bigrna.dot

digraph Workflow {

   node [shape="box"];	

   subgraph cluster_0  {
	  SRA_XML [shape=record label="SRA XML|{Study|Sample|Experiment|Run}"]
	  XML_TO_JSON [label="XML to JSON"]
	  SRA_JSON [shape=record label="SRA JSON|{Study|Sample|Experiment|Run}"]
	  BIOSAMPLE_XML [label="Biosample XML"]
	  BIOSAMPLE_JSON [label="Biosample JSON"]
	  SRA_TSV [shape=record label="SRA TSV|{file info|run info|livelist|accessions}"]
	  LOAD_TO_POSTGRES [label="Load to staging\ndatabase"]
	  ETL [label="Transform to\nPostgreSQL Schema"]
	  GRAPHQL_API [label="Populate\nGraphQL API"]
	  OMICIDX_OUTPUT [shape=parallelogram label="OmicIDX API"]

	  { rank=same SRA_XML BIOSAMPLE_XML SRA_TSV }
	  {SRA_XML BIOSAMPLE_XML} -> XML_TO_JSON
	  XML_TO_JSON -> {SRA_JSON BIOSAMPLE_JSON}
	  {BIOSAMPLE_JSON SRA_JSON SRA_TSV} -> LOAD_TO_POSTGRES
	  LOAD_TO_POSTGRES -> ETL -> GRAPHQL_API -> OMICIDX_OUTPUT
	  label="OmicIDX Metadata Pipeline";
	  fontsize=30;
	  color=black;
  }

  subgraph cluster_2 {
	  SRA [shape=record label="{Human|Mouse|Others}"]
	  GDC [shape=box]
	  dbGaP [shape=box]

	  HUMAN_TX [label="Human\nGencode v29 transcript reference"]
	  MOUSE_TX [label="Mouse\nGencode v19 transcript reference"]
	  OTHER_TX [label="Other organism-specific\ntranscript references"]
	  SALMON_IDX [label="Build Salmon index"]
	  SALMON_IDX -> SALMON_QUANT

	  {rank=same HUMAN_TX MOUSE_TX OTHER_TX} -> SALMON_IDX
	  

	  subgraph cluster_1 {
		  FASTQ [label="Generate FASTQ"]
		  
		  SALMON_QUANT [label="Salmon Quantification"]

		  SALMON_GENE [label="Gene counts" color=red]
		  SALMON_TX [label="Transcript counts" color=red]
		  SALMON_QC [label="Quality control metrics" color=red]
		  SALMON_PARAMS [label="Run parameters" color=red]
		  TRANSFER_TO_S3 [label="Automated transfer to s3"]

		  {SRA GDC dbGaP} -> FASTQ
		  FASTQ -> SALMON_QUANT
		  SALMON_QUANT -> {rank=same SALMON_GENE SALMON_TX SALMON_PARAMS SALMON_QC}
		  SALMON_GENE -> TRANSFER_TO_S3 
		  SALMON_TX -> TRANSFER_TO_S3 
		  SALMON_PARAMS -> TRANSFER_TO_S3 
		  SALMON_QC -> TRANSFER_TO_S3 

		  label="Per-sample pipeline\n(Runs on NIH HPC resources)";
		  color=black;
		  fontsize=20
	  }

	  AWS_S3 [label="AWS s3 storage"]
	  S3_TRIGGER [label="AWS s3 trigger"]
	  UPDATE_POSTGRES [label="Update postgresql metadata"]
	  TRANSFER_TO_S3 -> AWS_S3 -> S3_TRIGGER -> UPDATE_POSTGRES
	  
	  fontsize=30;
	  label="BigRNA Pipeline";
	  color=black
	  
	  }
  BIGRNA_API [label="BigRNA API\n& Data Access" fontsize=20 shape=parallelogram]
  UPDATE_POSTGRES -> BIGRNA_API
  OMICIDX_OUTPUT -> BIGRNA_API
  
}