python main.py
[Warning] ToolUniverse not available: No module named 'tooluniverse.tool_registry'
[Warning] Install tooluniverse to use Medea tools: pip install tooluniverse
/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/tool_space/gpt_utils.py:8: FutureWarning:
All support for the `google.generativeai` package has ended. It will no longer be receiving
updates or bug fixes. Please switch to the `google.genai` package as soon as possible.
See README for more details:
https://github.com/google-gemini/deprecated-generative-ai-python/blob/main/README.md
import google.generativeai as genai
[INFO] Loading FlagEmbedding library (may download reranker models)...
[INFO] FlagEmbedding loaded successfully
[INFO] Loading FlagEmbedding library (may download models on first run)...
[INFO] FlagEmbedding loaded successfully
================================================================================
MEDEA EVALUATION CONFIGURATION
================================================================================
Setting: medea
Task: targetID
Dataset Seed: 42
LLM Backbone Seed: 42
Temperature: 0.4
Evaluation Folder: ./evaluation
--------------------------------------------------------------------------------
Disease: ra
SCFM: PINNACLE
--------------------------------------------------------------------------------
LLM Backbone: gemini-3-flash-preview
LLM Paraphraser: gpt-4o
LLM Judge: gpt-4o
Quality Max Iter: 2
Debate Rounds: 2
Panelists: ['gemini-2.5-flash', 'o3-mini-0131', 'gemini-3-flash-preview']
================================================================================
Loading evaluation data from: ./evaluation/targetID/evaluation_samples/targetid-ra-query-42.csv
Query mode: True → Folder: evaluation_samples
Successfully loaded 140 rows
User Template:
Which gene among {candidate_genes} shows the highest cell-type specificity in {celltype} for targeting {disease}?
Agent Template: Identify the verified therapeutic targets for {disease} and retrieve their cell-type–specific embeddings from {scfm}. Determine which of the {cell_type}-activated genes in {scfm} overlap with these targets; this overlapping set constitutes the cell-type–specific targets for {disease}. Compute their average embedding as a reference, then calculate the cosine similarity between this reference and each individual target. If one gene shows a markedly higher similarity than the rest, nominate it as the top candidate and supply a concise, evidence-based rationale for its biological relevance.
=== LLM Provider: OpenRouter ===
=== Init Research Planning Agent Backbone ===
Initialized LLM: gemini-3-flash-preview (via OpenRouter)
=== Init Analysis Agent Backbone ===
Initialized LLM: gemini-3-flash-preview (via OpenRouter)
=== Init Literature Reasoning Agent Backbone ===
Initialized LLM: gemini-3-flash-preview (via OpenRouter)
=== Init Research Planning Actions ===
[Warning] ToolUniverse not available: No module named 'tooluniverse.tool_registry'
[Warning] Install tooluniverse to use Medea tools: pip install tooluniverse
Loaded 7 checker functions: ['celltype_avaliability_checker', 'concept_name_checker', 'context_avalibility_checker', 'disease_name_checker', 'enrichr_gene_name_checker', 'humanbase_context_checker', 'transcriptformer_context_checker']
=== Init Analysis Actions ===
=== Init Literature Reasoning Actions ===
LLM Backbone Seed: 42
[Research Plan] Agent Temp: 0.4
[Research Plan] Action Temp: 0.4
[Research Plan] Quality Iterations: 2
[Analysis] Agent Temp: 0.4
[Analysis] Action Temp: 0.4
[Panel] Panelists: ['gemini-2.5-flash', 'o3-mini-0131', 'gemini-3-flash-preview']
[User Query]: Which gene—GNAT2, TTLL10-AS1, CORO1A, EAF1-AS1, or AKAP17A (ENSG00000197976)—displays the highest cell-type specificity in effector memory CD4⁺ αβ T cells for targeting rheumatoid arthritis?
[MEDEA] Starting parallel execution: Research Planning + In-silico Experiment + Literature Reasoning
[MEDEA] Launching in-silico experiment process...
[MEDEA] Launching literature reasoning process...
[MEDEA] Experiment analysis process started (PID: 153676)
Agent research_plan_agent receives the following TaskPackage:
[
Task ID: 8916a1c0-86d9-441a-aa85-1b47fc6e063b
Instruction: {'user_query': 'Identify the verified therapeutic targets for rheumatoid arthritis using PINNACLE and obtain their corresponding cell-type–specific embeddings. Next, determine which genes from the effector memory CD4+ alpha beta T cell activation signature in PINNACLE overlap with these targets to define a set of cell-type–specific RA targets. Compute the average embedding of this set as a reference point, then calculate the cosine similarity between this reference and each individual gene embedding. If one gene displays a significantly higher similarity than the others, nominate it as the top candidate and provide a brief, evidence-based rationale for its biological relevance.'}
]
====research_plan_agent starts execution on TaskPackage 8916a1c0-86d9-441a-aa85-1b47fc6e063b====
[MEDEA] Literature reasoning process started (PID: 153679)
[MEDEA] Both processes running in parallel...
Agent reasoning_agent receives the following TaskPackage:
[
Task ID: 8916a1c0-86d9-441a-aa85-1b47fc6e063b
Instruction: {'user_query': 'Identify the verified therapeutic targets for rheumatoid arthritis using PINNACLE and obtain their corresponding cell-type–specific embeddings. Next, determine which genes from the effector memory CD4+ alpha beta T cell activation signature in PINNACLE overlap with these targets to define a set of cell-type–specific RA targets. Compute the average embedding of this set as a reference point, then calculate the cosine similarity between this reference and each individual gene embedding. If one gene displays a significantly higher similarity than the others, nominate it as the top candidate and provide a brief, evidence-based rationale for its biological relevance.', 'hypothesis': None}
]
====reasoning_agent starts execution on TaskPackage 8916a1c0-86d9-441a-aa85-1b47fc6e063b====
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent research_plan_agent takes 0-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Agent reasoning_agent takes 0-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
Research plan agent call failed: name 'ACTION_NOT_FOUND_MESS' is not defined
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 1-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 2-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 3-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 4-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 5-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 6-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
[chat_completion] Attempt 3/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] All 3 attempts exhausted.
Agent reasoning_agent takes 7-step Action:
{
name: I cannot help with it - Error: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
params: {}
}
Observation: [Error] Action not found in action list.
[chat_completion] Attempt 1/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 2s...
[chat_completion] Attempt 2/3 failed: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
[chat_completion] Retrying in 3s...
^CProcess Process-4:
Traceback (most recent call last):
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/main.py", line 842, in <module>
medea_unittest(
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/main.py", line 499, in medea_unittest
response = medea(user_instruction, experiment_instruction, research_planning_module, analysis_module, literature_module)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/core.py", line 273, in medea
literature_process.join(timeout=timeout)
File "/usr/lib/python3.10/multiprocessing/process.py", line 149, in join
res = self._popen.wait(timeout)
File "/usr/lib/python3.10/multiprocessing/popen_fork.py", line 40, in wait
if not wait([self.sentinel], timeout):
File "/usr/lib/python3.10/multiprocessing/connection.py", line 931, in wait
ready = selector.select(timeout)
File "/usr/lib/python3.10/selectors.py", line 416, in select
fd_event_list = self._selector.poll(timeout)
KeyboardInterrupt
Traceback (most recent call last):
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/tool_space/gpt_utils.py", line 134, in _openrouter_completion
completion = client.chat.completions.create(**request_params)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/openai/_utils/_utils.py", line 286, in wrapper
return func(*args, **kwargs)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/openai/resources/chat/completions/completions.py", line 1192, in create
return self._post(
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/openai/_base_client.py", line 1294, in post
return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/openai/_base_client.py", line 1067, in request
raise self._make_status_error_from_response(err.response) from None
openai.AuthenticationError: Error code: 401 - {'error': {'message': 'No cookie auth credentials found', 'code': 401}}
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/usr/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap
self.run()
File "/usr/lib/python3.10/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/core.py", line 131, in _reasoning_wrapper
result = literature_reasoning(*inputs_for_reasoning)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/core.py", line 104, in literature_reasoning
reasoning_response = literature_module(reason_taskpack)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/agentlite/agents/BaseAgent.py", line 115, in __call__
self.execute(task)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/agentlite/agents/BaseAgent.py", line 148, in execute
action = self.__next_act__(task, action_chain)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/modules/literature_reasoning.py", line 492, in __next_act__
raw_action = self.llm_layer(action_prompt)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/.venv/lib/python3.10/site-packages/agentlite/agents/BaseAgent.py", line 136, in llm_layer
return self.llm.run(prompt)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/modules/agent_llms.py", line 119, in run
response = chat_completion(
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/tool_space/gpt_utils.py", line 68, in chat_completion
return _openrouter_completion(messages, temperature, model, attempts, seed, response_format)
File "/mnt/c4a1c56d-0589-48cd-9717-f6c33a86fa3b/Tools/Medea/medea/tool_space/gpt_utils.py", line 146, in _openrouter_completion
time.sleep(wait_time)
KeyboardInterrupt
