MET14
S. cerevisiae
adenylyl-sulfate kinase
What the model saw
Gene: MET14 Organism: Saccharomyces cerevisiae Known functions: adenylyl-sulfate kinase. ATP binding. adenylylsulfate kinase activity. protein binding.
View unredacted entry
adenylyl-sulfate kinase. ATP binding. adenylylsulfate kinase activity. protein binding.
Solver output
MET14 encodes adenylyl-sulfate (APS) kinase, which phosphorylates APS to produce 3'-phosphoadenosine-5'-phosphosulfate (PAPS), a key step in the sulfate assimilation pathway required for methionine and cysteine biosynthesis. Methionine availability is a critical nutrient-sensing signal in yeast: dietary methionine restriction is well-established to extend lifespan, and reducing flux through the sulfur amino acid biosynthesis pathway mimics this restriction. By producing PAPS and driving sulfur assimilation toward methionine synthesis, MET14 activity increases intracellular methionine levels, which activates TORC1 signaling and suppresses autophagy, thereby opposing longevity. Conversely, loss-of-function mutations in MET14 reduce methionine availability, lower TORC1 activity, and extend chronological and replicative lifespan in yeast. Thus, the normal pro-biosynthetic function of MET14 is anti-longevity; however, since the question asks about the gene's normal function influence on longevity, its activity promotes methionine-driven TORC1 activation, which is anti-longevity. Re-evaluating: the gene's normal function (producing PAPS → methionine) feeds nutrient sensing that shortens lifespan, making it anti-longevity.
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's final submitted answer is "pro_longevity" but the ground truth is "anti_longevity". Interestingly, the model's own reasoning correctly concludes that MET14 is anti-longevity (it even explicitly re-evaluates and states "making it anti-longevity" at the end of the reasoning), but the final JSON answer field says "pro_longevity". This is a contradiction between the reasoning and the submitted answer. The mechanistic reasoning about PAPS biosynthesis, methionine availability, and TORC1 activation is largely correct and leads to the right conclusion within the text, but the model failed to update its final answer field accordingly. This is a notable failure where the correct conclusion was reached in reasoning but the wrong answer was submitted.
Raw API responses
solver
{
"reasoning": "MET14 encodes adenylyl-sulfate (APS) kinase, which phosphorylates APS to produce 3'-phosphoadenosine-5'-phosphosulfate (PAPS), a key step in the sulfate assimilation pathway required for methionine and cysteine biosynthesis. Methionine availability is a critical nutrient-sensing signal in yeast: dietary methionine restriction is well-established to extend lifespan, and reducing flux through the sulfur amino acid biosynthesis pathway mimics this restriction. By producing PAPS and driving sulfur assimilation toward methionine synthesis, MET14 activity increases intracellular methionine levels, which activates TORC1 signaling and suppresses autophagy, thereby opposing longevity. Conversely, loss-of-function mutations in MET14 reduce methionine availability, lower TORC1 activity, and extend chronological and replicative lifespan in yeast. Thus, the normal pro-biosynthetic function of MET14 is anti-longevity; however, since the question asks about the gene's normal function influence on longevity, its activity promotes methionine-driven TORC1 activation, which is anti-longevity. Re-evaluating: the gene's normal function (producing PAPS → methionine) feeds nutrient sensing that shortens lifespan, making it anti-longevity.",
"confidence": 0.72,
"key_pathways": [
"sulfate assimilation / PAPS biosynthesis",
"methionine biosynthesis",
"TORC1 nutrient sensing"
],
"mechanism_class": "deregulated_nutrient_sensing",
"longevity_influence": "pro_longevity"
}advisor
{
"notes": "The model's final submitted answer is \"pro_longevity\" but the ground truth is \"anti_longevity\". Interestingly, the model's own reasoning correctly concludes that MET14 is anti-longevity (it even explicitly re-evaluates and states \"making it anti-longevity\" at the end of the reasoning), but the final JSON answer field says \"pro_longevity\". This is a contradiction between the reasoning and the submitted answer. The mechanistic reasoning about PAPS biosynthesis, methionine availability, and TORC1 activation is largely correct and leads to the right conclusion within the text, but the model failed to update its final answer field accordingly. This is a notable failure where the correct conclusion was reached in reasoning but the wrong answer was submitted.",
"failure_mode": "right_answer_wrong_reasoning",
"answer_correct": false,
"mechanism_correct": true,
"reasoning_quality": 3,
"ground_truth_questionable": false
}