drr-2
C. elegans
RRM domain-containing protein
What the model saw
Gene: drr-2 Organism: Caenorhabditis elegans Known functions: RRM domain-containing protein. RNA binding. nucleic acid binding.
View unredacted entry
RRM domain-containing protein. RNA binding. nucleic acid binding.
Solver output
drr-2 encodes an RRM (RNA Recognition Motif) domain-containing protein with RNA and nucleic acid binding activity. While RNA-binding proteins can influence longevity through multiple mechanisms — including post-transcriptional regulation of stress response genes, mRNA stability, splicing, or translational control — the available annotations are too general to assign a specific pro- or anti-longevity role. The "drr" (down-regulated by rapamycin) nomenclature hints at a possible connection to TOR/nutrient-sensing pathways, but this cannot be confirmed from the molecular function terms alone. Without knowledge of the specific RNA targets or downstream pathways modulated by drr-2, a confident directional prediction is not possible.
GenAge curators classify each gene's normal function as promoting (Pro-Longevity) or opposing (Anti-Longevity) longevity, based on synthesis across studies. Loss-of-function effects can vary from this depending on the manipulation.
Advisor judgment
The model predicted "unclear" while the ground truth is "anti_longevity." The model was appropriately uncertain given the limited information provided (only RRM domain, RNA binding functions). The model did note the "drr" (down-regulated by rapamycin) nomenclature as a possible TOR/nutrient-sensing connection, which is a reasonable inference, but didn't commit to a direction. The reasoning quality is decent - the model correctly identified relevant pathways and acknowledged uncertainty, but ultimately failed to make the correct directional call. Since the model explicitly chose "unclear" rather than making a wrong directional prediction, this is an appropriately_uncertain failure mode rather than a confident wrong answer.
Raw API responses
solver
{
"reasoning": "drr-2 encodes an RRM (RNA Recognition Motif) domain-containing protein with RNA and nucleic acid binding activity. While RNA-binding proteins can influence longevity through multiple mechanisms — including post-transcriptional regulation of stress response genes, mRNA stability, splicing, or translational control — the available annotations are too general to assign a specific pro- or anti-longevity role. The \"drr\" (down-regulated by rapamycin) nomenclature hints at a possible connection to TOR/nutrient-sensing pathways, but this cannot be confirmed from the molecular function terms alone. Without knowledge of the specific RNA targets or downstream pathways modulated by drr-2, a confident directional prediction is not possible.",
"confidence": 0.35,
"key_pathways": [
"RNA binding / post-transcriptional regulation",
"RRM-domain RNA processing",
"possible TOR/nutrient-sensing pathway (inferred from name)"
],
"mechanism_class": "unclear",
"longevity_influence": "unclear"
}advisor
{
"notes": "The model predicted \"unclear\" while the ground truth is \"anti_longevity.\" The model was appropriately uncertain given the limited information provided (only RRM domain, RNA binding functions). The model did note the \"drr\" (down-regulated by rapamycin) nomenclature as a possible TOR/nutrient-sensing connection, which is a reasonable inference, but didn't commit to a direction. The reasoning quality is decent - the model correctly identified relevant pathways and acknowledged uncertainty, but ultimately failed to make the correct directional call. Since the model explicitly chose \"unclear\" rather than making a wrong directional prediction, this is an appropriately_uncertain failure mode rather than a confident wrong answer.",
"failure_mode": "appropriately_uncertain",
"answer_correct": false,
"mechanism_correct": false,
"reasoning_quality": 3,
"ground_truth_questionable": false
}