agentic-flow 1.9.0 → 1.9.2

package/agentic-flow/examples/crispr-cas13-pipeline/docs/architecture-diagrams/sequence-diagrams.md

@@ -1,499 +0,0 @@
-# Sequence Diagrams - Key Workflows
-
-## 1. User Authentication Flow
-
-```mermaid
-sequenceDiagram
-    actor User
-    participant Web as Web App
-    participant API as API Gateway
-    participant Redis as Redis
-    participant DB as PostgreSQL
-
-    User->>Web: Enter email & password
-    Web->>API: POST /api/v1/auth/login
-    activate API
-
-    API->>DB: SELECT user WHERE email = ?
-    activate DB
-    DB-->>API: User record
-    deactivate DB
-
-    API->>API: Verify password (bcrypt)
-
-    alt Password Valid
-        API->>API: Generate JWT token
-        API->>Redis: SET session:{token}
-        activate Redis
-        Redis-->>API: OK
-        deactivate Redis
-
-        API->>DB: INSERT INTO audit_log (login)
-        API-->>Web: 200 OK {token, user}
-        deactivate API
-        Web->>Web: Store token in localStorage
-        Web-->>User: Redirect to dashboard
-    else Password Invalid
-        API-->>Web: 401 Unauthorized
-        deactivate API
-        Web-->>User: Show error message
-    end
-```
-
----
-
-## 2. Experiment Creation & Sample Upload
-
-```mermaid
-sequenceDiagram
-    actor User
-    participant Web as Web App
-    participant API as API Gateway
-    participant DB as PostgreSQL
-    participant MinIO as MinIO S3
-
-    User->>Web: Fill experiment form
-    User->>Web: Upload FASTQ files (R1, R2)
-
-    Web->>API: POST /api/v1/experiments
-    activate API
-    API->>API: Validate JWT token
-    API->>DB: INSERT INTO experiments
-    activate DB
-    DB-->>API: experiment_id
-    deactivate DB
-
-    loop For each sample
-        Web->>API: POST /api/v1/samples/upload
-        activate API
-        API->>MinIO: PUT raw-data/{exp_id}/{sample}_R1.fastq.gz
-        activate MinIO
-        MinIO-->>API: Upload complete
-        deactivate MinIO
-
-        API->>DB: INSERT INTO samples
-        activate DB
-        DB-->>API: sample_id
-        deactivate DB
-        API-->>Web: 201 Created {sample_id}
-        deactivate API
-    end
-
-    API-->>Web: 201 Created {experiment_id}
-    deactivate API
-    Web-->>User: Show success message
-```
-
----
-
-## 3. Job Submission & Processing (Alignment)
-
-```mermaid
-sequenceDiagram
-    actor User
-    participant Web as Web App
-    participant API as API Gateway
-    participant DB as PostgreSQL
-    participant Kafka as Apache Kafka
-    participant Worker as Alignment Worker
-    participant MinIO as MinIO S3
-
-    User->>Web: Click "Start Analysis"
-    Web->>API: POST /api/v1/jobs {type: "alignment"}
-    activate API
-
-    API->>DB: INSERT INTO analysis_jobs (status: "queued")
-    activate DB
-    DB-->>API: job_id
-    deactivate DB
-
-    API->>Kafka: Publish to "jobs.alignment"
-    activate Kafka
-    Kafka-->>API: ACK
-    deactivate Kafka
-
-    API-->>Web: 202 Accepted {job_id}
-    deactivate API
-
-    Kafka->>Worker: Deliver job message
-    activate Worker
-
-    Worker->>DB: UPDATE jobs SET status="running"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>MinIO: Download FASTQ files
-    activate MinIO
-    MinIO-->>Worker: R1.fastq.gz, R2.fastq.gz
-    deactivate MinIO
-
-    Worker->>Worker: Run Bowtie2 alignment (8 threads)
-    Note over Worker: ~30-60 minutes for 50M reads
-
-    Worker->>Worker: Convert SAM to BAM (samtools)
-    Worker->>Worker: Index BAM (samtools)
-
-    Worker->>MinIO: Upload aligned/{sample}.bam
-    activate MinIO
-    MinIO-->>Worker: Upload complete
-    deactivate MinIO
-
-    Worker->>DB: UPDATE jobs SET status="completed"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>Kafka: Publish to "jobs.completed"
-    activate Kafka
-    Kafka-->>Worker: ACK
-    deactivate Kafka
-
-    deactivate Worker
-
-    Kafka->>API: Notify via WebSocket
-    API->>Web: WebSocket: {job_id, status: "completed"}
-    Web-->>User: Show completion notification
-```
-
----
-
-## 4. Real-Time Status Updates (WebSocket)
-
-```mermaid
-sequenceDiagram
-    actor User
-    participant Web as Web App
-    participant WS as WebSocket Server
-    participant Kafka as Apache Kafka
-    participant Worker as Processing Worker
-
-    User->>Web: Open Job Monitor page
-    Web->>WS: Connect WebSocket
-    activate WS
-    WS-->>Web: Connection established
-
-    Web->>WS: Subscribe to job_id=12345
-    WS->>WS: Add to subscribers list
-
-    Worker->>Kafka: Publish progress {job_id: 12345, progress: 45%}
-    activate Kafka
-    Kafka->>WS: Deliver progress update
-    deactivate Kafka
-
-    WS->>WS: Filter subscribers (job_id=12345)
-    WS->>Web: Send progress update
-    Web-->>User: Update progress bar (45%)
-
-    Worker->>Kafka: Publish progress {job_id: 12345, progress: 80%}
-    activate Kafka
-    Kafka->>WS: Deliver progress update
-    deactivate Kafka
-
-    WS->>Web: Send progress update
-    Web-->>User: Update progress bar (80%)
-
-    Worker->>Kafka: Publish completion {job_id: 12345, status: "completed"}
-    activate Kafka
-    Kafka->>WS: Deliver completion event
-    deactivate Kafka
-
-    WS->>Web: Send completion event
-    Web-->>User: Show "Analysis Complete" notification
-
-    User->>Web: Close page
-    Web->>WS: Disconnect WebSocket
-    WS->>WS: Remove from subscribers
-    deactivate WS
-```
-
----
-
-## 5. Off-Target Prediction Workflow
-
-```mermaid
-sequenceDiagram
-    participant API as API Gateway
-    participant Kafka as Apache Kafka
-    participant Worker as Off-Target Worker
-    participant MinIO as MinIO S3
-    participant Mongo as MongoDB
-    participant DB as PostgreSQL
-
-    API->>Kafka: Publish to "jobs.off_target"
-    activate Kafka
-    Kafka->>Worker: Deliver job message
-    deactivate Kafka
-    activate Worker
-
-    Worker->>DB: UPDATE jobs SET status="running"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>MinIO: Download aligned BAM file
-    activate MinIO
-    MinIO-->>Worker: sample.bam
-    deactivate MinIO
-
-    Worker->>Mongo: Query genome annotations
-    activate Mongo
-    Mongo-->>Worker: Gene positions, sequences
-    deactivate Mongo
-
-    Worker->>Worker: Extract guide RNA from experiment
-    Worker->>Worker: Scan BAM for mismatched alignments
-
-    loop For each potential off-target
-        Worker->>Worker: Calculate CFD score (ML model)
-        Worker->>Worker: Calculate MIT score
-        Worker->>DB: INSERT INTO off_targets
-        activate DB
-        DB-->>Worker: OK
-        deactivate DB
-    end
-
-    Worker->>DB: UPDATE jobs SET status="completed"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>Kafka: Publish to "jobs.completed"
-    activate Kafka
-    Kafka-->>Worker: ACK
-    deactivate Kafka
-
-    deactivate Worker
-```
-
----
-
-## 6. Differential Expression Analysis
-
-```mermaid
-sequenceDiagram
-    participant API as API Gateway
-    participant Kafka as Apache Kafka
-    participant Worker as DESeq2 Worker (R)
-    participant MinIO as MinIO S3
-    participant DB as PostgreSQL
-
-    API->>Kafka: Publish to "jobs.diff_expr"
-    activate Kafka
-    Kafka->>Worker: Deliver job message
-    deactivate Kafka
-    activate Worker
-
-    Worker->>DB: UPDATE jobs SET status="running"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    loop For each sample
-        Worker->>MinIO: Download BAM file
-        activate MinIO
-        MinIO-->>Worker: sample.bam
-        deactivate MinIO
-    end
-
-    Worker->>Worker: Run featureCounts (count reads per gene)
-    Note over Worker: ~10-20 minutes for 50M reads
-
-    Worker->>Worker: Create DESeqDataSet
-    Worker->>Worker: Run DESeq2 normalization
-    Worker->>Worker: Test for differential expression
-
-    loop For each gene
-        Worker->>DB: INSERT INTO differential_expression
-        activate DB
-        DB-->>Worker: OK
-        deactivate DB
-    end
-
-    Worker->>Worker: Generate plots (MA, volcano, heatmap)
-    Worker->>MinIO: Upload plots PDF
-    activate MinIO
-    MinIO-->>Worker: Upload complete
-    deactivate MinIO
-
-    Worker->>DB: UPDATE jobs SET status="completed"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>Kafka: Publish to "jobs.completed"
-    activate Kafka
-    Kafka-->>Worker: ACK
-    deactivate Kafka
-
-    deactivate Worker
-```
-
----
-
-## 7. Result Visualization & Export
-
-```mermaid
-sequenceDiagram
-    actor User
-    participant Web as Web App
-    participant API as API Gateway
-    participant DB as PostgreSQL
-    participant MinIO as MinIO S3
-
-    User->>Web: Navigate to Experiment Results
-    Web->>API: GET /api/v1/results/off-targets/{job_id}
-    activate API
-
-    API->>DB: SELECT * FROM off_targets WHERE job_id=?
-    activate DB
-    DB-->>API: List of off-targets
-    deactivate DB
-
-    API-->>Web: 200 OK {off_targets: [...]}
-    deactivate API
-
-    Web->>Web: Render interactive table
-    Web-->>User: Display off-targets
-
-    User->>Web: Click "Export to CSV"
-    Web->>API: GET /api/v1/results/export/{job_id}?format=csv
-    activate API
-
-    API->>DB: SELECT * FROM off_targets, differential_expression
-    activate DB
-    DB-->>API: Full results
-    deactivate DB
-
-    API->>API: Generate CSV file
-    API->>MinIO: Upload results/{job_id}/export.csv
-    activate MinIO
-    MinIO-->>API: Pre-signed URL
-    deactivate MinIO
-
-    API-->>Web: 200 OK {download_url}
-    deactivate API
-
-    Web->>MinIO: GET download_url
-    activate MinIO
-    MinIO-->>Web: export.csv
-    deactivate MinIO
-
-    Web-->>User: Download CSV file
-```
-
----
-
-## 8. Error Handling & Retry
-
-```mermaid
-sequenceDiagram
-    participant Kafka as Apache Kafka
-    participant Worker as Processing Worker
-    participant DB as PostgreSQL
-    participant DLQ as Dead Letter Queue
-
-    Kafka->>Worker: Deliver job message (attempt 1)
-    activate Worker
-
-    Worker->>Worker: Process job
-    Worker->>Worker: ERROR: Network timeout
-
-    Worker->>DB: UPDATE jobs SET status="retrying"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>Kafka: NACK (negative acknowledgment)
-    deactivate Worker
-
-    Note over Kafka: Wait 5 seconds
-
-    Kafka->>Worker: Deliver job message (attempt 2)
-    activate Worker
-
-    Worker->>Worker: Process job
-    Worker->>Worker: ERROR: Out of memory
-
-    Worker->>DB: UPDATE jobs SET status="retrying"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>Kafka: NACK
-    deactivate Worker
-
-    Note over Kafka: Wait 10 seconds
-
-    Kafka->>Worker: Deliver job message (attempt 3)
-    activate Worker
-
-    Worker->>Worker: Process job
-    Worker->>Worker: ERROR: Invalid BAM file
-
-    Worker->>DB: UPDATE jobs SET status="failed", error_message="Invalid BAM file"
-    activate DB
-    DB-->>Worker: OK
-    deactivate DB
-
-    Worker->>DLQ: Send to "jobs.failed" topic
-    activate DLQ
-    DLQ-->>Worker: ACK
-    deactivate DLQ
-
-    Worker->>Kafka: ACK (final acknowledgment)
-    deactivate Worker
-
-    Note over DLQ: Admin reviews failed jobs
-```
-
----
-
-## 9. Monitoring & Alerting
-
-```mermaid
-sequenceDiagram
-    participant Service as Any Service
-    participant Prom as Prometheus
-    participant Alert as Alertmanager
-    participant Slack as Slack
-    participant PD as PagerDuty
-
-    loop Every 15 seconds
-        Prom->>Service: Scrape /metrics endpoint
-        activate Service
-        Service-->>Prom: Metrics (CPU, memory, request rate)
-        deactivate Service
-    end
-
-    Prom->>Prom: Evaluate alerting rules
-
-    alt CPU > 90% for 5 minutes
-        Prom->>Alert: Fire alert: "HighCPU"
-        activate Alert
-
-        Alert->>Alert: Check routing rules
-
-        alt Severity: Warning
-            Alert->>Slack: Send message to #alerts
-            activate Slack
-            Slack-->>Alert: OK
-            deactivate Slack
-        else Severity: Critical
-            Alert->>PD: Trigger incident
-            activate PD
-            PD-->>Alert: Incident created
-            deactivate PD
-
-            PD->>PD: Notify on-call engineer
-        end
-
-        deactivate Alert
-    end
-```
-
----
-
-**Document Version**: 1.0
-**Last Updated**: 2025-10-12
-**See Also**: [Architecture Overview](../ARCHITECTURE.md)