OpenKinetics API Reference¶

This page documents the OpenKinetics API-backed kinetic scorers. The package provides an HTTP client for the OpenKinetics Predictor REST API, a family of dynamically-created scorer classes registered with the Magician plugin system, and helpers for PDB ligand discovery and SMILES resolution.

Architecture¶

The OpenKinetics package is structured as a Magician designer subpackage under REvoDesign.magician.designers.openkinetics/.

openkinetics/
├── __init__.py      # Re-exports all public symbols + dynamic scorer classes
├── _client.py       # REST API client, config loading, data I/O, result normalisation
├── _models.py       # Exception hierarchy, dataclasses, endpoint constants
├── _scorers.py      # Abstract base scorer + dynamic subclass creation via _SCORER_SPECS
└── _pdb.py          # PDB ligand discovery, SMILES extraction, mutation label helpers

Scorer classes are generated dynamically at import time from the _SCORER_SPECS tuple in _scorers.py. Each spec produces a concrete subclass of OpenKineticsScorerAbstract with a fixed method, prediction_type, and citation key. The __init__.py re-exports these classes by name.

Beyond the core scoring workflow (submit, status, result), the OpenKineticsClient also wraps four additional service endpoints: check_health() (/health/), list_methods() (/methods/), get_status() (/status/{job_id}/), and check_quota() (/quota/). The submit() method accepts optional parameters handle_long_sequences, use_experimental, include_similarity_columns, and canonicalize_substrates.

YAML Configuration¶

The package reads its runtime configuration from config/third_party/scorers/openkinetics_api.yaml:

scorers:
  openkinetics:
    enabled: false
    base_url: "https://predictor.openkinetics.org/api/v1"
    default_method: "CataPro"
    default_prediction_type: "kcat/Km"
    poll_interval_seconds: 3
    timeout_seconds: 600
    cache_enabled: true

Key	Type	Default	Description
`enabled`	`bool`	`false`	Whether the OpenKinetics scorers are available.
`base_url`	`str`	`https://predictor.openkinetics.org/api/v1`	Base URL of the OpenKinetics Predictor API.
`default_method`	`str`	`CataPro`	Default prediction method.
`default_prediction_type`	`str`	`kcat/Km`	Default prediction target.
`poll_interval_seconds`	`int`	`3`	Seconds between status-poll requests.
`timeout_seconds`	`int`	`600`	Maximum overall wall-clock time for a prediction job.
`cache_enabled`	`bool`	`true`	Whether per-variant SQLite caching is on by default.

The config is loaded at runtime via load_openkinetics_config().

API Key Configuration¶

The API key is auto-registered by default. OpenKineticsScorerAbstract defaults auto_register_api_key=True, which triggers automatic key generation and persistence on first use (see resolve_api_key() below). Manual setup in environ.yaml is still supported as a fallback and takes precedence if set.

Constants¶

`REvoDesign.magician.designers.openkinetics._models.DEFAULT_OPENKINETICS_API_KEY_ENV = 'OPENKINETICS_API_KEY'` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.OPENKINETICS_ENDPOINTS = {'health': '/health/', 'methods': '/methods/', 'validate': '/validate/', 'submit': '/submit/', 'status': '/status/{job_id}/', 'result': '/result/{job_id}/', 'quota': '/quota/'}` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.WATER_RESIDUE_NAMES = frozenset({'HOH', 'WAT'})` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.COFACTOR_EXCLUSIONS = frozenset({'HEM'})` `module-attribute` ¶

Exception Hierarchy¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsError` ¶

Bases: RuntimeError

Base OpenKinetics integration error.

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsConfigurationError` ¶

Bases: OpenKineticsError

Raised when required local configuration is missing or invalid.

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsAPIError` ¶

Bases: OpenKineticsError

Raised for HTTP or remote API failures.

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsTimeoutError` ¶

Bases: OpenKineticsError

Raised when a remote job does not complete within the timeout.

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsValidationError` ¶

Bases: OpenKineticsError

Raised when prepared OpenKinetics input data is invalid.

Dataclasses¶

`REvoDesign.magician.designers.openkinetics._models.LigandCandidate` `dataclass` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsFixturePaths` `dataclass` ¶

Configuration Functions¶

`REvoDesign.magician.designers.openkinetics._client.load_openkinetics_config()` ¶

Load OpenKinetics config from third_party/scorers/openkinetics_api.yaml.

`REvoDesign.magician.designers.openkinetics._client.resolve_api_key(*, api_key=None, auto_register=False, replace_existing=False, base_url=None, timeout_seconds=None, session=None)` ¶

Resolve the OpenKinetics API key from various sources in order of preference.

This function attempts to find an OpenKinetics API key from multiple sources: 1. Directly supplied API key parameter 2. Environment variable 3. Auto-registration (if enabled)

Parameters:

Name	Type	Description	Default
`api_key`	`str \| None`	Direct API key provided as a parameter. If not provided or empty, other sources will be checked.	`None`
`auto_register`	`bool`	Whether to automatically register and persist a new API key if no existing key is found.	`False`
`replace_existing`	`bool`	Whether to replace any existing API key during auto-registration.	`False`
`base_url`	`str \| None`	Base URL for API calls when fetching a new key during auto-registration.	`None`
`timeout_seconds`	`int \| None`	Timeout value for API requests during auto-registration.	`None`
`session`	`Session \| None`	Requests session object for making HTTP requests during auto-registration.	`None`

Returns:

Name	Type	Description
`str`	`str`	The resolved OpenKinetics API key as a string.

Raises:

Type	Description
`OpenKineticsConfigurationError`	If no API key is found and auto_registration is disabled.

`REvoDesign.magician.designers.openkinetics._client.fetch_openkinetics_api_key(*, base_url=None, timeout_seconds=None, session=None, replace_existing=False)` ¶

Generate a self-service OpenKinetics API key for the current client IP.

Parameters:

Name	Type	Description	Default
`base_url`	`str \| None`	The base URL for the OpenKinetics API. If None, uses the configured default.	`None`
`timeout_seconds`	`int \| None`	Request timeout in seconds. If None, uses the configured default.	`None`
`session`	`Session \| None`	Requests session to use for making HTTP requests. If None, creates a new session.	`None`
`replace_existing`	`bool`	Whether to replace any existing API key if one exists for this IP.	`False`

Returns:

Name	Type	Description
`str`	`str`	The generated OpenKinetics API key.

Raises:

Type	Description
`OpenKineticsAPIError`	If there's an error during API key generation or communication.
`OpenKineticsConfigurationError`	If an API key already exists but isn't available locally.

`REvoDesign.magician.designers.openkinetics._client.persist_openkinetics_api_key(api_key)` ¶

Persist an OpenKinetics API key and register it for this process.

Parameters:

Name	Type	Description	Default
`api_key`	`str`	The OpenKinetics API key to persist. Must be a non-empty string.	required

Returns:

Name	Type	Description
`str`	`str`	The persisted API key that was passed in.

Raises:

Type	Description
`OpenKineticsConfigurationError`	If the API key is empty or if persistence to environ.yaml fails.

Data I/O Helpers¶

`REvoDesign.magician.designers.openkinetics._client.build_openkinetics_data_rows(variant_rows, substrate_smiles)` ¶

Builds OpenKinetics data rows by combining protein sequences from variant rows with a common substrate.

Parameters:

Name	Type	Description	Default
`variant_rows`	`list[dict[str, str]]`	A list of dictionaries containing variant data, each with a 'protein_sequence' key	required
`substrate_smiles`	`str`	The SMILES string representing the substrate to be used for all entries	required

Returns:

Type	Description
`list[dict[str, str]]`	A list of dictionaries where each dictionary contains 'Protein Sequence' and 'Substrate' keys

`REvoDesign.magician.designers.openkinetics._client.build_openkinetics_request_payload(*, data_rows, method_metadata, method, prediction_type)` ¶

`REvoDesign.magician.designers.openkinetics._client.get_method_metadata(methods_response, *, method, prediction_type)` ¶

Extract metadata for a specific method from the OpenKinetics methods response.

Parameters:

Name	Type	Description	Default
`methods_response`	`dict[str, Any]`	Dictionary containing the methods response from OpenKinetics API	required
`method`	`str`	The identifier or display name of the method to find	required
`prediction_type`	`str`	The type of prediction to filter methods by	required

Returns:

Type	Description
`dict[str, Any]`	Dictionary containing the metadata for the requested method

Raises:

Type	Description
`OpenKineticsValidationError`	If the methods response is malformed or the requested method is not found

`REvoDesign.magician.designers.openkinetics._client.write_csv_rows(path, rows)` ¶

`REvoDesign.magician.designers.openkinetics._client.write_json(path, payload)` ¶

`REvoDesign.magician.designers.openkinetics._client.sha256_file(path)` ¶

`REvoDesign.magician.designers.openkinetics._client.write_normalized_scores_csv(path, rows)` ¶

`REvoDesign.magician.designers.openkinetics._client._normalize_result_rows(result_payload, *, method, prediction_type, substrate_smiles, variant_rows=None, job_id='')` ¶

OpenKineticsClient¶

The HTTP client for the OpenKinetics Predictor REST API. Used by both the scoring workflow (OpenKineticsScorerAbstract) and the manual fixture-collection workflow.

`REvoDesign.magician.designers.openkinetics._client.OpenKineticsClient` ¶

HTTP client for the OpenKinetics Predictor REST API.

Used by both the scoring workflow (OpenKineticsScorerAbstract) and the manual fixture-collection workflow (collect_openkinetics_fixture_dataset).

`list_methods()` ¶

Retrieve a list of available methods from the OpenKinetics API.

This method sends a GET request to the 'methods' endpoint to fetch information about the available methods in the OpenKinetics service.

Parameters:

Name	Type	Description	Default
`None`		This method does not take any parameters.	required

Returns:

Name	Type	Description
`Any`	`Any`	The response from the OpenKinetics API containing method information. The exact type and structure depends on the API response format.

`validate_file(csv_path, *, run_similarity=False)` ¶

Validates a CSV file using the OpenKinetics API.

Parameters:

Name	Type	Description	Default
`csv_path`	`str \| Path`	Path to the CSV file to be validated	required
`run_similarity`	`bool`	Whether to run similarity analysis during validation (default: False)	`False`

Returns:

Type	Description
`Any`	JSON response from the OpenKinetics API

Raises:

Type	Description
`OpenKineticsAPIError`	If the API request fails or returns an error status code

`validate(rows, *, run_similarity=False)` ¶

Validates a list of data rows by normalizing input and processing through temporary CSV file.

Parameters:

Name	Type	Description	Default
`rows`	`list[dict[str, Any]]`	A list of dictionaries containing protein sequence and substrate information	required
`run_similarity`	`bool`	Boolean flag to indicate whether similarity analysis should be performed	`False`

Returns:

Type	Description
`dict[str, Any]`	A dictionary containing validation results

`check_health()` ¶

Check the OpenKinetics service health endpoint.

Parameters:

Name	Type	Description	Default
`self`		The instance of the class containing this method	required

Returns:

Type	Description
`dict[str, Any]`	dict[str, Any]: A dictionary containing the health status response from the OpenKinetics service

`check_quota()` ¶

Check the account quota (daily usage / limit).

Parameters:

Name	Type	Description	Default
`self`		The instance of the class containing this method	required

Returns:

Type	Description
`dict[str, Any]`	dict[str, Any]: A dictionary containing quota information including daily usage and limit data

`submit(rows, method, prediction_type, *, handle_long_sequences='truncate', use_experimental=False, include_similarity_columns=True, canonicalize_substrates=True)` ¶

Submit a batch of protein sequence and substrate pairs to the OpenKinetics service for prediction.

Parameters:

Name	Type	Description	Default
`self`		The instance of the class containing this method	required
`rows`	`list[dict[str, Any]]`	A list of dictionaries, each containing 'Protein Sequence' and 'Substrate' keys with corresponding values	required
`method`	`str`	The prediction method to be used for the analysis	required
`prediction_type`	`str`	The type of prediction to perform	required
`handle_long_sequences`	`str`	Strategy for handling long sequences, defaults to "truncate"	`'truncate'`
`use_experimental`	`bool`	Whether to use experimental features, defaults to False	`False`
`include_similarity_columns`	`bool`	Whether to include similarity columns in the output, defaults to True	`True`
`canonicalize_substrates`	`bool`	Whether to canonicalize substrates, defaults to True	`True`

Returns:

Type	Description
`dict[str, Any]`	A dictionary containing the response from the OpenKinetics service, including job identifier

Raises:

Type	Description
`OpenKineticsValidationError`	If no rows are provided or if the submit response doesn't contain a job identifier

`get_result(job_id, result_format='json', *, poll_interval_seconds=3, timeout_seconds=None)` ¶

Retrieves the result of a job from the OpenKinetics API.

Parameters:

Name	Type	Description	Default
`job_id`	`str`	The unique identifier of the job to retrieve results for	required
`result_format`	`str`	Format of the result, either 'json' or 'csv' (default is 'json')	`'json'`
`poll_interval_seconds`	`int`	Interval in seconds between polling attempts when result is not ready (default is 3)	`3`
`timeout_seconds`	`int \| None`	Maximum time to wait for the result in seconds, None means default timeout (default is None)	`None`

Returns:

Type	Description
`dict[str, Any] \| str`	A dictionary containing the result data if format is JSON, or a string containing CSV data if format is CSV

Raises:

Type	Description
`OpenKineticsAPIError`	If the API request fails or returns an error status
`OpenKineticsTimeoutError`	If the timeout is exceeded while waiting for the result
`OpenKineticsValidationError`	If an unsupported result format is specified

`poll_until_complete(job_id, *, poll_interval_seconds=3, timeout_seconds=600)` ¶

Polls the job status until completion or timeout.

Parameters:

Name	Type	Description	Default
`job_id`	`str`	The identifier of the job to poll	required
`poll_interval_seconds`	`int`	Time interval between status checks in seconds (default: 3)	`3`
`timeout_seconds`	`int`	Maximum time to wait for job completion in seconds (default: 600)	`600`

Returns:

Type	Description
`list[Any]`	A list of status payloads collected during polling

Raises:

Type	Description
`OpenKineticsAPIError`	If the job fails or encounters an error
`OpenKineticsTimeoutError`	If the timeout is reached before job completion

PDB / Ligand Functions¶

`REvoDesign.magician.designers.openkinetics._pdb.discover_ligand_candidates(pdb_path, excluded_residue_names=WATER_RESIDUE_NAMES | COFACTOR_EXCLUSIONS)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.choose_primary_ligand(pdb_path)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.extract_ligand_pdb_block(pdb_path, ligand)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.smiles_from_ligand_pdb_block(pdb_block)` ¶

`REvoDesign.magician.designers.openkinetics._pdb._canonicalize_smiles(smiles)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.resolve_substrate_metadata(pdb_path, structure_id='1SUO')` ¶

`REvoDesign.magician.designers.openkinetics._pdb.load_chain_sequence_context(pdb_path, chain_id='A')` ¶

`REvoDesign.magician.designers.openkinetics._pdb.load_mutation_labels(mutation_path, limit=None)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.relabel_pdb_position_to_sequential(label, residue_numbers, *, chain_id='A')` ¶

Convert a PDB-numbered mutation label to its sequential-position equivalent.

Mutation labels like AE93D use PDB residue numbering which may be non-contiguous. :func:REvoDesign.tools.mutant_tools.extract_mutants_from_mutant_id expects sequential (1-based) positions, so this helper performs the remapping using the residue-number list obtained from :func:load_chain_sequence_context.

Scorer Classes¶

OpenKineticsScorerAbstract¶

Base class for all OpenKinetics API-based kinetic scorers. Concrete subclasses fix the method and prediction_type via the abstract built_in_defaults() class method.

`REvoDesign.magician.designers.openkinetics._scorers.OpenKineticsScorerAbstract` ¶

Bases: ExternalDesignerAbstract, ABC

Base class for OpenKinetics API-based kinetic scorers.

Concrete subclasses fix the method and prediction type via :meth:built_in_defaults.

`built_in_defaults()` `abstractmethod` `classmethod` ¶

Return {"method": ..., "prediction_type": ...}.

Dynamically Created Scorer Subclasses¶

Rather than defining one class per prediction method, _scorers.py declares a _SCORER_SPECS tuple and creates concrete subclasses with type() at import time. Each subclass sets name, prefer_lower, built_in_defaults, and __bibtex__.

The following 19 scorer classes are generated from the spec:

Class name	Scorer name	Method	Prediction type	Citation key
`CataProKcatScorer`	`OpenKinetics-CataPro-kcat`	CataPro	kcat	CataPro
`CatPredKcatScorer`	`OpenKinetics-CatPred-kcat`	CatPred	kcat	CatPred
`DLKcatScorer`	`OpenKinetics-DLKcat-kcat`	DLKcat	kcat	DLKcat
`EITLEMKcatScorer`	`OpenKinetics-EITLEM-kcat`	EITLEM	kcat	EITLEM
`KinFormHKcatScorer`	`OpenKinetics-KinForm-H-kcat`	KinForm-H	kcat	KinForm
`KinFormLKcatScorer`	`OpenKinetics-KinForm-L-kcat`	KinForm-L	kcat	KinForm
`OmniESIKcatScorer`	`OpenKinetics-OmniESI-kcat`	OmniESI	kcat	OmniESI
`RealKcatScorer`	`OpenKinetics-RealKcat-kcat`	RealKcat	kcat	RealKcat
`UniKPKcatScorer`	`OpenKinetics-UniKP-kcat`	UniKP	kcat	UniKP
`CataProKmScorer`	`OpenKinetics-CataPro-Km`	CataPro	Km	CataPro
`CatPredKmScorer`	`OpenKinetics-CatPred-Km`	CatPred	Km	CatPred
`EITLEMKmScorer`	`OpenKinetics-EITLEM-Km`	EITLEM	Km	EITLEM
`KinFormHKmScorer`	`OpenKinetics-KinForm-H-Km`	KinForm-H	Km	KinForm
`MMISAKMKmScorer`	`OpenKinetics-MMISA-KM-Km`	MMISA-KM	Km	MMISA-KM
`OmniESIKmScorer`	`OpenKinetics-OmniESI-Km`	OmniESI	Km	OmniESI
`RealKcatKmScorer`	`OpenKinetics-RealKcat-Km`	RealKcat	Km	RealKcat
`UniKPKmScorer`	`OpenKinetics-UniKP-Km`	UniKP	Km	UniKP
`CataProKcatKmScorer`	`OpenKinetics-CataPro-kcat/Km`	CataPro	kcat/Km	CataPro
`IECataKcatKmScorer`	`OpenKinetics-IECata-kcat/Km`	IECata	kcat/Km	IECata

Km-targeting scorers set prefer_lower = True (lower scores are better). kcat and kcat/Km scorers set prefer_lower = False (higher scores are better).

Re-exported list of scorer class names¶

`REvoDesign.magician.designers.openkinetics._scorers.OPENKINETICS_SCORER_CLASS_NAMES = tuple((spec[0]) for spec in _SCORER_SPECS)` `module-attribute` ¶

OpenKinetics API Reference¶

Architecture¶

YAML Configuration¶

API Key Configuration¶

Constants¶

REvoDesign.magician.designers.openkinetics._models.DEFAULT_OPENKINETICS_API_KEY_ENV = 'OPENKINETICS_API_KEY' module-attribute ¶

REvoDesign.magician.designers.openkinetics._models.OPENKINETICS_ENDPOINTS = {'health': '/health/', 'methods': '/methods/', 'validate': '/validate/', 'submit': '/submit/', 'status': '/status/{job_id}/', 'result': '/result/{job_id}/', 'quota': '/quota/'} module-attribute ¶

REvoDesign.magician.designers.openkinetics._models.WATER_RESIDUE_NAMES = frozenset({'HOH', 'WAT'}) module-attribute ¶

REvoDesign.magician.designers.openkinetics._models.COFACTOR_EXCLUSIONS = frozenset({'HEM'}) module-attribute ¶

Exception Hierarchy¶

REvoDesign.magician.designers.openkinetics._models.OpenKineticsError ¶

REvoDesign.magician.designers.openkinetics._models.OpenKineticsConfigurationError ¶

REvoDesign.magician.designers.openkinetics._models.OpenKineticsAPIError ¶

REvoDesign.magician.designers.openkinetics._models.OpenKineticsTimeoutError ¶

REvoDesign.magician.designers.openkinetics._models.OpenKineticsValidationError ¶

Dataclasses¶

REvoDesign.magician.designers.openkinetics._models.LigandCandidate dataclass ¶

REvoDesign.magician.designers.openkinetics._models.OpenKineticsFixturePaths dataclass ¶

Configuration Functions¶

REvoDesign.magician.designers.openkinetics._client.load_openkinetics_config() ¶

REvoDesign.magician.designers.openkinetics._client.resolve_api_key(*, api_key=None, auto_register=False, replace_existing=False, base_url=None, timeout_seconds=None, session=None) ¶

REvoDesign.magician.designers.openkinetics._client.fetch_openkinetics_api_key(*, base_url=None, timeout_seconds=None, session=None, replace_existing=False) ¶

REvoDesign.magician.designers.openkinetics._client.persist_openkinetics_api_key(api_key) ¶

Data I/O Helpers¶

REvoDesign.magician.designers.openkinetics._client.build_openkinetics_data_rows(variant_rows, substrate_smiles) ¶

REvoDesign.magician.designers.openkinetics._client.build_openkinetics_request_payload(*, data_rows, method_metadata, method, prediction_type) ¶

REvoDesign.magician.designers.openkinetics._client.get_method_metadata(methods_response, *, method, prediction_type) ¶

REvoDesign.magician.designers.openkinetics._client.write_csv_rows(path, rows) ¶

REvoDesign.magician.designers.openkinetics._client.write_json(path, payload) ¶

REvoDesign.magician.designers.openkinetics._client.sha256_file(path) ¶

REvoDesign.magician.designers.openkinetics._client.write_normalized_scores_csv(path, rows) ¶

REvoDesign.magician.designers.openkinetics._client._normalize_result_rows(result_payload, *, method, prediction_type, substrate_smiles, variant_rows=None, job_id='') ¶

OpenKineticsClient¶

REvoDesign.magician.designers.openkinetics._client.OpenKineticsClient ¶

list_methods() ¶

validate_file(csv_path, *, run_similarity=False) ¶

validate(rows, *, run_similarity=False) ¶

check_health() ¶

check_quota() ¶

submit(rows, method, prediction_type, *, handle_long_sequences='truncate', use_experimental=False, include_similarity_columns=True, canonicalize_substrates=True) ¶

get_result(job_id, result_format='json', *, poll_interval_seconds=3, timeout_seconds=None) ¶

poll_until_complete(job_id, *, poll_interval_seconds=3, timeout_seconds=600) ¶

PDB / Ligand Functions¶

REvoDesign.magician.designers.openkinetics._pdb.discover_ligand_candidates(pdb_path, excluded_residue_names=WATER_RESIDUE_NAMES | COFACTOR_EXCLUSIONS) ¶

REvoDesign.magician.designers.openkinetics._pdb.choose_primary_ligand(pdb_path) ¶

REvoDesign.magician.designers.openkinetics._pdb.extract_ligand_pdb_block(pdb_path, ligand) ¶

REvoDesign.magician.designers.openkinetics._pdb.smiles_from_ligand_pdb_block(pdb_block) ¶

REvoDesign.magician.designers.openkinetics._pdb._canonicalize_smiles(smiles) ¶

REvoDesign.magician.designers.openkinetics._pdb.resolve_substrate_metadata(pdb_path, structure_id='1SUO') ¶

REvoDesign.magician.designers.openkinetics._pdb.load_chain_sequence_context(pdb_path, chain_id='A') ¶

REvoDesign.magician.designers.openkinetics._pdb.load_mutation_labels(mutation_path, limit=None) ¶

REvoDesign.magician.designers.openkinetics._pdb.relabel_pdb_position_to_sequential(label, residue_numbers, *, chain_id='A') ¶

Scorer Classes¶

OpenKineticsScorerAbstract¶

REvoDesign.magician.designers.openkinetics._scorers.OpenKineticsScorerAbstract ¶

built_in_defaults() abstractmethod classmethod ¶

Dynamically Created Scorer Subclasses¶

Re-exported list of scorer class names¶

REvoDesign.magician.designers.openkinetics._scorers.OPENKINETICS_SCORER_CLASS_NAMES = tuple((spec[0]) for spec in _SCORER_SPECS) module-attribute ¶

`REvoDesign.magician.designers.openkinetics._models.DEFAULT_OPENKINETICS_API_KEY_ENV = 'OPENKINETICS_API_KEY'` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.OPENKINETICS_ENDPOINTS = {'health': '/health/', 'methods': '/methods/', 'validate': '/validate/', 'submit': '/submit/', 'status': '/status/{job_id}/', 'result': '/result/{job_id}/', 'quota': '/quota/'}` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.WATER_RESIDUE_NAMES = frozenset({'HOH', 'WAT'})` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.COFACTOR_EXCLUSIONS = frozenset({'HEM'})` `module-attribute` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsError` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsConfigurationError` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsAPIError` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsTimeoutError` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsValidationError` ¶

`REvoDesign.magician.designers.openkinetics._models.LigandCandidate` `dataclass` ¶

`REvoDesign.magician.designers.openkinetics._models.OpenKineticsFixturePaths` `dataclass` ¶

`REvoDesign.magician.designers.openkinetics._client.load_openkinetics_config()` ¶

`REvoDesign.magician.designers.openkinetics._client.resolve_api_key(*, api_key=None, auto_register=False, replace_existing=False, base_url=None, timeout_seconds=None, session=None)` ¶

`REvoDesign.magician.designers.openkinetics._client.fetch_openkinetics_api_key(*, base_url=None, timeout_seconds=None, session=None, replace_existing=False)` ¶

`REvoDesign.magician.designers.openkinetics._client.persist_openkinetics_api_key(api_key)` ¶

`REvoDesign.magician.designers.openkinetics._client.build_openkinetics_data_rows(variant_rows, substrate_smiles)` ¶

`REvoDesign.magician.designers.openkinetics._client.build_openkinetics_request_payload(*, data_rows, method_metadata, method, prediction_type)` ¶

`REvoDesign.magician.designers.openkinetics._client.get_method_metadata(methods_response, *, method, prediction_type)` ¶

`REvoDesign.magician.designers.openkinetics._client.write_csv_rows(path, rows)` ¶

`REvoDesign.magician.designers.openkinetics._client.write_json(path, payload)` ¶

`REvoDesign.magician.designers.openkinetics._client.sha256_file(path)` ¶

`REvoDesign.magician.designers.openkinetics._client.write_normalized_scores_csv(path, rows)` ¶

`REvoDesign.magician.designers.openkinetics._client._normalize_result_rows(result_payload, *, method, prediction_type, substrate_smiles, variant_rows=None, job_id='')` ¶

`REvoDesign.magician.designers.openkinetics._client.OpenKineticsClient` ¶

`list_methods()` ¶

`validate_file(csv_path, *, run_similarity=False)` ¶

`validate(rows, *, run_similarity=False)` ¶

`check_health()` ¶

`check_quota()` ¶

`submit(rows, method, prediction_type, *, handle_long_sequences='truncate', use_experimental=False, include_similarity_columns=True, canonicalize_substrates=True)` ¶

`get_result(job_id, result_format='json', *, poll_interval_seconds=3, timeout_seconds=None)` ¶

`poll_until_complete(job_id, *, poll_interval_seconds=3, timeout_seconds=600)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.discover_ligand_candidates(pdb_path, excluded_residue_names=WATER_RESIDUE_NAMES | COFACTOR_EXCLUSIONS)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.choose_primary_ligand(pdb_path)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.extract_ligand_pdb_block(pdb_path, ligand)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.smiles_from_ligand_pdb_block(pdb_block)` ¶

`REvoDesign.magician.designers.openkinetics._pdb._canonicalize_smiles(smiles)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.resolve_substrate_metadata(pdb_path, structure_id='1SUO')` ¶

`REvoDesign.magician.designers.openkinetics._pdb.load_chain_sequence_context(pdb_path, chain_id='A')` ¶

`REvoDesign.magician.designers.openkinetics._pdb.load_mutation_labels(mutation_path, limit=None)` ¶

`REvoDesign.magician.designers.openkinetics._pdb.relabel_pdb_position_to_sequential(label, residue_numbers, *, chain_id='A')` ¶

`REvoDesign.magician.designers.openkinetics._scorers.OpenKineticsScorerAbstract` ¶

`built_in_defaults()` `abstractmethod` `classmethod` ¶

`REvoDesign.magician.designers.openkinetics._scorers.OPENKINETICS_SCORER_CLASS_NAMES = tuple((spec[0]) for spec in _SCORER_SPECS)` `module-attribute` ¶