{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://scischema.org/schemas/biology/rna-seq-dge-workflow/1.0.0/master-schema.json",
  "version": "1.0.0",
  "title": "RNA-Seq Differential Gene Expression Analysis Workflow",
  "description": "Schema describing an RNA-seq differential gene expression analysis workflow, including experimental design, quality control, read processing, alignment, quantification, statistical analysis, and reporting.",
  "type": "object",
  "properties": {
    "biological_source_material": {
      "type": "object",
      "description": "Biological source, experimental conditions, and replicate structure.",
      "properties": {
        "organism": {
          "type": "string",
          "description": "Scientific name of the organism."
        },
        "tissue_or_cell_type": {
          "type": "string",
          "description": "Tissue or cell type from which RNA was extracted."
        },
        "treatment_condition": {
          "type": "string",
          "description": "Experimental condition applied to the biological sample."
        },
        "biological_replicates": {
          "type": "integer",
          "minimum": 3,
          "description": "Number of independent biological replicates per condition."
        },
        "sample_collection_protocol": {
          "type": "string",
          "description": "Method used to collect and handle samples prior to RNA extraction."
        }
      },
      "required": [
        "organism",
        "tissue_or_cell_type",
        "treatment_condition",
        "biological_replicates"
      ]
    },
    "rna_extraction_and_quality_metrics": {
      "type": "object",
      "description": "RNA isolation protocol and quality control measurements.",
      "properties": {
        "protocol": {
          "type": "string",
          "description": "RNA isolation method used."
        },
        "rin": {
          "type": "number",
          "minimum": 0,
          "maximum": 10,
          "description": "RNA Integrity Number score on a 0-10 scale."
        },
        "concentration": {
          "type": "number",
          "minimum": 0,
          "description": "RNA concentration in ng/µL."
        },
        "a260_280_ratio": {
          "type": "number",
          "description": "Absorbance ratio indicating RNA purity."
        },
        "quantification_method": {
          "type": "string",
          "description": "Instrument or method used to measure RNA concentration."
        }
      },
      "required": [
        "protocol",
        "rin",
        "concentration"
      ]
    },
    "library_preparation_protocol": {
      "type": "object",
      "description": "cDNA library construction parameters.",
      "properties": {
        "method": {
          "type": "string",
          "description": "RNA enrichment or depletion strategy used for library preparation."
        },
        "strandedness": {
          "type": "string",
          "enum": [
            "forward",
            "reverse",
            "unstranded"
          ],
          "description": "Whether strand-of-origin information is preserved during library synthesis."
        },
        "fragmentation_strategy": {
          "type": "string",
          "description": "Method used to fragment RNA prior to cDNA synthesis."
        },
        "amplification_cycles": {
          "type": "integer",
          "minimum": 0,
          "description": "Number of PCR amplification cycles applied during library preparation."
        },
        "umi_details": {
          "type": "object",
          "description": "Unique Molecular Identifier configuration. Only applicable when UMIs are incorporated.",
          "properties": {
            "umi_length": {
              "type": "integer",
              "minimum": 4,
              "description": "Length of the UMI sequence in base pairs."
            },
            "umi_position": {
              "type": "string",
              "enum": [
                "5_prime",
                "3_prime",
                "internal"
              ],
              "description": "Position of the UMI within the sequencing read."
            },
            "deduplication_method": {
              "type": "string",
              "enum": [
                "unique",
                "percentile",
                "cluster",
                "adjacency",
                "directional"
              ],
              "description": "Algorithm used to resolve PCR duplicates based on UMI sequences."
            }
          },
          "required": [
            "umi_length",
            "umi_position",
            "deduplication_method"
          ]
        }
      },
      "required": [
        "method",
        "strandedness",
        "fragmentation_strategy",
        "amplification_cycles"
      ]
    },
    "sequencing_platform_and_configuration": {
      "type": "object",
      "description": "Sequencing instrument and run configuration.",
      "properties": {
        "platform": {
          "type": "string",
          "description": "Sequencing instrument model used."
        },
        "read_length": {
          "type": "integer",
          "minimum": 36,
          "description": "Length of each sequencing read in base pairs."
        },
        "read_type": {
          "type": "string",
          "enum": [
            "single-end",
            "paired-end"
          ],
          "description": "Whether reads are sequenced from one or both ends of the fragment."
        },
        "sequencing_depth": {
          "type": "number",
          "minimum": 10,
          "description": "Average number of reads per sample in millions."
        }
      },
      "required": [
        "platform",
        "read_length",
        "read_type",
        "sequencing_depth"
      ]
    },
    "raw_data_quality_control": {
      "type": "object",
      "description": "Preprocessing and quality control steps applied to raw sequencing reads.",
      "properties": {
        "adapter_trimming": {
          "type": "string",
          "description": "Tool used to remove sequencing adapter sequences from reads."
        },
        "quality_filtering": {
          "type": "string",
          "description": "Criteria or tool used to remove low-quality reads or bases."
        },
        "qc_assessment_tool": {
          "type": "string",
          "description": "Tool used for per-base quality assessment and QC reporting."
        }
      },
      "required": [
        "adapter_trimming",
        "quality_filtering",
        "qc_assessment_tool"
      ]
    },
    "read_alignment": {
      "type": "object",
      "description": "Read alignment or pseudoalignment to the reference genome or transcriptome.",
      "properties": {
        "tool": {
          "type": "string",
          "description": "Alignment or pseudoalignment tool used."
        },
        "genome_assembly": {
          "type": "string",
          "description": "Reference genome assembly identifier."
        },
        "annotation_source": {
          "type": "string",
          "description": "Transcript annotation database used."
        },
        "annotation_version": {
          "type": "string",
          "description": "Version of the transcript annotation used."
        },
        "alignment_rate": {
          "type": "number",
          "minimum": 0,
          "maximum": 1,
          "description": "Proportion of reads successfully aligned to the reference."
        }
      },
      "required": [
        "tool",
        "genome_assembly",
        "annotation_source",
        "annotation_version"
      ]
    },
    "expression_quantification": {
      "type": "object",
      "description": "Strategy for estimating gene or transcript expression levels from aligned reads.",
      "properties": {
        "level": {
          "type": "string",
          "enum": [
            "gene",
            "transcript"
          ],
          "description": "Resolution at which expression is quantified."
        },
        "count_summarization_method": {
          "type": "string",
          "description": "Tool or method used to summarize aligned reads into expression counts."
        },
        "normalization_aware": {
          "type": "boolean",
          "description": "Whether the quantification step incorporates bias correction."
        },
        "low_count_filtering": {
          "type": "object",
          "description": "Criteria for removing lowly expressed genes prior to statistical testing.",
          "properties": {
            "min_counts_per_sample": {
              "type": "integer",
              "minimum": 0,
              "description": "Minimum read count required in a given sample for gene retention."
            },
            "min_samples_with_min_counts": {
              "type": "integer",
              "minimum": 1,
              "description": "Minimum number of samples that must meet the count threshold."
            },
            "tool": {
              "type": "string",
              "description": "Tool used to apply low-count filtering."
            }
          }
        }
      },
      "required": [
        "level",
        "count_summarization_method"
      ]
    },
    "normalization_method": {
      "type": "object",
      "description": "Between-sample normalization to correct for library size and composition biases. RPKM/FPKM are within-sample metrics and not appropriate for DGE.",
      "properties": {
        "method": {
          "type": "string",
          "enum": [
            "median_of_ratios",
            "TMM",
            "upper_quartile",
            "RUV",
            "other"
          ],
          "description": "Normalization strategy applied across samples."
        },
        "tool": {
          "type": "string",
          "description": "Software package implementing the normalization."
        }
      },
      "required": [
        "method",
        "tool"
      ]
    },
    "statistical_analysis": {
      "type": "object",
      "description": "Statistical framework for differential expression testing, including model specification, batch correction, and significance criteria.",
      "properties": {
        "model_framework": {
          "type": "string",
          "description": "Statistical model used for differential expression testing."
        },
        "design_formula": {
          "type": "string",
          "description": "Formula specifying experimental factors and covariates included in the model."
        },
        "covariates": {
          "type": "array",
          "items": {
            "type": "string"
          },
          "description": "Additional variables included in the model to account for technical or biological variability."
        },
        "dispersion_estimation": {
          "type": "string",
          "enum": [
            "common",
            "trended",
            "per_gene",
            "local"
          ],
          "description": "Method for estimating gene-wise overdispersion in count data."
        },
        "batch_correction": {
          "type": "object",
          "description": "Correction applied to remove technical variation introduced by experimental batches.",
          "properties": {
            "applied": {
              "type": "boolean",
              "description": "Whether batch correction was performed."
            },
            "method": {
              "type": "string",
              "enum": [
                "ComBat",
                "limma_removeBatchEffect",
                "SVA",
                "RUV",
                "none"
              ],
              "description": "Batch correction method used."
            },
            "batch_variable": {
              "type": "string",
              "description": "Name of the variable representing batch in the experimental design."
            }
          }
        },
        "multiple_testing_correction": {
          "type": "string",
          "enum": [
            "Benjamini-Hochberg",
            "Bonferroni",
            "Holm",
            "Storey_q_value"
          ],
          "description": "Method for controlling the false discovery rate across multiple hypothesis tests."
        },
        "significance_thresholds": {
          "type": "object",
          "description": "Thresholds used to declare a gene as differentially expressed.",
          "properties": {
            "adjusted_pvalue_threshold": {
              "type": "number",
              "minimum": 0,
              "maximum": 1,
              "description": "Maximum adjusted p-value for a gene to be considered significant."
            },
            "log2_fold_change_threshold": {
              "type": "number",
              "minimum": 0,
              "description": "Minimum absolute log2 fold change required for biological significance."
            }
          },
          "required": [
            "adjusted_pvalue_threshold",
            "log2_fold_change_threshold"
          ]
        }
      },
      "required": [
        "model_framework",
        "design_formula",
        "multiple_testing_correction",
        "significance_thresholds"
      ]
    },
    "output_artifacts": {
      "type": "array",
      "description": "Files and visualizations generated by the workflow.",
      "items": {
        "type": "object",
        "required": [
          "type",
          "path"
        ],
        "properties": {
          "type": {
            "type": "string",
            "enum": [
              "normalized_count_matrix",
              "differential_expression_table",
              "ma_plot",
              "volcano_plot",
              "pathway_enrichment_results",
              "qc_report",
              "gene_expression_heatmap",
              "other"
            ],
            "description": "Category of the output artifact."
          },
          "path": {
            "type": "string",
            "description": "File path or URI where the artifact is stored."
          },
          "format": {
            "type": "string",
            "enum": [
              "csv",
              "tsv",
              "pdf",
              "png",
              "html",
              "rds",
              "h5"
            ],
            "description": "File format of the output artifact."
          }
        }
      },
      "minItems": 1
    },
    "reproducibility_metadata": {
      "type": "object",
      "description": "Metadata required to reproduce the workflow and ensure FAIR compliance.",
      "properties": {
        "software_versions": {
          "type": "object",
          "unevaluatedProperties": {
            "type": "string"
          },
          "description": "Mapping of software tool names to their exact versions."
        },
        "workflow_engine": {
          "type": "string",
          "enum": [
            "Snakemake",
            "Nextflow",
            "CWL",
            "WDL",
            "Galaxy",
            "other"
          ],
          "description": "Workflow management system used to orchestrate the pipeline."
        },
        "computational_environment": {
          "type": "string",
          "description": "Container or environment specification used for execution."
        },
        "random_seed": {
          "type": "integer",
          "description": "Seed value used in stochastic steps to ensure reproducibility."
        },
        "parameter_settings": {
          "type": "object",
          "unevaluatedProperties": true,
          "description": "Non-default parameter configurations applied across workflow tools."
        }
      },
      "required": [
        "software_versions",
        "workflow_engine",
        "computational_environment"
      ]
    }
  },
  "required": [
    "biological_source_material",
    "rna_extraction_and_quality_metrics",
    "library_preparation_protocol",
    "sequencing_platform_and_configuration",
    "read_alignment",
    "expression_quantification",
    "normalization_method",
    "statistical_analysis",
    "reproducibility_metadata"
  ],
  "unevaluatedProperties": false
}