
The researchers started with peptides (short protein chains) derived from three families of gliadin — the immune-triggering fraction of gluten: α-gliadin (alpha), γ-gliadin (gamma), and ω-gliadin (omega). These fragments are what immune cells actually encounter and evaluate.
They then ran those fragments through a simulated digestive process. First, they applied tissue transglutaminase (tTG) — an enzyme in the intestinal wall that chemically modifies gluten fragments, making them more recognizable to immune cells. Then they modeled the effects of stomach and intestinal enzymes breaking everything down further.
The resulting library of peptide fragments was tested computationally against all four HLA-DQ variants using NetMHCIIpan 4.3, a validated tool for predicting how strongly a peptide will bind to a given HLA molecule. Each peptide received a composite score the researchers called the Biological Plausibility score — a combined measure of binding strength, digestive stability, and how abundant the peptide is in gluten proteins.
Two Distinct Patterns
The modeling produced two distinct findings.
HLA-DQ9 overlaps with HLA-DQ8. When the researchers grouped variants by which peptides they favored — a technique called hierarchical clustering — DQ9 and DQ8 landed in the same group. DQ9 appears to grab many of the same gluten fragments that DQ8 does, meaning it can sustain the same immune-triggering pathway. The researchers call this convergent antigen presentation: a different genetic starting point arriving at the same dangerous outcome.
HLA-DQ5 works in the opposite direction. Rather than overlapping with DQ2.5 or DQ8, DQ5 presented a distinct set of gliadin peptides — fragments the classic risk genes largely ignore. This is complementary antigen presentation: DQ5 expands the range of gluten fragments that can trigger an immune response, rather than duplicating the classic pathway.
Gamma-gliadin dominated across all four variants. Regardless of which HLA molecule held the tray, γ-gliadin fragments were the most consistently recognized. The researchers attribute this to two properties: gamma-gliadin resists digestion better than other gliadin types (more of it survives intact to reach immune cells), and it shows high allelic promiscuity — meaning its peptides bind effectively to multiple different HLA variants, not just one.
Why Computational Modeling?
This study is entirely in silico — conducted through computer simulation rather than laboratory experiments on tissue or live patients. That’s worth naming clearly.
Computational approaches are powerful for hypothesis generation at scale. The researchers could test thousands of peptide-HLA binding combinations in a way that would be impractical in a wet lab. But predicted binding affinities need experimental validation before they translate into clinical tools. The authors are explicit about this — they frame the work as a mechanistic rationale for future laboratory and patient-based studies.
We’ve previously covered the genetic complexity of celiac disease and how antibody genes shape immune responses in celiac patients. This study fits into that growing picture: celiac disease is not a single genetic equation, and the immune system has more ways to react to gluten than the classic DQ2.5/DQ8 model captures.
What Could Come Next
The authors call for experimental validation — lab studies using actual T-cells and HLA-typed tissue samples — to confirm whether the binding patterns the model predicts translate into real immune responses. If they do, the results could support expanded genetic panels that include DQ5 and DQ9 as celiac risk factors. Over time, that kind of testing could reduce the number of patients who are told, incorrectly, that their genetics rule out celiac disease.
This is long-horizon work. But for anyone who has watched a celiac patient cycle through inconclusive tests and conflicting results, understanding that the immune system has multiple routes to the same destructive endpoint is both scientifically important and, frankly, a relief to see being studied.
Related Coverage
- Celiac Disease: A Comprehensive Review of Epidemiology, Pathogenesis, and Therapeutic Strategies
- Gene-Environment Interactions in Celiac Disease: The Role of the Intestinal Epithelium
- New Research Reveals How Antibody Genes May Shape the Immune Response in Celiac Disease
References
- Milano A, Crocchiolo R, Anzaldi N, et al. In silico analysis of functional overlap and complementarity of HLA-DQ5 and DQ9 in gliadin presentation: expanding the immunopeptidomic landscape of celiac disease. HLA. 2026 Jul;108(1):e70820. doi:10.1111/tan.70820. PubMed