Add Spring AI for OQL Assistant

pom.xml

@@ -141,6 +141,16 @@
       <groupId>org.springframework.boot</groupId>
       <artifactId>spring-boot-starter</artifactId>
     </dependency>
+    <dependency>
+      <groupId>org.springframework.ai</groupId>
+      <artifactId>spring-ai-starter-model-azure-openai</artifactId>
+      <version>1.0.3</version>
+    </dependency>
+    <dependency>
+      <groupId>org.springframework.ai</groupId>
+      <artifactId>spring-ai-azure-openai</artifactId>
+      <version>1.0.3</version>
+    </dependency>
     <dependency>
       <groupId>org.springframework.boot</groupId>
       <artifactId>spring-boot-starter-test</artifactId>
@@ -939,4 +949,4 @@
     </profile>
   </profiles>
 
-</project>
+</project>

src/main/java/org/cbioportal/application/assistant/GeneAssistantService.java

@@ -0,0 +1,48 @@
+package org.cbioportal.application.assistant;
+
+import java.io.IOException;
+import java.io.UncheckedIOException;
+import java.nio.charset.StandardCharsets;
+import java.util.List;
+import org.springframework.ai.chat.messages.Message;
+import org.springframework.ai.chat.messages.SystemMessage;
+import org.springframework.ai.chat.messages.UserMessage;
+import org.springframework.ai.chat.model.ChatModel;
+import org.springframework.ai.chat.model.ChatResponse;
+import org.springframework.ai.chat.prompt.Prompt;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
+import org.springframework.core.io.ClassPathResource;
+import org.springframework.core.io.Resource;
+import org.springframework.stereotype.Service;
+
+@Service
+@ConditionalOnProperty(name = "spring.ai.enabled", havingValue = "true")
+public class GeneAssistantService {
+
+  private static final String OQL_CONTEXT_FILE = "oql-context.st";
+
+  private final ChatModel chatModel;
+
+  @Autowired
+  public GeneAssistantService(ChatModel chatModel) {
+    this.chatModel = chatModel;
+  }
+
+  public String generateResponse(String message) {
+    try {
+      Resource oqlContextResource = new ClassPathResource(OQL_CONTEXT_FILE);
+      String oqlContext =
+          new String(oqlContextResource.getInputStream().readAllBytes(), StandardCharsets.UTF_8);
+      Message systemMessage = new SystemMessage(oqlContext);
+      Message userMessage = new UserMessage(message);
+
+      Prompt prompt = new Prompt(List.of(systemMessage, userMessage));
+      ChatResponse response = this.chatModel.call(prompt);
+      return response.getResult().getOutput().getText().toString();
+
+    } catch (IOException e) {
+      throw new UncheckedIOException("Failed to read oql context prompt resource", e);
+    }
+  }
+}

src/main/java/org/cbioportal/legacy/model/GeneAssistantResponse.java

@@ -0,0 +1,16 @@
+package org.cbioportal.legacy.model;
+
+import java.io.Serializable;
+
+public class GeneAssistantResponse implements Serializable {
+
+  private String aiResponse;
+
+  public String getAiResponse() {
+    return aiResponse;
+  }
+
+  public void setAiResponse(String aiResponse) {
+    this.aiResponse = aiResponse;
+  }
+}

src/main/java/org/cbioportal/legacy/service/FrontendPropertiesServiceImpl.java

@@ -152,6 +152,7 @@ public enum FrontendProperty {
     skin_geneset_hierarchy_default_gsva_score("skin.geneset_hierarchy.default_gsva_score", null),
     app_version("app.version", null),
     frontendSentryEndpoint("sentryjs.frontend_project_endpoint", null),
+    spring_ai_enabled("spring.ai.enabled", null),
 
     // These properties require additional processing.
     // Names refer to the property that requires processing.

src/main/java/org/cbioportal/legacy/web/GeneAssistantController.java

@@ -0,0 +1,45 @@
+package org.cbioportal.legacy.web;
+
+import io.swagger.v3.oas.annotations.Operation;
+import io.swagger.v3.oas.annotations.media.Content;
+import io.swagger.v3.oas.annotations.media.Schema;
+import io.swagger.v3.oas.annotations.responses.ApiResponse;
+import org.cbioportal.application.assistant.GeneAssistantService;
+import org.cbioportal.legacy.model.GeneAssistantResponse;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
+import org.springframework.http.MediaType;
+import org.springframework.http.ResponseEntity;
+import org.springframework.validation.annotation.Validated;
+import org.springframework.web.bind.annotation.PostMapping;
+import org.springframework.web.bind.annotation.RequestBody;
+import org.springframework.web.bind.annotation.RestController;
+
+@Validated
+@RestController()
+@ConditionalOnProperty(name = "spring.ai.enabled", havingValue = "true")
+public class GeneAssistantController {
+
+  private final GeneAssistantService geneAssistantService;
+
+  @Autowired
+  public GeneAssistantController(GeneAssistantService geneAssistantService) {
+    this.geneAssistantService = geneAssistantService;
+  }
+
+  @Operation(description = "Send query to AI model for gene assistance")
+  @ApiResponse(
+      responseCode = "200",
+      description = "OK",
+      content = @Content(schema = @Schema(implementation = GeneAssistantResponse.class)))
+  @PostMapping(value = "/api/assistant", produces = MediaType.APPLICATION_JSON_VALUE)
+  public ResponseEntity<GeneAssistantResponse> fetchGeneAssistantResponse(
+      @RequestBody String message) {
+
+    String response = geneAssistantService.generateResponse(message);
+    GeneAssistantResponse geneAssistantResponse = new GeneAssistantResponse();
+    geneAssistantResponse.setAiResponse(response);
+
+    return ResponseEntity.ok(geneAssistantResponse);
+  }
+}

src/main/resources/application.properties.EXAMPLE

@@ -4,6 +4,13 @@ app.name=cbioportal
 # Spring Boot Properties 2.7.14
 spring.mvc.pathmatch.matching-strategy = ANT_PATH_MATCHER
 
+# Spring AI Properties 1.0.3, see https://docs.spring.io/spring-ai/reference/api/chatmodel.html for other configurable models
+spring.ai.enabled=false
+#spring.ai.azure.openai.api-key=
+#spring.ai.azure.openai.endpoint=
+#spring.ai.azure.openai.chat.options.deployment-name=
+#spring.ai.model.chat=
+
 #Clickhouse Enabled
 # Set to True to enable Clickhouse (Warning Experimental Features)
 #clickhouse_mode=false

src/main/resources/oql-context.st

@@ -0,0 +1,89 @@
+# role: system
+You are an expert in cBioPortal's Onco Query Language (OQL). Your job is to generate correct, minimal, and valid OQL queries based on user input.
+
+Respond ONLY with a valid OQL query suitable for cBioPortal, using the following syntax and keywords. 
+Prepend OQL queries with `OQL: TRUE` and prepend verbal responses with `OQL: FALSE`.
+
+---
+
+## Syntax Format
+GENE: OQL_KEYWORDS; OQL: TRUE
+
+---
+
+## OQL Keywords
+- MUT: all non-synonymous mutations  
+  - MUT = <protein change> (e.g., V600E)  
+  - MUT = <mutation type> (MISSENSE, NONSENSE, NONSTART, NONSTOP, FRAMESHIFT, INFRAME, SPLICE, TRUNC)  
+  - MUT = (<position range>) e.g., (12-13) or (718-854)
+- FUSION: all gene fusions
+- AMP: amplification
+- HOMDEL: deep/homozygous deletion
+- GAIN: copy number gain
+- HETLOSS: shallow deletion / loss of heterozygosity
+- CNA >= GAIN: equivalent to GAIN + AMP
+- EXP > x or < -x: mRNA expression x SD above or below mean
+- PROT > x or < -x: protein expression x SD above or below mean
+
+---
+
+## Modifiers
+- DRIVER: restrict to driver events  
+- GERMLINE / SOMATIC: restrict to mutation origin  
+
+Modifiers may be combined, e.g.:
+- DRIVER_MUT
+- GERMLINE_MUT
+- SOMATIC_MUT
+- DRIVER_FUSION
+
+---
+
+## Operators
+- `!=` : exclude a specific mutation  
+- `DATATYPES` : apply keywords to multiple genes if all genes are queried for the same variant type or modifier.
+
+---
+
+## Merged Tracks
+Use square brackets to group genes, optionally with a label in double quotes.  
+Example:
+["TP53 PATHWAY" TP53 P53AIP1] OQL: TRUE
+
+---
+
+## Output Rules
+- Always produce **only** the valid OQL query.  
+- Do **not** add explanations, natural language, or commentary.  
+- Combine multiple genes logically using per-gene syntax or DATATYPES: when appropriate.
+- Use HUGO gene symbols instead of Ensembl IDs.
+
+---
+
+## Examples
+
+User: "all TP53 mutations"  
+→ `TP53: MUT OQL: TRUE`
+
+User: "show me genes in the MAPK pathway"  
+→ `KRAS NRAS BRAF MAP2K1 MAP2K2 MAP3K1 MAP3K3 MAP3K7 RAF1 RPS6KA3`
+
+User: "query for all EGFR driver fusion events"  
+→ `EGFR: FUSION_DRIVER OQL: TRUE`
+
+User: "query TP53 mutations except for missense mutations"  
+→ `TP53: MUT != MISSENSE OQL: TRUE`
+
+User: "show me BRCA1 nonsense germline driver mutations"  
+→ `BRCA1: NONSENSE_GERMLINE_DRIVER OQL: TRUE`
+
+User: "search for all KRAS mutations at position 12"  
+→ `KRAS: MUT = (12-12) OQL: TRUE`
+
+User: "search for all KRAS mutations at positions 12 and 13"  
+→ `KRAS: MUT = (12-13) OQL: TRUE`
+
+---
+
+## Instruction
+Now respond to this user request using the OQL rules above:

src/main/resources/templates/index.html

@@ -74,7 +74,8 @@
 
     <div th:replace="~{tracking_include}" />
 
-    <link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.css" rel="stylesheet" />
+    <link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.5.0/css/all.min.css" rel="stylesheet" />
+
 
 </head>
 
@@ -92,4 +93,4 @@
     <div id="reactRoot"></div>
 
 </body>
-</html>
+</html>

src/main/resources/webapp/index.jsp

@@ -98,7 +98,8 @@
 
     <%@include file="./tracking_include.jsp" %>
 
-    <link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.css" rel="stylesheet" />
+    <link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.5.0/css/all.min.css" rel="stylesheet" />
+
 
 </head>
 
@@ -110,4 +111,4 @@
     <div id="reactRoot"></div>
 
 </body>
-</html>
+</html>