Testing new OpenAI functions as a potential SPIRES augmentation.

src/ontogpt/engines/spires2_engine.py

@@ -0,0 +1,79 @@
+"""
+Rewrite of SPIRES to use OpenAI function feature.
+
+See https://github.com/monarch-initiative/ontogpt/issues/132
+"""
+import json
+import logging
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple, Union
+
+import openai
+from linkml_runtime.linkml_model import ClassDefinition
+
+from ontogpt.engines.knowledge_engine import (
+    ANNOTATION_KEY_PROMPT,
+    ANNOTATION_KEY_PROMPT_SKIP,
+    EXAMPLE,
+    FIELD,
+    OBJECT,
+    KnowledgeEngine,
+    chunk_text,
+)
+from ontogpt.templates.core import ExtractionResult
+
+this_path = Path(__file__).parent
+
+
+logger = logging.getLogger(__name__)
+
+
+@dataclass
+class SPIRES2Engine(KnowledgeEngine):
+    """Knowledge extractor."""
+
+    model: str = "gpt-3.5-turbo-0613"
+
+    def extract_from_text(
+        self, text: str, cls: ClassDefinition = None, object: OBJECT = None
+    ) -> ExtractionResult:
+        """
+        Extract annotations from the given text.
+
+        :param text:
+        :param cls:
+        :param object: optional stub object
+        :return:
+        """
+        if cls is None:
+            [cls] = [c for c in self.schemaview.all_classes().values() if c.tree_root]
+        py_cls = self.template_module.__dict__[cls.name]
+        schema = py_cls.schema()
+        functions = [
+            {
+                "name": "extract_data",
+                #"description":  cls.description,
+                "description":  "paper",
+                "parameters": schema,
+            },
+        ]
+        # TODO: introspect schema to customize system content
+        messages = [
+            {"role": "system",
+             "content": "You are a helpful assistant that extracts summaries from text as JSON for a database."},
+            {"role": "user",
+             "content": 'Extract a summary from the following text: ' + text},
+        ]
+        logger.info(json.dumps(functions, indent=2))
+        # TODO: abstract this so as not hardcoded
+        response = openai.ChatCompletion.create(
+            model=self.model, functions=functions, messages=messages)
+        logger.info(f"Response: {response}")
+        r = response.choices[0]['message']['function_call']['arguments']
+        extracted_object = py_cls(**json.loads(r))
+        return ExtractionResult(
+            input_text=text,
+            extracted_object=extracted_object,
+            named_entities=self.named_entities,
+        )

src/ontogpt/models.yaml

@@ -8,6 +8,22 @@ models:
       - OpenAI
     is_default: true
 
+  - name: MODEL_GPT_3_5_TURBO
+    alternative_names:
+      - "gpt-3.5-turbo-0613"
+      - "openai-gpt-3.5-turbo-0613"
+    provider: OpenAI
+    creators:
+      - OpenAI
+
+  - name: MODEL_GPT_3_5_TURBO_16K
+    alternative_names:
+      - "gpt-3.5-turbo-16k"
+      - "openai-gpt-3.5-turbo-16k"
+    provider: OpenAI
+    creators:
+      - OpenAI
+
   - name: MODEL_TEXT_DAVINCI_003
     alternative_names:
       - "text-davinci-003"

tests/integration/test_knowledge_engines/test_spires2_engine.py

@@ -0,0 +1,151 @@
+"""Core tests."""
+import unittest
+
+import yaml
+from linkml_runtime.linkml_model import ClassDefinitionName
+from oaklib import get_implementation_from_shorthand
+
+from ontogpt.clients.pubmed_client import PubmedClient
+from ontogpt.engines import create_engine
+from ontogpt.engines.knowledge_engine import chunk_text
+from ontogpt.engines.spires2_engine import SPIRES2Engine
+from ontogpt.engines.spires_engine import SPIRESEngine
+from ontogpt.io.yaml_wrapper import dump_minimal_yaml
+from ontogpt.templates.biological_process import BiologicalProcess
+from ontogpt.templates.gocam import (
+    ExtractionResult,
+    Gene,
+    GeneLocation,
+    GeneOrganismRelationship,
+    GoCamAnnotations,
+)
+
+TEMPLATE = "gocam.GoCamAnnotations"
+
+PAPER = """
+Title: β-Catenin Is Required for the cGAS/STING Signaling Pathway but
+       Antagonized by the Herpes Simplex Virus 1 US3 Protein.
+Text:
+The cGAS/STING-mediated DNA-sensing signaling pathway is crucial
+for interferon (IFN) production and host antiviral
+responses. Herpes simplex virus I (HSV-1) is a DNA virus that has
+evolved multiple strategies to evade host immune responses. Here,
+we demonstrate that the highly conserved β-catenin protein in the
+Wnt signaling pathway is an important factor to enhance the
+transcription of type I interferon (IFN-I) in the cGAS/STING
+signaling pathway, and the production of IFN-I mediated by
+β-catenin was antagonized by HSV-1 US3 protein via its kinase
+activity. Infection by US3-deficienct HSV-1 and its kinase-dead
+variants failed to downregulate IFN-I and IFN-stimulated
+gene (ISG) production induced by β-catenin. Consistent with this,
+absence of β-catenin enhanced the replication of US3-deficienct
+HSV-1, but not wild-type HSV-1. The underlying mechanism was the
+interaction of US3 with β-catenin and its hyperphosphorylation of
+β-catenin at Thr556 to block its nuclear translocation. For the
+first time, HSV-1 US3 has been shown to inhibit IFN-I production
+through hyperphosphorylation of β-catenin and to subvert host
+antiviral innate immunity.IMPORTANCE Although increasing evidence
+has demonstrated that HSV-1 subverts host immune responses and
+establishes lifelong latent infection, the molecular mechanisms
+by which HSV-1 interrupts antiviral innate immunity, especially
+the cGAS/STING-mediated cellular DNA-sensing signaling pathway,
+have not been fully explored. Here, we show that β-catenin
+promotes cGAS/STING-mediated activation of the IFN pathway, which
+is important for cellular innate immune responses and intrinsic
+resistance to DNA virus infection. The protein kinase US3
+antagonizes the production of IFN by targeting β-catenin via its
+kinase activity. The findings in this study reveal a novel
+mechanism for HSV-1 to evade host antiviral immunity and add new
+knowledge to help in understanding the interaction between the
+host and HSV-1 infection.
+
+Keywords: HSV-1; US3; type I IFN; β-catenin.
+"""
+
+EXAMPLE_RESULTS = """
+genes: β-Catenin; cGAS; STING; US3; IFN; ISG
+organisms: Herpes Simplex Virus I (HSV-1);
+gene_organisms: β-Catenin:host; cGAS:host; STING:host; US3:HSV-1; IFN:host; ISG:host
+activities: production of type I IFN; transcription of type I IFN; replication of HSV-1;
+nuclear translocation of β-catenin.
+gene_functions: β-catenin:enhance the transcription of type I IFN; US3:antagonize
+the production of IFN; β-catenin:block nuclear translocation.
+cellular_processes: cGAS/STING-mediated DNA-sensing signaling; activation of IFN pathway
+pathways: IFN pathway; Wnt signalling pathway
+gene_gene_interactions: US3:β-catenin
+gene_localizations: US3:host; β-catenin:host
+"""
+
+EXAMPLE_RESULTS_ALT = """
+genes: β-Catenin; cGAS; STING; US3; IFN; ISG
+organisms: Herpes Simplex Virus I (HSV-1);
+gene_organisms: β-Catenin - Human; cGAS - Human; STING - Human;
+US3 - Human; IFN - Human; ISG - Human.
+activities: Transcription; Production; Downregulation; Replication; Nuclear Translocation
+gene_functions: β-Catenin - Enhances Transcription; US3 - Antagonizes Production;
+US3 - Downregulates IFN-I; US3 - Blocks Nuclear Translocation; β-Catenin - Enhances Production
+cellular_processes: DNA-sensing; Interferon Production; Antiviral Innate Immunity;
+Host Innate Immune Responses; Interaction with Host; Evade Host Antiviral Immunity
+pathways: cGAS/STING-mediated DNA-sensing; Wnt Signaling; IFN pathway
+gene_gene_interactions: US3 - β-Catenin; β-Catenin - US3
+gene_localizations: β-Catenin - Nuclear; US3 - Hyperphosphorylation
+"""
+
+TEST_PROCESS = BiologicalProcess(
+    label="autophagosome assembly",
+    description="The formation of a double membrane-bounded structure, the autophagosome,\
+        that occurs when a specialized membrane sac, called the isolation membrane,\
+        starts to enclose a portion of the cytoplasm",
+    subclass_of="GO:0022607",
+    outputs=["GO:0005776"],
+)
+
+DIRECT_PARSE = {
+    "genes": ["β-Catenin", "cGAS", "STING", "US3", "IFN", "ISG"],
+    "gene_organisms": [
+        ("β-Catenin", "host"),
+        ("cGAS", "host"),
+        ("STING", "host"),
+        ("US3", "HSV-1"),
+        ("IFN", "host"),
+        ("ISG", "host"),
+    ],
+}
+
+
+class TestCore(unittest.TestCase):
+    """Test annotation."""
+
+    def setUp(self) -> None:
+        """Set up."""
+        #self.ke = create_engine(TEMPLATE, SPIRESEngine)
+        self.ke = SPIRES2Engine(template=TEMPLATE)
+
+    def test_setup(self):
+        """Tests template and module is loaded."""
+        ke = self.ke
+        pyc = ke.template_pyclass
+        print(pyc)
+        obj = pyc(genes=["a"], gene_organisms=[{"gene": "a", "organism": "b"}])
+        print(yaml.dump(obj.dict()))
+        self.assertEqual(obj.genes, ["a"])
+        self.assertEqual(obj.gene_organisms[0].gene, "a")
+        self.assertEqual(obj.gene_organisms[0].organism, "b")
+        slot = ke.schemaview.induced_slot("gene", "GeneOrganismRelationship")
+        self.assertEqual(slot.name, "gene")
+        self.assertIsNone(slot.multivalued)
+        self.assertEqual(slot.range, "Gene")
+
+
+    def test_extract(self):
+        """Tests end to end knowledge extraction."""
+        ke = self.ke
+        ann = ke.extract_from_text(PAPER)
+        print(f"RESULTS={ann}")
+        print(yaml.dump(ann.dict()))
+        results = ann.extracted_object
+        print(results)
+        if not isinstance(results, GoCamAnnotations):
+            raise ValueError(f"Expected GoCamAnnotations, got {type(results)}")
+        self.assertIn("HGNC:2514", results.genes)
+

tests/integration/test_knowledge_engines/test_spires_engine.py

@@ -129,9 +129,9 @@ def test_setup(self):
         self.assertEqual(obj.genes, ["a"])
         self.assertEqual(obj.gene_organisms[0].gene, "a")
         self.assertEqual(obj.gene_organisms[0].organism, "b")
-        slot = ke.schemaview.induced_slot("genes", "GeneOrganismRelationship")
-        self.assertEqual(slot.name, "genes")
-        self.assertEqual(slot.multivalued, True)
+        slot = ke.schemaview.induced_slot("gene", "GeneOrganismRelationship")
+        self.assertEqual(slot.name, "gene")
+        self.assertIsNone(slot.multivalued)
         self.assertEqual(slot.range, "Gene")
 
     def test_chunk_text(self):