openprotein.prompt

Create prompts to be used with PoET models, along with queries which opens up use-cases like inverse folding with PoET-2.

Interface

class openprotein.prompt.PromptAPI(session)

Prompt API providing the interface to create prompts for use with PoET models.

create_prompt(context, name=None, description=None)

Create a prompt.

Parameters:

• context (Context | Sequence[Context]) – Context or list of contexts. Each context is a sequence of entries where each entry is a raw sequence (bytes/str, optionally with : chain breaks for multichain), Protein, or Complex. Currently only protein chains are accepted; passing a Complex with DNA, RNA, or Ligand chains raises InvalidParameterError. This restriction may be relaxed in the future.

• name (str) – Name of the prompt.

• description (Optional[str]) – Description of the prompt.

Returns:

The created prompt.

Return type:

Prompt

get_prompt(prompt_id)

Get the prompt for a given prompt ID.

Parameters:

prompt_id (str) – The prompt ID.

Returns:

The prompt.

Return type:

Prompt

list_prompts()

List all prompts.

Returns:

List of prompts.

Return type:

List[Prompt]

create_query(query, force_structure=False)

Create a query.

Parameters:

• query (bytes or str or Protein or Complex) – A query protein or complex. Raw bytes/str inputs may include : chain breaks to denote a multichain protein. Currently only protein chains are accepted; passing a Complex with DNA, RNA, or Ligand chains raises InvalidParameterError. This restriction may be relaxed in the future.

• force_structure (bool, optional) – Optionally force a query to be interpreted with a structure. Useful for creating structure prediction queries which can have no structure.

Returns:

The created query.

Return type:

Query

get_query(query_id)

Get the query for a given query ID.

Parameters:

query_id (str) – The query ID.

Returns:

The query.

Return type:

Query

Classes

class openprotein.prompt.Prompt(session, job=None, metadata=None, num_replicates=None)

Prompt which contains a set of sequences and/or structures used to condition the PoET models.

get_as_complexes()

Retrieve the prompt context with every entry as a Complex.

Single-chain entries are wrapped as Complex({"A": protein}) so the return type is uniform regardless of chain count.

get_as_proteins()

Retrieve the prompt context with every entry as a Protein.

Raises InvalidParameterError if any entry is multichain — use get_as_complexes() instead when multichain entries may be present.

property id

The unique identifier of the prompt.

property name

The name of the prompt.

property description

The description of the prompt.

property created_date

The timestamp when the prompt was created.

property num_replicates

The number of replicates in the prompt for an ensemble prompt.

property status

The status of the prompt if sampling from an MSAFuture.

property args: dict[str, Any]

The registered job arguments.

cancelled()

Check if the job has been cancelled.

Returns:

True if the job is cancelled, False otherwise.

Return type:

bool

done()

Check if the job has completed.

Returns:

True if the job is done, False otherwise.

Return type:

bool

property end_date: datetime | None

The end timestamp of the job.

property job_id: str

The unique identifier of the job.

property job_type: str

The type of the job.

property progress_counter: int

The progress counter of the job.

refresh()

Refresh the job status and internal job object.

property start_date: datetime | None

The start timestamp of the job.

wait(interval=5, timeout=None, verbose=False)

Wait for the job to complete, then fetch results.

Parameters:

• interval (int, optional) – Time in seconds between polling. Defaults to config.POLLING_INTERVAL.

• timeout (int | None, optional) – Maximum time in seconds to wait. Defaults to None.

• verbose (bool, optional) – Verbosity flag. Defaults to False.

Returns:

The results of the job.

Return type:

Any

wait_until_done(interval=5, timeout=None, verbose=False)

Wait for the job to complete.

Parameters:

• interval (float, optional) – Time in seconds between polling. Defaults to config.POLLING_INTERVAL.

• timeout (int, optional) – Maximum time in seconds to wait. Defaults to None.

• verbose (bool, optional) – Verbosity flag. Defaults to False.

Returns:

True if the job completed successfully.

Return type:

bool

Notes

This method does not fetch the job results, unlike wait().

class openprotein.prompt.Query(session, metadata)

Query containing a sequence/structure used to query the design models which opens up new workflows.

Create a query with a masked sequence using mask_sequence_at() for PoET2Model to run inverse folding.

Create a query with a masked structure using mask_structure_at() for RFdiffusionModel to run inverse folding.

get()

Retrieve the query as a Protein or Complex.

Single-chain queries collapse to Protein; multichain queries are returned as Complex. For a uniform return type, see get_as_complex() or get_as_protein().

Returns:

Protein or Complex representing the query

Return type:

Protein | Complex

get_as_complex()

Retrieve the query as a Complex.

A single-chain Protein result is wrapped as Complex({"A": protein}) so the return type is uniform.

get_as_protein()

Retrieve the query as a Protein.

Raises InvalidParameterError if the query is multichain — use get_as_complex() instead when multichain queries may be present.

property id

The unique identifier of the query.

property created_date

The timestamp when the query was created.

class openprotein.prompt.PromptMetadata(*, id, name, description=None, created_date, num_replicates, job_id=None, status)

Metadata about a prompt.

class openprotein.prompt.QueryMetadata(*, id, created_date)

Metadata about a query.

class openprotein.prompt.PromptJob(*, job_id, job_type, status, created_date, start_date=None, end_date=None, prerequisite_job_id=None, progress_message=None, progress_counter=None, sequence_length=None, failure_message=None, **extra_data)

A representation of a prompt job.

property msa_id

ID of the underlying MSA.

property prompt_id

Prompt ID.