Unlocking the Regulatory Genome by ARGUS: An Evidence-Constrained Agentic Framework for Interpreting Single Nucleotide Variants
Abstract
Over 90% of disease-associated variants from genome-wide association studies fall in noncoding regulatory regions, yet their functional interpretation remains a central open problem in genomic medicine. Large language models prompted to interpret such variants routinely hallucinate transcription factor (TF) binding changes, fabricate experimental support, and assign biological significance to statistically negligible signals. We present ARGUS (Agentic Regulatory Genomics for an Uncertainty-aware Scientist), which strictly separates deterministic biological computation from LLM-mediated reasoning. ARGUS wraps 458 DNABERT-based TF binding models in a hypothesis-directed investigation loop where a planner selects evidence sources based on current uncertainty, a verifier deterministically interprets each observation, and intermediate results change the investigation path. On variant rs6983267 at the 8q24 cancer risk locus, the same planner produces four divergent trajectories for four TFs. FOXA1 is rescued in 3 steps when real ADASTRA allele-specific binding data (15 experiments, FDR = 0.030) reveals a model false negative masked by saturation. KLF6 traverses 8 steps across ADASTRA, JASPAR motif analysis, and ENCODE cCRE regulatory annotation before abstaining due to mixed indirect evidence. RAD21 abstains in 8 steps after ADASTRA returns a coverage-qualified but nonsignificant allelic test (5 experiments, FDR = 0.65), and SP1, which shares FOXA1's saturated retained prediction, abstains because no direct experimental evidence exists at this locus. All observations come from real ADASTRA, JASPAR, and ENCODE cCRE queries; none are simulated. A comparison of fixed-priority and LLM-mediated planning shows that the LLM planner reaches identical verdicts with fewer tool calls by declining evidence that cannot resolve the claim under test.
Related
- Unlocking associative nitrogen fixation in Poaceae: molecular mechanisms, regulatory constraints, and pathways to sustainable productivity
- VarNexus: an integrative database for interpreting disease-associated regulatory variants through multimodal predictions and functional evidence
- Unlocking the three-node regulatory axis: polyamines, antioxidants, and photosynthetic protection in stressed fruit crops
- Sensory variation and behavioural degeneracy: a framework for interpreting heterogeneity in the gut-brain axis
- Bridging the gap between omics and structural data: A framework for interpreting protein-RNA interaction specificity
BibTeX
@misc{dutta2026unlocking,
title = {{Unlocking the Regulatory Genome by ARGUS: An Evidence-Constrained Agentic Framework for Interpreting Single Nucleotide Variants}},
author = {Pratik Dutta and Matthew B. Obusan and Max Chao and Rekha Sathian and Nimisha Papineni and Ramana V. Davuluri},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.12281},
eprint = {2610.12281},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.12281},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Unlocking the Regulatory Genome by ARGUS: An Evidence-Constrained Agentic Framework for Interpreting Single Nucleotide Variants
AU - Pratik Dutta
AU - Matthew B. Obusan
AU - Max Chao
AU - Rekha Sathian
AU - Nimisha Papineni
AU - Ramana V. Davuluri
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.12281
UR - https://arxiv.org/abs/2610.12281
AB - Over 90% of disease-associated variants from genome-wide association studies fall in noncoding regulatory regions, yet their functional interpretation remains a central open problem in genomic medicine. Large language models prompted to interpret such variants routinely hallucinate transcription factor (TF) binding changes, fabricate experimental support, and assign biological significance to statistically negligible signals. We present ARGUS (Agentic Regulatory Genomics for an Uncertainty-aware Scientist), which strictly separates deterministic biological computation from LLM-mediated reasoning. ARGUS wraps 458 DNABERT-based TF binding models in a hypothesis-directed investigation loop where a planner selects evidence sources based on current uncertainty, a verifier deterministically interprets each observation, and intermediate results change the investigation path. On variant rs6983267 at the 8q24 cancer risk locus, the same planner produces four divergent trajectories for four TFs. FOXA1 is rescued in 3 steps when real ADASTRA allele-specific binding data (15 experiments, FDR = 0.030) reveals a model false negative masked by saturation. KLF6 traverses 8 steps across ADASTRA, JASPAR motif analysis, and ENCODE cCRE regulatory annotation before abstaining due to mixed indirect evidence. RAD21 abstains in 8 steps after ADASTRA returns a coverage-qualified but nonsignificant allelic test (5 experiments, FDR = 0.65), and SP1, which shares FOXA1's saturated retained prediction, abstains because no direct experimental evidence exists at this locus. All observations come from real ADASTRA, JASPAR, and ENCODE cCRE queries; none are simulated. A comparison of fixed-priority and LLM-mediated planning shows that the LLM planner reaches identical verdicts with fewer tool calls by declining evidence that cannot resolve the claim under test.
ER -