""" MCP server for metadata access """
import json
import re
from pathlib import Path
from typing import Any, Dict, Optional, Union
from urllib.parse import urlparse
import boto3
from aind_data_access_api.document_db import MetadataDbClient
from fastmcp import FastMCP
from hdmf_zarr import NWBZarrIO
from suffix_trees import STree
mcp = FastMCP("aind_data_access")
def setup_mongodb_client():
"""Set up and return the MongoDB client"""
API_GATEWAY_HOST = "api.allenneuraldynamics.org"
DATABASE = "metadata_index"
COLLECTION = "data_assets"
return MetadataDbClient(
host=API_GATEWAY_HOST,
database=DATABASE,
collection=COLLECTION,
)
@mcp.tool()
def get_records(filter: dict = {}, projection: dict = {}, limit: int = 5):
"""
Retrieves documents from MongoDB database using simple filters
and projections.
WHEN TO USE THIS FUNCTION:
- For straightforward document retrieval based on specific criteria
- When you need only a subset of fields from documents
- When the query logic doesn't require multi-stage processing
- For better performance with simpler queries
NOT RECOMMENDED FOR:
- Complex data transformations (use aggregation_retrieval instead)
- Grouping operations or calculations across documents
- Joining or relating data across collections
- Trying to fetch an entire data asset (data assets are long and
will clog up the context window)
Parameters
----------
filter : dict
MongoDB query filter to narrow down the documents to retrieve.
Example: {"subject.sex": "Male"}
If empty dict object, returns all documents.
projection : dict
Fields to include or exclude in the returned documents.
Use 1 to include a field, 0 to exclude.
Example: {"subject.genotype": 1, "_id": 0}
will return only the genotype field.
If empty dict object, returns all documents.
limit: int
Limit retrievals to a reasonable number, try to not exceed 100
Returns
-------
list
List of dictionary objects representing the matching documents.
Each dictionary contains the requested fields based on the projection.
"""
docdb_api_client = setup_mongodb_client()
try:
records = docdb_api_client.retrieve_docdb_records(
filter_query=filter, projection=projection, limit=limit
)
return records
except Exception as ex:
template = "An exception of type {0} occurred. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
return message
@mcp.tool()
def aggregation_retrieval(agg_pipeline: list):
"""
Executes a MongoDB aggregation pipeline for complex data transformations
and analysis.
For additional context on how to create filters and projections,
use the retrieve_schema_context tool.
WHEN TO USE THIS FUNCTION:
- When you need to perform multi-stage data processing operations
- For complex queries requiring grouping, filtering, sorting in sequence
- When you need to calculate aggregated values (sums, averages, counts)
- For data transformation operations that can't be done with simple queries
NOT RECOMMENDED FOR:
- Simple document retrieval (use get_records instead)
- When you only need to filter data without transformations
Parameters
----------
agg_pipeline : list
A list of dictionary objects representing MongoDB aggregation stages.
Each stage should be a valid MongoDB aggregation operator.
Common stages include: $match, $project, $group, $sort, $unwind.
Returns
-------
list
Returns a list of documents resulting from the aggregation pipeline.
If an error occurs, returns an error message string describing
the exception.
Notes
-----
- Include a $project stage early in the pipeline to reduce data transfer
- Avoid using $map operator in $project stages as it requires array inputs
"""
docdb_api_client = setup_mongodb_client()
try:
result = docdb_api_client.aggregate_docdb_records(
pipeline=agg_pipeline
)
return result
except Exception as ex:
template = "An exception of type {0} occurred. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
return message
@mcp.tool()
def count_records(filter: dict):
"""
Retrieves number of documents from MongoDB database using
a simple MongoDB filter
WHEN TO USE THIS FUNCTION:
- For counting number of documents based on a straightforward criteria
NOT RECOMMENDED FOR:
- Complex data transformations (use aggregation_retrieval instead)
Parameters
----------
filter : dict
MongoDB query filter to narrow down the documents to retrieve.
Example: {"subject.sex": "Male"}
If empty dict object, returns all documents.
Returns
-------
int
number of records retrieved
"""
docdb_api_client = setup_mongodb_client()
try:
count = docdb_api_client._count_records(filter_query=filter)
return count
except Exception as ex:
template = "An exception of type {0} occurred. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
return message
@mcp.tool()
def get_summary(_id: str):
"""
Get an LLM-generated summary for a data asset, based on the _id field
"""
docdb_api_client = setup_mongodb_client()
try:
result = docdb_api_client.generate_data_summary(_id)
return result
except Exception as ex:
template = "An exception of type {0} occurred. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
return message
@mcp.tool()
def identify_nwb_contents_in_code_ocean(subject_id, date):
"""
Searches the /data directory in a code ocean repository for a folder
and subfolder containing the subject_id and date,
and loads the corresponding NWB file.
Parameters:
- subject_id (str): Subject identifier to search for in directory names
- date (str): Date string (e.g. '2023-05-09') to search for in directory names
Returns:
- nwbfile: Loaded NWBFile object if found, else None
"""
# Create pattern for matching
pattern = rf".*{subject_id}.*{date}.*"
base_path = Path("/data")
# Find matching first-level directories
first_matches = [
d
for d in base_path.iterdir()
if d.is_dir() and re.search(pattern, d.name)
]
if not first_matches:
# print(f"Directory matching subject_id={subject_id} and date={date} not found in /data.")
return None
first_dir = first_matches[0]
# print(f"Found first-level directory: {first_dir.name}")
# Find matching second-level directories
second_matches = [
d
for d in first_dir.iterdir()
if d.is_dir() and re.search(pattern, d.name)
]
if not second_matches:
# print(f"No second-level directory matching subject_id={subject_id} and date={date} found.")
return None
nwb_path = second_matches[0]
print(f"Found second-level directory: {nwb_path.name}")
# Check if path exists and load NWB file
try:
with NWBZarrIO(str(nwb_path), "r") as io:
nwbfile = io.read()
# print('Loaded NWB file from:', nwb_path)
return (
nwbfile.all_children()
) # combination of files in data/asset/asset_nwb
except Exception as e:
# print(f'Error loading file from {nwb_path}: {e}')
return None
@mcp.tool()
def identify_nwb_contents_with_s3_link(s3_link):
"""
Identifies NWB folder in the given S3 link and opens it as a
NWBZarrIO object.
Parameters:
s3_link (str): The S3 link to the folder or file.
Returns:
list: List of contents in NWB folder if found, otherwise None.
"""
# Parse the S3 link
parsed_url = urlparse(s3_link)
bucket_name = parsed_url.netloc
prefix = parsed_url.path.lstrip("/")
# Initialize S3 client
s3 = boto3.client("s3")
try:
# List objects in the given S3 path
response = s3.list_objects_v2(Bucket=bucket_name, Prefix=prefix)
if "Contents" in response:
list_nwb_files = []
for obj in response["Contents"]:
directory_name = obj["Key"]
if "nwb" in directory_name.lower():
list_nwb_files.append(directory_name)
# Identifying common substring in all nwb files (ideally, nwb folder)
s3_nwb_folder = STree.STree(list_nwb_files).lcs()
s3_link_to_nwb = f"s3://{bucket_name}/{s3_nwb_folder}"
# Opening s3 link to nwb folder as an nwb object
with NWBZarrIO(str(s3_link_to_nwb), "r") as io:
nwbfile = io.read() # type pynwb.file.NWBFile
file_contents = (
nwbfile.all_children()
) # return nwbfile.allchildren() list contents as a list
print("Loaded NWB file from:", s3_link_to_nwb)
return file_contents
except Exception as e:
# print(f"Error accessing S3: {e}")
return None
def _flatten_dict(
d: Union[Dict, list],
parent_key: str = "",
sep: str = ".",
depth: Optional[int] = None,
current_depth: int = 0,
) -> Dict[str, Any]:
"""
Recursively flattens a nested dict/list into dot-notation up to `depth`.
If depth=None, fully flatten.
"""
items = []
if isinstance(d, dict) and (depth is None or current_depth < depth):
for k, v in d.items():
new_key = f"{parent_key}{sep}{k}" if parent_key else k
items.extend(
_flatten_dict(
v, new_key, sep, depth, current_depth + 1
).items()
)
elif isinstance(d, list) and (depth is None or current_depth < depth):
for i, v in enumerate(d):
new_key = f"{parent_key}{sep}{i}"
items.extend(
_flatten_dict(
v, new_key, sep, depth, current_depth + 1
).items()
)
else:
items.append((parent_key, d))
return dict(items)
@mcp.tool()
def flatten_records(
filter,
limit,
records: list[dict],
depth: Optional[int] = None,
) -> list[dict]:
"""
Flatten a list of records into dot-notation.
Args:
records (list): List of dicts.
depth (int, optional): How deep to flatten.
Returns:
list[dict]: Each record flattened.
"""
docdb_api_client = setup_mongodb_client()
try:
records = docdb_api_client.retrieve_docdb_records(
filter_query=filter, limit=limit
)
return [_flatten_dict(record, depth=depth) for record in records]
except Exception as ex:
template = "An exception of type {0} occurred. Arguments:\n{1!r}"
message = template.format(type(ex).__name__, ex.args)
return message
@mcp.tool()
def get_project_names() -> list:
"""
Exposes project names in database
"""
docdb_api_client = setup_mongodb_client()
names = docdb_api_client.aggregate_docdb_records(
pipeline=[
{
"$match": {
"data_description.project_name": {
"$exists": True,
"$ne": None,
}
}
},
{"$group": {"_id": "$data_description.project_name"}},
{"$sort": {"_id": 1}},
{"$project": {"_id": 0, "project_name": "$_id"}},
]
)
return names
@mcp.tool()
def get_top_level_nodes() -> list:
"""
This tool exposes the top level nodes of the data schema. In order to access any of the fields using
tools like get_records/aggregation retrieval, you would just have to call the field name like "_id".
Note that most of the fields have further nesting. So in order to call a field within the nesting sturcture,
you would have to use something like "subject.subject_id". Use this tool as a resource. To find out more
about the nesting structure of the fields, you can access the relevant get_<field_name>_example tools in this server.
"""
top_level_nodes = {
"_id": "Unique identifier for data asset assigned by MongoDB, usually a series of numbers",
"name": "Name of data asset (also an identifier assigned during schema curation), follows specific structure <modality>_<subject_id>_<date>)",
"quality_control": "A collection of metrics evaluated on a data asset. Has further nesting",
"acquisition": "Single episode of data collection that creates one data asset for imaging/spim assets. Has further nesting",
"data_description": "Tracks administrative information about a data asset, including affiliated researchers/organizations, projects, data modalities, dates of collection, and more. Has further nesting",
"instrument": "Information about the components, mostly hardware devices, used to collect data for imaging/spim assets. Has further nesting",
"procedures": "Information about anything done to the subject or specimen prior to data collection. Has further nesting",
"processing": "Captures the data processing and analysis steps that have been carried out. Has further nesting",
"rig": "Information about the components, mostly hardware devices, used to collect data for physiology assets. Has further nesting",
"session": "Single episode of data collection that creates one data asset for physiology assets. Has further nesting",
"subject": "Describes the subject from which data was obtained. Has further nesting",
"external_links": "Use external_links.Code Ocean to access code ocean IDs - example: external_links: {Code Ocean: [97189da9-88ea-4d85-b1b0-ceefb9299f1a]",
"location": "s3 link for this data asset",
"schema_version": "version of the schema the asset follows",
}
return top_level_nodes
@mcp.tool()
def get_additional_schema_help():
"""
Advice to follow for creating MongoDB aggregations for the metadata
"""
return """
Key Requirements when creating MongoDB queries:
Always unwind procedures field
Use data_description.modality.name for modality queries
For questions on modalities, always unwind the modalities field.
Use $regex over $elemmatch
Handle duration queries carefully:
To find the duration of a session, strictly follow the following aggregation stage -
{{
$addFields: {{
session_duration_ms: {{
$subtract: [
{{ $dateFromString: {{ dateString: "$session.session_end_time" }} }},
{{ $dateFromString: {{ dateString: "$session.session_start_time" }} }}
]
}}
}}
}}
To find the duration of an acquisition, strictly follow the following aggregation stage -
{{
$addFields: {{
session_duration_ms: {{
$subtract: [
{{ $dateFromString: {{ dateString: "$acquisition.session_end_time" }} }},
{{ $dateFromString: {{ dateString: "$acquisition.session_start_time" }} }}
]
}}
}}
}}
Note that the session field is used for physiology based modalities like behaviour,
optical physiology etc and acquisition is used for modalities like smartspim or exaspim.
If a question is ambiguous or you are unsure which field the data might exist in,
use both fields, do not prioritize one over the other.
"""
@mcp.tool()
def get_modality_types():
"""
Exposes how to access data modality information within data assets
"""
return """
Here are the different modality types:
Access them through data_description.modality.name or data_description.modality.abbreviation
1.
abbreviation: "behavior"
2.
name: "Behavior videos"
abbreviation: "behavior -videos"
3.
name: "Confocal microscopy"
abbreviation: "confocal"
4.
name: "Electromyography"
abbreviation: "EMG"
5.
name: "Extracellular electrophysiology"
abbreviation: "ecephys"
6.
name: "Fiber photometry"
abbreviation: "fib"
7.
name: "Fluorescence micro-optical sectioning tomography"
abbreviation: "fMOST"
8.
name: "Intracellular electrophysiology"
abbreviation: "icephys"
9.
name:"Intrinsic signal imaging"
abbreviation:"ISI"
10.
name:"Magnetic resonance imaging"
abbreviation: "MRI"
11.
name: "Multiplexed error-robust fluorescence in situ hybridization"
abbreviation: "merfish"
12.
name: "Planar optical physiology"
abbreviation: "pophys"
13.
name: "Scanned line projection imaging"
abbreviation: "slap"
14.
name: "Selective plane illumination microscopy"
abbreviation: "SPIM"
"""
@mcp.tool()
def get_quality_control_example() -> dict:
"""
Example of the quality control schema.
"""
return """
The quality_control schema defines how quality metrics are organized and evaluated for data assets:
- Each data asset has an array of "evaluations"
- Each evaluation contains:
- modality: The type of data (SPIM, ecephys, behavior, etc.)
- stage: When quality was assessed (Raw data, Processing, Analysis, Multi-asset)
- metrics: Array of individual measurements with name, value, and status history
- status: Overall Pass/Fail/Pending status of the evaluation
Important quality_control query patterns:
1. To query evaluation properties:
{{"quality_control.evaluations": {{"\\$elemMatch": {{<conditions>}}}}}}
2. To unwind and query individual evaluations:
[{{"\\$unwind": "$quality_control.evaluations"}}, {{"\\$match": {{"quality_control.evaluations.<field>": <value>}}}}]
3. To query metrics within evaluations:
[{{"\\$unwind": "$quality_control.evaluations"}},
{{"\\$unwind": "$quality_control.evaluations.metrics"}},
{{"\\$match": {{"quality_control.evaluations.metrics.name": <metric_name>}}}}]
4. For aggregating QC statistics:
[{{"\\$unwind": "$quality_control.evaluations"}},
{{"\\$group": {{"_id": "$quality_control.evaluations.modality.abbreviation", "count": {{"\\$sum": 1}}}}}}]
Example queries:
- Find assets with failed quality control evaluations:
{{"quality_control.evaluations.latest_status": "Fail"}}
- Find SPIM data with pending QC:
{{"quality_control.evaluations": {{"\\$elemMatch": {{"modality.abbreviation": "SPIM", "latest_status": "Pending"}}}}}}
- Count metrics per evaluation:
{{"\\$project": {{"metricCount": {{"\\$size": "$quality_control.evaluations.metrics"}}}}}}
"""
@mcp.tool()
def get_acquisition_example() -> dict:
"""
Example of the acquisition schema.
Access fields like this - acquisition.<field_name>
"""
sample_acquisition = json.dumps(
{
"acquisition": {
"active_objectives": None,
"axes": [
{
"dimension": 2,
"direction": "Left_to_right",
"name": "X",
"unit": "micrometer",
},
{
"dimension": 1,
"direction": "Posterior_to_anterior",
"name": "Y",
"unit": "micrometer",
},
{
"dimension": 0,
"direction": "Superior_to_inferior",
"name": "Z",
"unit": "micrometer",
},
],
"chamber_immersion": {
"medium": "Cargille oil",
"refractive_index": 1.5208,
},
"experimenter_full_name": "John Rohde",
"external_storage_directory": "",
"instrument_id": "SmartSPIM1-1",
"local_storage_directory": "D:",
"sample_immersion": None,
"schema_version": "0.4.2",
"session_end_time": "2023-03-06T22:59:16",
"session_start_time": "2023-03-06T17:47:13",
"specimen_id": "",
"subject_id": "662616",
"tiles": [
{
"channel": {
"channel_name": "488.0",
"filter_wheel_index": 1,
"laser_power": 20,
"laser_power_unit": "milliwatt",
"laser_wavelength": 488,
"laser_wavelength_unit": "nanometer",
},
"coordinate_transformations": [
{
"translation": [42033, 41585, 10.8],
"type": "translation",
},
{"scale": [1.8, 1.8, 2], "type": "scale"},
],
"file_name": "Ex_488_Em_525/420330/420330_415850/",
"imaging_angle": 0,
"imaging_angle_unit": "degree",
"notes": "\nLaser power is in percentage of total -- needs calibration",
},
{
"channel": {
"channel_name": "445.0",
"filter_wheel_index": 0,
"laser_power": 30,
"laser_power_unit": "milliwatt",
"laser_wavelength": 445,
"laser_wavelength_unit": "nanometer",
},
"coordinate_transformations": [
{
"translation": [45273, 41585, 10.8],
"type": "translation",
},
{"scale": [1.8, 1.8, 2], "type": "scale"},
],
"file_name": "Ex_445_Em_469/452730/452730_415850/",
"imaging_angle": 0,
"imaging_angle_unit": "degree",
"notes": "\nLaser power is in percentage of total -- needs calibration",
},
],
}
}
)
return sample_acquisition
@mcp.tool()
def get_data_description_example():
"""
Example of the data description schema.
Access fields like this - data_description.<field_name>
"""
sample_data_description = json.dumps(
{
"data_description": {
"schema_version": "1.0.0",
"license": "CC-BY-4.0",
"platform": {
"name": "SmartSPIM platform",
"abbreviation": "SmartSPIM",
},
"subject_id": "662616",
"creation_time": "2023-04-14T15:11:04-07:00",
"label": None,
"name": "SmartSPIM_662616_2023-04-14_15-11-04",
"institution": {
"name": "Allen Institute for Neural Dynamics",
"abbreviation": "AIND",
"registry": {
"name": "Research Organization Registry",
"abbreviation": "ROR",
},
"registry_identifier": "04szwah67",
},
"funding_source": [
{
"funder": {
"name": "National Institute of Neurological Disorders and Stroke",
"abbreviation": "NINDS",
"registry": {
"name": "Research Organization Registry",
"abbreviation": "ROR",
},
"registry_identifier": "01s5ya894",
},
"grant_number": "NIH1U19NS123714-01",
"fundee": "Jayaram Chandreashekar, Mathew Summers",
}
],
"data_level": "raw",
"group": "MSMA",
"investigators": [
{
"name": "Mathew Summers",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
{
"name": "Jayaram Chandrashekar",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
],
"project_name": "Thalamus in the middle",
"restrictions": None,
"modality": [
{
"name": "Selective plane illumination microscopy",
"abbreviation": "SPIM",
}
],
"related_data": [],
"data_summary": None,
},
}
)
return sample_data_description
@mcp.tool()
def get_instrument_example():
"""
Example of the instrument schema.
Access fields like this - instrument.<field_name>
"""
sample_instrument = json.dumps(
{
"instrument": {
"schema_version": "0.5.4",
"instrument_id": "SmartSPIM1-2",
"instrument_type": "SmartSPIM",
"location": "615 Westlake",
"manufacturer": "LifeCanvas",
"temperature_control": None,
"humidity_control": False,
"optical_tables": [
{
"name": None,
"serial_number": "Unknown",
"manufacturer": "MKS Newport",
"model": "VIS3648-PG4-325A",
"notes": None,
"length": 36,
"width": 48,
"table_size_unit": "inch",
"vibration_control": True,
}
],
"objectives": [
{
"name": None,
"serial_number": "Unknown",
"manufacturer": "Thorlabs",
"model": "TL4X-SAP",
"notes": "Thorlabs TL4X-SAP with LifeCanvas dipping cap and correction optics",
"numerical_aperture": 0.2,
"magnification": 3.6,
"immersion": "multi",
}
],
"detectors": [
{
"name": None,
"serial_number": "220302-SYS-060443",
"manufacturer": "Hamamatsu",
"model": "C14440-20UP",
"notes": None,
"type": "Camera",
"data_interface": "USB",
"cooling": "water",
}
],
"light_sources": [
{
"name": None,
"serial_number": "VL08223M03",
"manufacturer": "Vortran",
"model": "Stradus",
"notes": "All lasers controlled via Vortran VersaLase System",
"type": "laser",
"coupling": "Single-mode fiber",
"wavelength": 445,
"wavelength_unit": "nanometer",
"max_power": 150,
"power_unit": "milliwatt",
}
],
"fluorescence_filters": [
{
"name": None,
"serial_number": "Unknown-0",
"manufacturer": "Semrock",
"model": "FF01-469/35-25",
"notes": None,
"filter_type": "Band pass",
"diameter": 25,
"diameter_unit": "millimeter",
"thickness": 2,
"thickness_unit": "millimeter",
"filter_wheel_index": 0,
"cut_off_frequency": None,
"cut_off_frequency_unit": "Hertz",
"cut_on_frequency": None,
"cut_on_frequency_unit": "Hertz",
"description": None,
}
],
"motorized_stages": [
{
"name": None,
"serial_number": "Unknown-0",
"manufacturer": "Applied Scientific Instrumentation",
"model": "LS-100",
"notes": "Focus stage",
"travel": 100,
"travel_unit": "millimeter",
}
],
"scanning_stages": [
{
"name": None,
"serial_number": "Unknown-1",
"manufacturer": "Applied Scientific Instrumentation",
"model": "LS-50",
"notes": "Sample stage X",
"travel": 50,
"travel_unit": "millimeter",
"stage_axis_direction": "Illumination axis",
"stage_axis_name": "X",
},
],
"daqs": None,
"additional_devices": [
{
"name": None,
"serial_number": "10436130",
"manufacturer": "Julabo",
"model": "200F",
"notes": None,
"type": "Other",
}
],
"calibration_date": None,
"calibration_data": None,
"com_ports": [
{"hardware_name": "Laser Launch", "com_port": "COM3"}
],
"notes": None,
},
}
)
return sample_instrument
@mcp.tool()
def get_procedures_example():
"""
Example of the procedures schema.
Access fields like this - procedures.<field_name>
"""
sample_procedures = json.dumps(
{
"procedures": {
"schema_version": "0.11.2",
"subject_id": "662616",
"subject_procedures": [
{
"procedure_type": "Surgery",
"start_date": "2023-02-03",
"experimenter_full_name": "30509",
"iacuc_protocol": None,
"animal_weight_prior": None,
"animal_weight_post": None,
"weight_unit": "gram",
"anaesthesia": None,
"workstation_id": None,
"procedures": [
{
"procedure_type": "Perfusion",
"protocol_id": "dx.doi.org/10.17504/protocols.io.bg5vjy66",
"output_specimen_ids": ["662616"],
}
],
"notes": None,
},
{
"procedure_type": "Surgery",
"start_date": "2023-01-05",
"experimenter_full_name": "NSB-5756",
"iacuc_protocol": "2109",
"animal_weight_prior": "16.6",
"animal_weight_post": "16.7",
"weight_unit": "gram",
"anaesthesia": {
"type": "isoflurane",
"duration": "120.0",
"duration_unit": "minute",
"level": "1.5",
},
"workstation_id": "SWS 1",
"procedures": [
{
"injection_materials": [
{
"material_type": "Virus",
"name": "SL1-hSyn-Cre",
"tars_identifiers": {
"virus_tars_id": None,
"plasmid_tars_alias": None,
"prep_lot_number": "221118-11",
"prep_date": None,
"prep_type": None,
"prep_protocol": None,
},
"addgene_id": None,
"titer": {
"$numberLong": "37500000000000"
},
"titer_unit": "gc/mL",
}
],
"recovery_time": "10.0",
"recovery_time_unit": "minute",
"injection_duration": None,
"injection_duration_unit": "minute",
"instrument_id": "NJ#2",
"protocol_id": "dx.doi.org/10.17504/protocols.io.bgpujvnw",
"injection_coordinate_ml": "0.35",
"injection_coordinate_ap": "2.2",
"injection_coordinate_depth": ["2.1"],
"injection_coordinate_unit": "millimeter",
"injection_coordinate_reference": "Bregma",
"bregma_to_lambda_distance": "4.362",
"bregma_to_lambda_unit": "millimeter",
"injection_angle": "0",
"injection_angle_unit": "degrees",
"targeted_structure": "mPFC",
"injection_hemisphere": "Right",
"procedure_type": "Nanoject injection",
"injection_volume": ["200"],
"injection_volume_unit": "nanoliter",
},
{
"injection_materials": [
{
"material_type": "Virus",
"name": "AAV-Syn-DIO-TVA66T-dTomato-CVS N2cG",
"tars_identifiers": {
"virus_tars_id": None,
"plasmid_tars_alias": None,
"prep_lot_number": "220916-4",
"prep_date": None,
"prep_type": None,
"prep_protocol": None,
},
"addgene_id": None,
"titer": {
"$numberLong": "18000000000000"
},
"titer_unit": "gc/mL",
}
],
"recovery_time": "10.0",
"recovery_time_unit": "minute",
"injection_duration": None,
"injection_duration_unit": "minute",
"instrument_id": "NJ#2",
"protocol_id": "dx.doi.org/10.17504/protocols.io.bgpujvnw",
"injection_coordinate_ml": "2.9",
"injection_coordinate_ap": "-0.6",
"injection_coordinate_depth": ["3.6"],
"injection_coordinate_unit": "millimeter",
"injection_coordinate_reference": "Bregma",
"bregma_to_lambda_distance": "4.362",
"bregma_to_lambda_unit": "millimeter",
"injection_angle": "30",
"injection_angle_unit": "degrees",
"targeted_structure": "VM",
"injection_hemisphere": "Right",
"procedure_type": "Nanoject injection",
"injection_volume": ["200"],
"injection_volume_unit": "nanoliter",
},
],
"notes": None,
},
],
"specimen_procedures": [
{
"procedure_type": "Fixation",
"procedure_name": "SHIELD OFF",
"specimen_id": "662616",
"start_date": "2023-02-10",
"end_date": "2023-02-12",
"experimenter_full_name": "DT",
"protocol_id": "none",
"reagents": [
{
"name": "SHIELD Epoxy",
"source": "LiveCanvas Technologies",
"rrid": None,
"lot_number": "unknown",
"expiration_date": None,
}
],
"hcr_series": None,
"immunolabeling": None,
"notes": "None",
},
],
"notes": None,
},
}
)
return sample_procedures
@mcp.tool()
def get_subject_example():
"""
Example of the subject schema.
Access fields like this - subject.<field_name>
"""
sample_subject = json.dumps(
{
"subject": {
"schema_version": "0.4.2",
"species": {
"name": "Mus musculus",
"abbreviation": None,
"registry": {
"name": "National Center for Biotechnology Information",
"abbreviation": "NCBI",
},
"registry_identifier": "10090",
},
"subject_id": "662616",
"sex": "Female",
"date_of_birth": "2022-11-29",
"genotype": "wt/wt",
"mgi_allele_ids": None,
"background_strain": None,
"source": None,
"rrid": None,
"restrictions": None,
"breeding_group": None,
"maternal_id": None,
"maternal_genotype": None,
"paternal_id": None,
"paternal_genotype": None,
"wellness_reports": None,
"housing": None,
"notes": None,
}
}
)
return sample_subject
@mcp.tool()
def get_processing_example():
"""
Example of the processing schema.
Access fields like this - processing.<field_name>
"""
sample_processing = json.dumps(
{
"processing": {
"schema_version": "1.0.0",
"processing_pipeline": {
"data_processes": [],
"processor_full_name": "AIND Scientific Computing",
"pipeline_version": None,
"pipeline_url": None,
"note": None,
},
"analyses": [],
"notes": None,
}
}
)
return sample_processing
@mcp.tool()
def get_rig_example():
"""
Example of the rig schema.
Access fields like this - rig.<field_name>
"""
sample_rig = json.dumps(
{
"rig": {
"schema_version": "0.3.8",
"rig_id": "447-2-B_20240827",
"modification_date": "2024-08-27",
"mouse_platform": {
"device_type": "Tube",
"name": "mouse_tube_foraging",
"serial_number": None,
"manufacturer": {
"name": "Custom",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
"model": None,
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
"surface_material": None,
"date_surface_replaced": None,
"diameter": "3.0",
"diameter_unit": "centimeter",
},
"stimulus_devices": [
{
"device_type": "Reward delivery",
"stage_type": {
"device_type": "Motorized stage",
"name": "NewScaleMotor for LickSpouts",
"serial_number": "46801",
"manufacturer": {
"name": "New Scale Technologies",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
"model": "XYZ Stage with M30LS-3.4-15 linear stages",
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
"travel": "15.0",
"travel_unit": "millimeter",
"firmware": "https://github.com/AllenNeuralDynamics/python-newscale, branch: axes-on-target, commit #7c17497",
},
"reward_spouts": [
{
"device_type": "Reward spout",
"name": "Left lick spout",
"serial_number": None,
"manufacturer": None,
"model": None,
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
"side": "Left",
"spout_diameter": "1.2",
"spout_diameter_unit": "millimeter",
"spout_position": None,
"solenoid_valve": {
"device_type": "Solenoid",
"name": "Solenoid Left",
"serial_number": None,
"manufacturer": {
"name": "The Lee Company",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
"model": "LHDA1233415H",
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
},
"lick_sensor": {
"device_type": "Lick Sensor",
"name": "Lick Sensor Left",
"serial_number": None,
"manufacturer": {
"name": "Janelia Research Campus",
"abbreviation": "Janelia",
"registry": {
"name": "Research Organization Registry",
"abbreviation": "ROR",
},
"registry_identifier": "013sk6x84",
},
"model": None,
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
},
"lick_sensor_type": "Capacitive",
}
],
},
{
"device_type": "Speaker",
"name": "Stimulus Speaker",
"serial_number": None,
"manufacturer": {
"name": "Tymphany",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
"model": "XT25SC90-04",
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
"position": None,
},
],
"cameras": [
{
"name": "BehaviorVideography_FaceSide",
"camera_target": "Face side right",
"camera": {
"device_type": "Detector",
"name": "Side face camera",
"serial_number": None,
"manufacturer": {
"name": "Ailipu Technology Co",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
"model": "ELP-USBFHD05MT-KL170IR",
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": "The light intensity sensor was removed; IR illumination is constantly on",
"detector_type": "Camera",
"data_interface": "USB",
"cooling": "Air",
"computer_name": "W10DT714084",
"max_frame_rate": "120",
"frame_rate_unit": "hertz",
"immersion": None,
"chroma": "Color",
"sensor_width": 640,
"sensor_height": 480,
"size_unit": "pixel",
"sensor_format": None,
"sensor_format_unit": None,
"bit_depth": None,
"bin_mode": "Additive",
"bin_width": None,
"bin_height": None,
"bin_unit": "pixel",
"gain": None,
"crop_width": None,
"crop_height": None,
"crop_unit": "pixel",
"recording_software": {
"name": "Bonsai",
"version": "2.8.0",
"url": None,
"parameters": {},
},
"driver": None,
"driver_version": None,
},
"lens": {
"device_type": "Lens",
"name": "Xenocam 1",
"serial_number": None,
"manufacturer": {
"name": "Other",
"abbreviation": None,
"registry": None,
"registry_identifier": None,
},
"model": "XC0922LENS",
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": "Manufacturer is Xenocam",
"focal_length": "9",
"focal_length_unit": "millimeter",
"size": None,
"lens_size_unit": "inch",
"optimized_wavelength_range": None,
"wavelength_unit": "nanometer",
"max_aperture": "f/1.4",
},
"filter": None,
"position": None,
}
],
"enclosure": {
"device_type": "Enclosure",
"name": "Behavior enclosure",
"serial_number": None,
"manufacturer": {
"name": "Allen Institute for Neural Dynamics",
"abbreviation": "AIND",
"registry": {
"name": "Research Organization Registry",
"abbreviation": "ROR",
},
"registry_identifier": "04szwah67",
},
"model": None,
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
"size": {
"width": 54,
"length": 54,
"height": 54,
"unit": "centimeter",
},
"internal_material": "",
"external_material": "",
"grounded": False,
"laser_interlock": False,
"air_filtration": False,
},
"ephys_assemblies": [],
"fiber_assemblies": [],
"stick_microscopes": [],
"laser_assemblies": [],
"patch_cords": [],
"light_sources": [
{
"device_type": "Light emitting diode",
"name": "IR LED",
"serial_number": None,
"manufacturer": {
"name": "Thorlabs",
"abbreviation": None,
"registry": {
"name": "Research Organization Registry",
"abbreviation": "ROR",
},
"registry_identifier": "04gsnvb07",
},
"model": "M810L5",
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": None,
"wavelength": 810,
"wavelength_unit": "nanometer",
}
],
"detectors": [],
"objectives": [],
"filters": [],
"lenses": [],
"digital_micromirror_devices": [],
"polygonal_scanners": [],
"pockels_cells": [],
"additional_devices": [],
"daqs": [
{
"device_type": "Harp device",
"name": "Harp Behavior",
"serial_number": None,
"manufacturer": {
"name": "Champalimaud Foundation",
"abbreviation": None,
"registry": {
"name": "Research Organization Registry",
"abbreviation": "ROR",
},
"registry_identifier": "03g001n57",
},
"model": None,
"path_to_cad": None,
"port_index": None,
"additional_settings": {},
"notes": "Left lick spout and Right lick spout, as well as reward delivery solenoids are connected via ethernet cables",
"data_interface": "Ethernet",
"computer_name": "W10DT714084",
"channels": [],
"firmware_version": None,
"hardware_version": None,
"harp_device_type": {
"name": "Behavior",
"whoami": 1216,
},
"core_version": "1.11",
"tag_version": None,
"is_clock_generator": False,
}
],
"calibrations": [
{
"calibration_date": "2024-08-27T00:00:00-07:00",
"device_name": "Lick spout Left",
"description": "Water calibration for Lick spout Left. The input is the valve open time in seconds and the output is the volume of water delivered in microliters.",
"input": {"valve open time (s):": [0.02, 0.025, 0.03]},
"output": {
"water volume (ul):": [
1.35,
2,
2.5,
]
},
"notes": None,
}
],
"ccf_coordinate_transform": None,
"origin": None,
"rig_axes": None,
"modalities": [
{
"name": "Behavior videos",
"abbreviation": "behavior-videos",
},
{"name": "Behavior", "abbreviation": "behavior"},
],
"notes": None,
}
}
)
return sample_rig
@mcp.tool()
def get_session_example():
"""
Example of the session schema.
Access fields like this - session.<field_name>
"""
sample_session = json.dumps(
{
"session": {
"schema_version": "0.2.6",
"protocol_id": [""],
"experimenter_full_name": ["Bowen Tan"],
"session_start_time": "2024-09-03T15:49:53.900668-07:00",
"session_end_time": "2024-09-03T17:04:59.585169-07:00",
"session_type": "Uncoupled Baiting",
"iacuc_protocol": "2109",
"rig_id": "447-2-B_20240827",
"calibrations": [],
"maintenance": [],
"subject_id": "730945",
"animal_weight_prior": None,
"animal_weight_post": "23.6",
"weight_unit": "gram",
"anaesthesia": None,
"data_streams": [
{
"stream_start_time": "2024-09-03T15:49:53.900668-07:00",
"stream_end_time": "2024-09-03T17:04:59.585169-07:00",
"daq_names": [
"Harp Behavior",
"Harp Sound",
"Harp clock synchronization board",
"Harp sound amplifier",
"Lick Sensor Left",
"Lick Sensor Right",
],
"camera_names": [],
"light_sources": [],
"ephys_modules": [],
"stick_microscopes": [],
"manipulator_modules": [],
"detectors": [],
"fiber_connections": [],
"fiber_modules": [],
"ophys_fovs": [],
"slap_fovs": None,
"stack_parameters": None,
"mri_scans": [],
"stream_modalities": [
{"name": "Behavior", "abbreviation": "behavior"}
],
"software": [
{
"name": "dynamic-foraging-task",
"version": "behavior branch:main commit ID:04bcaa08294ead9ae3b9a08ff040c09f0999c782 version:1.4.4; metadata branch: main commit ID:04bcaa08294ead9ae3b9a08ff040c09f0999c782 version:1.4.4",
"url": "https://github.com/AllenNeuralDynamics/dynamic-foraging-task.git",
"parameters": {},
}
],
"notes": None,
}
],
"stimulus_epochs": [
{
"stimulus_start_time": "2024-09-03T15:49:53.900668-07:00",
"stimulus_end_time": "2024-09-03T17:04:59.585169-07:00",
"stimulus_name": "auditory go cue",
"session_number": None,
"software": [
{
"name": "dynamic-foraging-task",
"version": "behavior branch:main commit ID:04bcaa08294ead9ae3b9a08ff040c09f0999c782 version:1.4.4; metadata branch: main commit ID:04bcaa08294ead9ae3b9a08ff040c09f0999c782 version:1.4.4",
"url": "https://github.com/AllenNeuralDynamics/dynamic-foraging-task.git",
"parameters": {},
}
],
"script": None,
"stimulus_modalities": ["Auditory"],
"stimulus_parameters": [
{
"stimulus_type": "Auditory Stimulation",
"sitmulus_name": "auditory go cue",
"sample_frequency": "96000",
"amplitude_modulation_frequency": 7500,
"frequency_unit": "hertz",
"bandpass_low_frequency": None,
"bandpass_high_frequency": None,
"bandpass_filter_type": None,
"bandpass_order": None,
"notes": None,
}
],
"stimulus_device_names": [],
"speaker_config": {
"name": "Stimulus Speaker",
"volume": "76",
"volume_unit": "decibels",
},
"light_source_config": None,
"output_parameters": {
"meta": {
"box": "447-2-B",
"session_end_time": "2024-09-03 17:04:59.585169",
"session_run_time_in_min": 75,
},
"water": {
"water_in_session_foraging": 0.372,
"water_in_session_manual": 0.09599999999999997,
"water_in_session_total": 0.46799999999999997,
"water_after_session": 0.532,
"water_day_total": 1,
},
"weight": {
"base_weight": 23.18,
"target_weight": 19.703,
"target_weight_ratio": 0.85,
"weight_after": 23.6,
},
"performance": {
"foraging_efficiency": 0.5931917753463685,
"foraging_efficiency_with_actual_random_seed": 0.6138613861386139,
},
"task_parameters": {
"SessionlistSpin": "1",
"qt_spinbox_lineedit": "3.00",
"Load": False,
"NewSession": False,
"Clear": False,
"Start": False,
"Save": False,
"Sessionlist": "",
"HideLegend": False,
"ID": "730945",
"Experimenter": "Bowen Tan",
"AutoTrain": False,
"pushButton_streamlit": False,
"StartFIP": False,
"StartExcitation": False,
"StartBleaching": False,
"FIPMode": "Normal",
"baselinetime": "10",
"PhotometryB": "off",
"OptogeneticsB": "off",
"StartEphysRecording": False,
"OpenEphysRecordingType": "Behavior",
"MoveYN": False,
"MoveZN": False,
"StageStop": False,
"MoveYP": False,
"MoveZP": False,
"MoveXN": False,
"MoveXP": False,
"PositionZ": "6358.5",
"PositionY": "14387.5",
"PositionX": "6400.0",
"Step": "200",
"SetReference": False,
"GetPositions": False,
"StepSize": "5",
"WindowSize": "100",
"RunLength": "10",
"MartchingType": "log ratio",
"TargetRatio": "0.85",
"TotalWater": "1.0",
"SuggestedWater": "0.532",
"BaseWeight": "23.18",
"TargetWeight": "19.703",
"WeightAfter": "23.6",
"ShowNotes": "",
"ITIMin": "1.0",
"DelayMin": "1.0",
"InitiallyInactiveN": "2",
"DelayBeta": "0.0",
"DelayMax": "1.0",
"BlockMinReward": "0",
"AutoReward": False,
"ITIIncrease": "0",
"AutoWaterType": "Natural",
"Task": "Uncoupled Baiting",
"BlockMax": "35",
"RewardDelay": "0.0",
"SwitchThr": "0.5",
"ITIMax": "30.0",
"ITIBeta": "3.0",
"ResponseTime": "1.0",
"PointsInARow": "5",
"AdvancedBlockAuto": "now",
"Ignored": "10",
"BlockBeta": "10",
"BlockMin": "20",
"StopIgnores": "25",
"RewardPairsN": "1",
"MaxTime": "75",
"RewardConsumeTime": "3.0",
"NextBlock": False,
"IncludeAutoReward": "no",
"Multiplier": "0.5",
"RewardFamily": "1",
"RightValue_volume": "2.00",
"UncoupledReward": "0.1, 0.4, 0.7",
"RightValue": "0.023",
"BaseRewardSum": "0.8",
"LeftValue": "0.025",
"MaxTrial": "1000",
"Unrewarded": "10",
"LeftValue_volume": "2.00",
"warm_windowsize": "20",
"warm_max_choice_ratio_bias": "0.1",
"warmup": "off",
"warm_min_finish_ratio": "0.8",
"warm_min_trial": "50",
"Randomness": "Exponential",
"AddOneTrialForNoresponse": "Yes",
"GiveRight": False,
"GiveWaterL": "0.034",
"GiveWaterR": "0.035",
"GiveWaterL_volume": "3.00",
"GiveWaterR_volume": "3.00",
"GiveLeft": False,
"AlignToGoCue": "yes",
"MoveXP_2": False,
"MoveXN_2": False,
"LaserCalibration_dialog": {
"Frequency_1": "40",
"Protocol_1": "Sine",
"Location_1": "Both",
"Duration_1": "10",
"RD_1": "1",
"LaserColor_1": "Blue",
"PulseDur_1": "0.002",
"KeepOpen": False,
"SampleFrequency": "5000",
"CopyFromOpto": False,
"LaserPowerMeasured": "",
"Save": False,
"voltage": "0.1",
"Capture": False,
"Flush_DO0": False,
"Flush_DO1": False,
"Flush_DO2": False,
"Flush_DO3": False,
"Flush_Port2": False,
"showrecentlaser": "",
"qt_spinbox_lineedit": "",
"showspecificcalilaser": "NA",
"CopyToSession": False,
"CopyCondition": "Condition_1",
"CopyLaser": "Laser_1",
"Open": False,
},
"Opto_dialog": {
"MinOptoInterval": "0",
"SessionWideControl": "off",
"FractionOfSession": "0.5",
"SessionStartWith": "on",
"SessionAlternating": "on",
"SampleFrequency": "5000",
"LatestCalibrationDate": "NA",
"Laser_calibration": "Blue",
"laser_1_target": "",
"laser_2_target": "",
"laser_2_calibration_voltage": "",
"laser_1_calibration_power": "",
"laser_1_calibration_voltage": "",
"laser_2_calibration_power": "",
"OffsetStart_2": "0",
"Probability_2": "0.25",
"OffsetEnd_2": "0",
"Duration_2": "5",
"LaserEnd_2": "Trial start",
"RD_2": "1",
"LaserColor_2": "NA",
"Location_2": "Both",
"Frequency_2": "40",
"Laser2_power_2": "0 mw",
"Laser1_power_2": "0 mw",
"LaserStart_2": "Trial start",
"Protocol_2": "Sine",
"Condition_2": "NA",
"Laser2_power_3": "0 mw",
"Location_3": "Both",
"LaserStart_3": "Trial start",
"LaserEnd_3": "Trial start",
"OffsetEnd_3": "0",
"Condition_3": "NA",
"OffsetStart_3": "0",
"LaserColor_3": "NA",
"Protocol_3": "Sine",
"Laser1_power_3": "0 mw",
"Duration_3": "5",
"Probability_3": "0.25",
"Frequency_3": "40",
"RD_3": "1",
"LaserStart_4": "Trial start",
"Laser1_power_4": "0 mw",
"Duration_4": "5",
"LaserColor_4": "NA",
"OffsetStart_4": "0",
"Laser2_power_4": "0 mw",
"LaserEnd_4": "Trial start",
"OffsetEnd_4": "0",
"Location_4": "Both",
"Probability_4": "0.25",
"Condition_4": "NA",
"Frequency_4": "40",
"Protocol_4": "Sine",
"RD_4": "1",
"LaserStart_5": "Trial start",
"LaserColor_5": "NA",
"Probability_1": "0.25",
"Laser1_power_1": "0 mw",
"LaserColor_1": "NA",
"Location_1": "Both",
"Laser2_power_1": "0 mw",
"RD_1": "1",
"Frequency_1": "40",
"OffsetEnd_1": "0",
"LaserEnd_1": "Trial start",
"Duration_1": "5",
"Condition_1": "NA",
"LaserStart_1": "Trial start",
"OffsetStart_1": "0",
"Protocol_1": "Sine",
"OffsetEnd_5": "0",
"PulseDur_2": "0.002",
"Laser2_power_5": "0 mw",
"LaserEnd_5": "Trial start",
"Probability_5": "0.25",
"Frequency_5": "40",
"Duration_5": "5",
"Protocol_5": "Sine",
"ConditionP_5": "1",
"PulseDur_4": "0.002",
"Condition_5": "NA",
"RD_5": "1",
"PulseDur_1": "0.002",
"PulseDur_3": "0.002",
"ConditionP_4": "1",
"ConditionP_2": "1",
"PulseDur_5": "0.002",
"ConditionP_3": "1",
"OffsetStart_5": "0",
"Laser1_power_5": "0 mw",
"Location_5": "Both",
"ConditionP_1": "1",
"LaserColor_6": "NA",
"LaserStart_6": "Trial start",
"Location_6": "Both",
"OffsetStart_6": "0",
"Laser1_power_6": "0 mw",
"LaserEnd_6": "Trial start",
"Laser2_power_6": "0 mw",
"OffsetEnd_6": "0",
"Probability_6": "0.25",
"Protocol_6": "Sine",
"Duration_6": "5",
"Frequency_6": "40",
"Condition_6": "NA",
"RD_6": "1",
"ConditionP_6": "1",
"PulseDur_6": "0.002",
},
"Camera_dialog": {
"StartRecording": False,
"AutoControl": "No",
"FrameRate": "500",
"qt_spinbox_lineedit": "500",
"OpenSaveFolder": False,
"StartPreview": False,
"camera_start_time": "",
"camera_stop_time": "",
},
"Metadata_dialog": {
"RotationAngle": "",
"EphysProbes": "",
"ModuleAngle": "",
"ArcAngle": "",
"ManipulatorX": "",
"ManipulatorZ": "",
"ManipulatorY": "",
"ProbeTarget": "",
"ExperimentDescription": "Behavior training\n",
"SaveMeta": False,
"IACUCProtocol": "2109",
"ProtocolID": "",
"RigMetadataFile": "rig_447-2-B_2024-08-27_10_53_22.json",
"SelectRigMetadata": False,
"ClearMetadata": False,
"StickMicroscopes": "",
"Stick_ModuleAngle": "",
"Stick_ArcAngle": "",
"Stick_RotationAngle": "",
"DataSummary": "",
"FundingSource": "",
"Investigators": "",
"ProjectName": "",
"GrantNumber": "",
"Fundee": "",
"LoadMeta": False,
"GoCueDecibel": "76",
"LickSpoutReferenceZ": "",
"LickSpoutReferenceY": "",
"LickSpoutReferenceX": "",
"LickSpoutReferenceArea": "",
"LickSpoutDistance": "5000",
},
"Other_CurrentTime": "2024-09-03 17:04:59.585169",
"Other_RunningTime": 75,
"Other_SessionStartTime": "2024-09-03 15:49:53.900668",
"Other_current_box": "447-2-B",
"Other_go_cue_decibel": "76",
"Other_lick_spout_distance": "5000",
"info_performance_essential_1": "Current trial: 493\nResponded trial: 434/493 (0.88)\nReward Trial: 186/493 (0.38)\nEarned Reward: 0.372 mL\nWater in session: 0.468 mL",
"info_performance_essential_2": "Foraging eff: 0.59\nForaging eff (r.s.): 0.61\n\nBias: -1.26 (left)",
"info_performance_others": "Left choice rewarded: 125/352 (0.36)\nRight choice rewarded: 61/82 (0.74)\n\nEarly licking (EL)\n Frac of EL trial start_goCue: 243/492 (0.49)\n Frac of EL trial start_delay: 212/492 (0.43)\n Frac of EL trial delay_goCue: 90/492 (0.18)\n Left/Right early licks start_goCue: 778/152 (5.12)\n\nDouble dipping (DD)\n Frac of DD trial start_goCue: 27/492 (0.05)\n Frac of DD trial start_delay: 15/492 (0.03)\n Frac of DD trial delay_goCue: 13/492 (0.03)\n Frac of DD trial goCue_goCue1: 16/492 (0.03)\n DD per finish trial start_goCue: 0.1\n DD per finish trial goCue_goCue1: 0.04\n\n Frac of DD trial goCue_nextStart: 31/491 (0.06)\n DD per finish trial goCue_nextStart: 0.08\n",
"info_task": "Session started: 15:49\nCurrent time: 17:04\nRun time: 75 mins\n\nCurrent left block: 11/24\nCurrent right block: 30/32",
"Ot_log_folder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\behavior\\raw.harp",
"box": "447-2-B",
"settings": {
"default_saveFolder": "C:\\behavior_data\\",
"current_box": "447-2",
"Teensy_COM_box1": "",
"Teensy_COM_box2": "",
"Teensy_COM_box3": "",
"Teensy_COM_box4": "",
"bonsai_path": "C:\\Users\\svc_aind_behavior\\Documents\\Github\\dynamic-foraging-task\\bonsai\\Bonsai.exe",
"bonsaiworkflow_path": "C:\\Users\\svc_aind_behavior\\Documents\\Github\\dynamic-foraging-task\\src\\workflows\\foraging.bonsai",
"temporary_video_folder": "C:\\Users\\svc_aind_behavior\\Documents\\temporaryvideo\\",
"show_log_info_in_console": False,
"newscale_serial_num_box1": "46103",
"newscale_serial_num_box2": "46801",
"newscale_serial_num_box3": "",
"newscale_serial_num_box4": "",
"default_ui": "ForagingGUI.ui",
"create_rig_metadata": True,
"go_cue_decibel_box1": 75,
"go_cue_decibel_box2": 76,
"go_cue_decibel_box3": 75,
"go_cue_decibel_box4": 75,
"FIP_workflow_path": "",
"FIP_settings": "C:\\Users\\svc_aind_behavior\\Documents\\FIPSettings",
"bonsai_config_path": "C:\\Users\\svc_aind_behavior\\Documents\\GitHub\\dynamic-foraging-task\\bonsai\\Bonsai.config",
"open_ephys_machine_ip_address": "",
"metadata_dialog_folder": "C:\\Users\\svc_aind_behavior\\Documents\\ForagingSettings\\metadata_dialog\\",
"rig_metadata_folder": "C:\\Users\\svc_aind_behavior\\Documents\\ForagingSettings\\rig_metadata\\",
"project_info_file": "C:\\Users\\svc_aind_behavior\\Documents\\ForagingSettings\\Project Name and Funding Source v2.csv",
"schedule_path": "Z:\\dynamic_foraging\\DynamicForagingSchedule.csv",
"lick_spout_distance_box1": 5000,
"lick_spout_distance_box2": 5000,
"lick_spout_distance_box3": 5000,
"lick_spout_distance_box4": 5000,
"name_mapper_file": "C:\\Users\\svc_aind_behavior\\Documents\\ForagingSettings\\name_mapper.json",
"save_each_trial": True,
"AutomaticUpload": True,
"manifest_flag_dir": "C:\\Users\\svc_aind_behavior\\Documents\\aind_watchdog_service\\manifest",
"auto_engage": True,
"clear_figure_after_save": True,
"default_openFolder": "C:\\behavior_data\\",
},
"settings_box": {
"Behavior": "COM7",
"Soundcard": "COM5",
"BonsaiOsc1": "4012",
"BonsaiOsc2": "4013",
"BonsaiOsc3": "4014",
"BonsaiOsc4": "4015",
"HighSpeedCamera": "0",
"AINDLickDetector": "0",
"AttenuationLeft": "50",
"AttenuationRight": "50",
"current_box": "447-2-B",
},
"commit_ID": "04bcaa08294ead9ae3b9a08ff040c09f0999c782",
"repo_url": "https://github.com/AllenNeuralDynamics/dynamic-foraging-task.git",
"current_branch": "main",
"repo_dirty_flag": False,
"dirty_files": "",
"version": "1.4.4",
"open_ephys": [],
"ManualWaterVolume": [0, 0],
"saving_type_label": "normal saving",
"SessionFolder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53",
"TrainingFolder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\behavior",
"HarpFolder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\behavior\\raw.harp",
"VideoFolder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\behavior-videos",
"PhotometryFolder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\fib",
"MetadataFolder": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\metadata-dir",
"SaveFile": "C:\\behavior_data\\447-2-B\\730945\\behavior_730945_2024-09-03_15-49-53\\behavior\\730945_2024-09-03_15-49-53.json",
"fiber_photometry_start_time": "",
"fiber_photometry_end_time": "",
"fiber_mode": "Normal",
"session_metadata": {},
"generate_session_metadata_success": True,
"generate_rig_metadata_success": True,
"generate_data_description_success": False,
"drop_frames_tag": 0,
"trigger_length": 0,
"drop_frames_warning_text": "",
"frame_num": {},
},
},
"reward_consumed_during_epoch": "372.0",
"reward_consumed_unit": "microliter",
"trials_total": 493,
"trials_finished": 434,
"trials_rewarded": 186,
"notes": "Notes on reward consumption",
}
],
"mouse_platform_name": "mouse_tube_foraging",
"active_mouse_platform": False,
"reward_delivery": None,
"reward_consumed_total": "372.0",
"reward_consumed_unit": "microliter",
"notes": "",
}
}
)
return sample_session
@mcp.resource("resource://aind_api")
def get_aind_data_access_api() -> str:
"""
Get context on how to use the AIND data access api to show users how to
wrap tool calls
"""
resource_path = Path(__file__).parent / "resources" / "aind_api_prompt.txt"
with open(resource_path, "r") as file:
file_content = file.read()
return file_content
@mcp.resource("resource://load_nwbfile")
def get_nwbfile_download_script() -> str:
"""
Get context on how to return an NWBfile from the /data folder in current repository
"""
resource_path = Path(__file__).parent / "resources" / "load_nwbfile.txt"
with open(resource_path, "r") as file:
file_content = file.read()
return file_content
def main():
"""Main entry point for the MCP server."""
mcp.run(transport="stdio")
if __name__ == "__main__":
main()
