{ "$schema": "https://json-schema.org/draft/2020-12/schema", "$id": "https://scischema.org/schemas/biology/rna-seq-library-preparation/1.0.0/master-schema.json", "version": "1.0.0", "title": "Bulk RNA-seq Library Preparation and Sequencing Process", "description": "Gold-standard schema representing the complete workflow for Bulk RNA-seq library preparation and sequencing, incorporating best practices from iterative expert evaluation across five rounds of Schema Miner refinement.", "type": "object", "$defs": { "kit_details": { "type": "object", "description": "Reusable commercial kit identification block — referenced across RNA extraction, library preparation, and transcript enrichment to avoid redundant property definitions.", "required": [ "kit_name", "manufacturer" ], "properties": { "kit_name": { "type": "string", "description": "Full commercial name of the kit used (e.g., Qiagen RNeasy Plus Mini Kit, NEBNext Ultra II RNA Library Prep Kit)", "examples": [ "Qiagen RNeasy Plus", "NEBNext Ultra II" ] }, "manufacturer": { "type": "string", "description": "Name of the kit manufacturer (e.g., Qiagen, New England Biolabs, Illumina, Thermo Fisher Scientific)", "examples": [ "Qiagen", "NEB", "Illumina" ] }, "catalog_number": { "type": "string", "description": "Manufacturer catalog number for precise lot-to-lot traceability and reproducibility verification", "examples": [ "74134", "E7770S" ] }, "kit_version": { "type": "string", "description": "Version or lot number of the kit used, important for tracking batch-to-batch variation in reagent performance", "examples": [ "v2", "Lot_20240101" ] } } } }, "required": [ "sample", "rna_extraction", "rna_input", "quality_control", "library_preparation", "sequencing", "primary_data_processing", "experimental_design", "metadata" ], "properties": { "sample": { "type": "object", "description": "Biological sample identity and source information, separated from extraction and QC as an independent workflow stage.", "required": [ "sample_id", "organism", "tissue_type", "source_material" ], "properties": { "sample_id": { "type": "string", "description": "Unique identifier for the biological sample within the study, used to track the sample across all workflow stages and link to public repository records", "examples": [ "Sample_001", "GSM123456" ] }, "organism": { "type": "string", "description": "Scientific name of the organism from which the sample was derived, following NCBI taxonomy conventions", "examples": [ "Homo sapiens", "Mus musculus" ] }, "tissue_type": { "type": "string", "description": "Specific tissue or organ from which RNA was extracted. Should be as precise as possible (e.g., prefrontal cortex rather than brain) to enable accurate cross-study comparisons", "examples": [ "liver", "breast tumor", "whole blood", "prefrontal cortex" ] }, "cell_type": { "type": "string", "description": "Specific cell line or cell population if applicable. For primary cells, use established nomenclature (e.g., CD4+ T cells); for cell lines, use ATCC or equivalent registry name", "examples": [ "MCF7", "CD4+ T cells", "HEK293T" ] }, "source_material": { "type": "string", "description": "Type and preservation state of the biological material at the point of collection. Affects RNA integrity expectations and extraction protocol selection", "examples": [ "fresh frozen", "FFPE", "cultured cells", "RNAlater-stabilized" ] }, "storage_conditions": { "type": "object", "description": "Storage conditions for the biological sample at different stages prior to and after RNA extraction, relevant for assessing RNA integrity", "properties": { "prior_to_extraction": { "type": "string", "description": "How and at what temperature the sample was stored between collection and RNA extraction", "examples": [ "Flash-frozen at -80°C", "FFPE block at room temperature", "RNAlater at -20°C" ] }, "post_extraction": { "type": "string", "description": "Storage conditions for the extracted RNA prior to library preparation", "examples": [ "-80°C in RNase-free water", "-80°C in ethanol precipitation", "-80°C in TE buffer" ] } } }, "treatment_or_condition": { "type": "string", "description": "Experimental treatment or condition applied to the sample prior to RNA extraction. Should include dose, duration, and vehicle where applicable", "examples": [ "10µM doxorubicin 24h", "Wild-type control", "siRNA knockdown 48h", "hypoxia 1% O2 6h" ] } } }, "rna_extraction": { "type": "object", "description": "RNA isolation methodology and associated parameters. Post-extraction quality assessment is handled exclusively in the quality_control block to avoid duplication.", "required": [ "method", "dnase_treatment" ], "properties": { "method": { "type": "string", "description": "General approach or protocol used for RNA isolation. Determines expected RNA purity and integrity and informs downstream QC thresholds", "examples": [ "TRIzol phenol-chloroform extraction", "Qiagen RNeasy column-based", "mechanical homogenization followed by column purification" ] }, "kit_details": { "$ref": "#/$defs/kit_details", "description": "Commercial kit used for RNA extraction, if applicable. Not all extractions use a commercial kit." }, "protocol_reference": { "type": "string", "description": "DOI, URL, or citation for the extraction protocol used, enabling exact replication of the isolation procedure", "examples": [ "https://doi.org/10.1038/nprot.2006.83", "Manufacturer protocol v2.0" ] }, "dnase_treatment": { "type": "boolean", "description": "Whether DNase I treatment was applied post-extraction to remove residual genomic DNA contamination, which can confound alignment and quantification" }, "yield": { "type": "number", "minimum": 0, "description": "Total RNA yield after extraction, measured in nanograms (ng). Used to determine whether sufficient material is available for library preparation and to assess extraction efficiency", "examples": [ 1000, 5000, 25000 ] }, "starting_material": { "type": "object", "description": "Pre-extraction biological input material quantity, used to calculate extraction efficiency and normalize yield across samples", "properties": { "quantity": { "type": "number", "minimum": 0, "description": "Amount of biological starting material used for RNA extraction", "examples": [ 25, 100, 1000000 ] }, "unit": { "type": "string", "enum": [ "mg", "g", "cells", "uL" ], "description": "Unit of measurement for the starting material quantity (mg or g for tissue; cells for cell cultures; uL for liquid samples)" } } }, "extraction_date": { "type": "string", "format": "date", "description": "Date when RNA extraction was performed, in ISO 8601 format (YYYY-MM-DD). Required for temporal traceability and batch effect documentation", "examples": [ "2024-01-10" ] } } }, "rna_input": { "type": "object", "description": "RNA input quantity used for library preparation — documented as a separate workflow stage from extraction to capture the specific amount committed to library construction, which may differ from total yield.", "required": [ "quantity", "quantity_unit" ], "properties": { "quantity": { "type": "number", "minimum": 0.001, "description": "Amount of RNA used as input for library preparation. Influences PCR cycle number requirements and downstream library complexity", "examples": [ 50, 100, 500, 1000 ] }, "quantity_unit": { "type": "string", "enum": [ "ng", "µg" ], "description": "Unit of measurement for RNA input quantity. Nanograms (ng) for standard inputs; micrograms (µg) for high-input protocols" } } }, "quality_control": { "type": "object", "description": "Comprehensive two-stage quality control block covering post-extraction RNA integrity assessment and post-library-preparation library quality assessment. Both sub-blocks are required to fully document sample suitability for sequencing.", "required": [ "rna_integrity", "library_quality" ], "properties": { "rna_integrity": { "type": "object", "description": "Post-extraction RNA quality metrics assessing integrity, purity, and concentration using Bioanalyzer, TapeStation, or spectrophotometric instruments", "required": [ "rin" ], "properties": { "rin": { "type": "number", "minimum": 1, "maximum": 10, "description": "RNA Integrity Number (RIN) from Bioanalyzer or equivalent (scale 1-10). Values above 7 are generally recommended for standard poly-A RNA-seq; lower values may be acceptable for rRNA-depleted or FFPE samples", "examples": [ 8.5, 9.2, 7.1 ] }, "dv200": { "type": "number", "minimum": 0, "maximum": 100, "description": "Percentage of RNA fragments greater than 200 nucleotides in length. Critical quality metric for FFPE samples where RIN is not applicable; values above 30% are generally considered acceptable", "examples": [ 65, 85, 95 ] }, "bioanalyzer_profile": { "type": "string", "description": "Qualitative description of the electropherogram trace from Bioanalyzer or TapeStation, documenting the visual appearance of the 18S and 28S ribosomal RNA peaks and any degradation markers", "examples": [ "intact 28S/18S peaks", "moderately degraded with 28S shoulder", "high quality with RIN 9.2" ] }, "quality_assessment_date": { "type": "string", "format": "date", "description": "Date when RNA quality assessment was performed, in ISO 8601 format (YYYY-MM-DD). Used to track time elapsed between extraction and quality assessment", "examples": [ "2024-01-11" ] }, "spectrophotometric_ratios": { "type": "object", "description": "Spectrophotometric purity ratios measured by NanoDrop or equivalent instrument, used to detect protein and organic compound contamination in the RNA sample", "required": [ "a260_a280", "a260_a230" ], "properties": { "a260_a280": { "type": "number", "minimum": 1.5, "maximum": 2.5, "description": "Absorbance ratio at 260nm to 280nm — indicator of protein contamination. Pure RNA is expected to have a ratio of approximately 2.0; values below 1.8 suggest protein or phenol contamination", "examples": [ 2.0, 1.95, 2.05 ] }, "a260_a230": { "type": "number", "minimum": 1.5, "maximum": 3.0, "description": "Absorbance ratio at 260nm to 230nm — indicator of organic compound contamination (guanidinium salts, phenol, EDTA). Expected range 2.0–2.2 for pure RNA; values below 1.8 indicate carryover contamination", "examples": [ 2.1, 2.3, 2.15 ] } } } } }, "library_quality": { "type": "object", "description": "Post-library-preparation quality metrics assessing fragment size distribution, concentration, and sequencing suitability of the prepared library before loading onto the sequencer", "properties": { "average_insert_size": { "type": "integer", "description": "Average insert size of library fragments after adapter ligation, measured in base pairs (bp) by Bioanalyzer or TapeStation. Typical range 150–500 bp for standard RNA-seq", "examples": [ 300, 250, 350 ] }, "min_size": { "type": "integer", "description": "Minimum fragment size detected in the library by Bioanalyzer or TapeStation, in base pairs (bp). Used to detect adapter dimer contamination (typically <150 bp)", "examples": [ 150, 180 ] }, "max_size": { "type": "integer", "description": "Maximum fragment size detected in the library by Bioanalyzer or TapeStation, in base pairs (bp). Large fragments may indicate incomplete fragmentation", "examples": [ 500, 600 ] }, "library_concentration": { "type": "number", "description": "Final library concentration before sequencing, measured in nanograms per microliter (ng/µL) by Qubit fluorometry or equivalent. Used to calculate loading concentration for the sequencer", "examples": [ 12.5, 8.3, 25.0 ] }, "quantification_method": { "type": "string", "enum": [ "Qubit fluorometry", "Bioanalyzer", "TapeStation", "qPCR" ], "description": "Instrument or method used to quantify library concentration. Qubit fluorometry is preferred for accurate dsDNA-specific quantification; qPCR measures only amplifiable molecules" }, "adapter_dimer_contamination": { "type": "boolean", "description": "Whether adapter dimer artifacts (typically 120–150 bp fragments) were detected in the final library. Presence indicates incomplete size selection and may reduce effective sequencing yield" }, "gc_content_bias": { "type": "object", "description": "Assessment of GC content bias in the library, which can affect quantification accuracy of high- or low-GC transcripts", "properties": { "detected": { "type": "boolean", "description": "Whether GC content bias was detected during library QC or post-sequencing FastQC analysis" }, "bias_direction": { "type": "string", "enum": [ "high GC", "low GC", "both" ], "description": "Direction of GC bias if detected — high GC bias favors amplification of GC-rich transcripts; low GC bias favors AT-rich transcripts" }, "assessment_method": { "type": "string", "description": "Method used to assess GC content bias (e.g., FastQC per-sequence GC content module, RSeQC, MultiQC)", "examples": [ "FastQC", "RSeQC", "MultiQC" ] } } }, "alignment_rate": { "type": "number", "minimum": 0, "maximum": 100, "description": "Percentage of sequenced reads that successfully aligned to the reference genome. Values below 70% may indicate contamination, wrong reference genome, or library quality issues", "examples": [ 92.5, 87.3, 95.1 ] }, "exon_capture_rate": { "type": "number", "minimum": 0, "maximum": 100, "description": "Percentage of annotated exons with at least one read of coverage. Low values indicate poor library complexity or coverage uniformity", "examples": [ 85.2, 91.0, 78.5 ] }, "coverage_variance": { "type": "number", "description": "Coefficient of variation (CV) of base-level coverage across exons. Higher values indicate uneven coverage, which may reflect GC bias, fragmentation bias, or low library complexity", "examples": [ 0.45, 0.62, 0.31 ] } } } } }, "library_preparation": { "type": "object", "description": "Complete library construction workflow from transcript enrichment through PCR amplification. Transcript enrichment is nested here as an intrinsic construction step rather than a standalone block.", "required": [ "transcript_enrichment", "protocol_or_kit", "library_type", "umi_usage", "fragmentation_and_amplification" ], "properties": { "transcript_enrichment": { "type": "object", "description": "Transcript population targeting strategy applied before or during library construction to selectively capture specific RNA species and reduce representation of unwanted transcripts", "required": [ "strategy" ], "properties": { "strategy": { "type": "string", "enum": [ "poly-A selection", "rRNA depletion (RiboZero)", "globin depletion", "total RNA" ], "description": "Method used to enrich for specific transcript populations. Poly-A selection captures mRNA via oligo-dT beads; rRNA depletion removes ribosomal RNA to retain all RNA species including non-polyadenylated transcripts" }, "target_rna": { "type": "string", "description": "Specific RNA population targeted by the enrichment strategy, if more specific than the general strategy indicates", "examples": [ "mRNA", "lncRNA", "total RNA excluding rRNA", "small RNA" ] }, "kit_details": { "$ref": "#/$defs/kit_details", "description": "Commercial kit used for transcript enrichment (e.g., Illumina TruSeq Stranded mRNA, Ribo-Zero Gold)" }, "protocol_reference": { "type": "string", "description": "DOI, URL, or citation for the transcript enrichment protocol used", "examples": [ "https://doi.org/10.1101/pdb.prot5350" ] } } }, "protocol_or_kit": { "type": "string", "description": "Name of the library preparation protocol or commercial kit used for the complete library construction workflow", "examples": [ "Illumina TruSeq Stranded mRNA Library Prep", "NEBNext Ultra II RNA Library Prep", "KAPA RNA HyperPrep" ] }, "kit_details": { "$ref": "#/$defs/kit_details", "description": "Detailed kit identification for the library preparation kit, including catalog number for lot traceability" }, "library_type": { "type": "string", "enum": [ "stranded-RF (first-strand)", "stranded-FR (second-strand)", "unstranded" ], "description": "Library strandedness and directionality using tool-convention labels compatible with STAR, Salmon, HTSeq, and featureCounts. RF = reverse-forward (first-strand synthesis, e.g., dUTP method); FR = forward-reverse (second-strand synthesis)" }, "strand_specific_method": { "type": "string", "enum": [ "dUTP/UDG", "ligation-based", "other" ], "description": "Specific biochemical method used to achieve strand specificity. The dUTP/UDG method incorporates deoxyuridine triphosphate (dUTP) during second-strand synthesis and selectively degrades the dUTP-containing strand using uracil-DNA glycosylase (UDG), preserving original strand information" }, "umi_usage": { "type": "boolean", "description": "Whether Unique Molecular Identifiers (UMIs) were incorporated during library preparation to enable PCR duplicate identification and correction, improving quantification accuracy especially for low-input samples" }, "adapter_sequences": { "type": "object", "description": "Specific adapter oligonucleotide sequences ligated to library fragments for sequencing platform compatibility and sample identification", "required": [ "read1_adapter" ], "properties": { "read1_adapter": { "type": "string", "description": "Adapter sequence ligated to the first read end, required for all library types", "examples": [ "AGATCGGAAGAGC", "CTGTCTCTTATACACATCT" ] }, "read2_adapter": { "type": "string", "description": "Adapter sequence ligated to the second read end — only applicable for paired-end sequencing libraries", "examples": [ "AGATCGGAAGAGC", "CTGTCTCTTATACACATCT" ] } } }, "adapter_ligation": { "type": "object", "description": "Detailed parameters of the adapter ligation step, providing complete documentation of the library construction chemistry", "properties": { "adapter_type": { "type": "string", "enum": [ "Y-shaped", "A-tailed", "blunt-end", "other" ], "description": "Structural type of adapter used. Y-shaped adapters are standard for strand-specific libraries; A-tailed adapters use a 3' adenine overhang for ligation" }, "adapter_sequence_5prime": { "type": "string", "description": "Full nucleotide sequence of the adapter ligated to the 5' end of the RNA or cDNA fragment" }, "adapter_sequence_3prime": { "type": "string", "description": "Full nucleotide sequence of the adapter ligated to the 3' end of the RNA or cDNA fragment" }, "ligation_strategy": { "type": "string", "description": "Specific ligation chemistry or approach used to attach adapters to library fragments", "examples": [ "A-tailing followed by Y-adapter ligation", "blunt-end ligation", "RNA ligation without adapter" ] } } }, "reverse_transcription": { "type": "string", "description": "Method used for converting RNA to complementary DNA (cDNA) during library preparation. Priming strategy affects coverage uniformity and transcript representation", "examples": [ "oligo-dT priming", "random hexamers", "gene-specific priming", "template switching" ] }, "phosphorothioate_oligos_used": { "type": "boolean", "description": "Whether phosphorothioate-modified oligonucleotides were used during library preparation to prevent adapter-adapter ligation (adapter dimer formation) and improve library complexity" }, "indexing": { "type": "object", "description": "Sample multiplexing index information, enabling multiple samples to be pooled and sequenced in a single lane with subsequent demultiplexing by barcode", "required": [ "index_type" ], "properties": { "index_type": { "type": "string", "enum": [ "single", "dual" ], "description": "Whether single or dual indexing was used. Dual indexing reduces index hopping artifacts and is required for high-throughput multiplexing on patterned flow cells" }, "index_sequence": { "type": "string", "description": "Actual barcode nucleotide sequence used to uniquely identify this sample in the multiplexed pool", "examples": [ "ATCACG", "TTAGGC-ATCACG" ] }, "index_length": { "type": "integer", "minimum": 4, "maximum": 10, "description": "Length of the index sequence in nucleotides. Standard Illumina indexes are 6–8 nucleotides", "examples": [ 6, 8, 10 ] } } }, "fragmentation_and_amplification": { "type": "object", "description": "Unified block documenting RNA or cDNA fragmentation parameters and subsequent PCR amplification conditions. Both steps are intrinsically linked in library construction as fragmentation size directly influences amplification efficiency.", "required": [ "fragmentation_method", "fragmentation_temperature", "fragmentation_time", "pcr_cycles" ], "properties": { "fragmentation_method": { "type": "string", "enum": [ "enzymatic", "thermal", "sonication", "chemical", "acoustic shearing", "mechanical shearing", "none" ], "description": "Method used to fragment RNA or cDNA into sequenceable lengths. Enzymatic fragmentation uses RNA fragmentation reagents or transposases; thermal fragmentation uses heat-induced hydrolysis; acoustic shearing is advantageous for samples with high rRNA content" }, "target_size": { "type": "integer", "minimum": 100, "maximum": 1000, "description": "Target fragment size in base pairs (bp) after fragmentation. Typical range 150–500 bp for standard RNA-seq; shorter fragments (150–200 bp) are used for small RNA or FFPE samples", "examples": [ 200, 300, 400 ] }, "fragmentation_temperature": { "type": "number", "minimum": 4, "description": "Temperature in degrees Celsius at which fragmentation was performed. Minimum 4°C to exclude physically impossible values; thermal fragmentation typically performed at 94°C", "examples": [ 94, 95, 85 ] }, "fragmentation_time": { "type": "number", "minimum": 1, "description": "Duration of the fragmentation step in minutes. Minimum 1 minute; longer times produce smaller fragments. Typical range 1–15 minutes depending on target size and method", "examples": [ 3, 5, 8, 15 ] }, "pcr_cycles": { "type": "integer", "minimum": 5, "maximum": 20, "description": "Number of PCR amplification cycles used to generate sufficient library material for sequencing. Typical range 5–15 cycles for standard RNA-seq inputs; optimal cycle count should be determined by real-time PCR to avoid over-amplification and GC bias", "examples": [ 8, 12, 15 ] }, "polymerase": { "type": "string", "description": "DNA polymerase used for PCR amplification during library preparation. High-fidelity polymerases reduce amplification-introduced errors and GC bias", "examples": [ "KAPA HiFi", "Phusion High-Fidelity", "Q5 Hot Start" ] }, "amplification_method": { "type": "string", "enum": [ "rolling circle", "bridge PCR", "other" ], "description": "Mechanism used for library amplification on the sequencing platform. Rolling circle amplification (DNBseq/MGI platforms) generates DNA nanoballs by isothermal linear amplification without PCR errors. Bridge PCR (Illumina platforms) uses exponential cluster amplification and is prone to GC content bias" }, "purification_method": { "type": "string", "enum": [ "SPRI beads", "gel electrophoresis", "column-based" ], "description": "Method used to purify library fragments between workflow steps, removing enzymes, primers, and unwanted fragment sizes" }, "size_selection_method": { "type": "string", "enum": [ "SPRI bead subtraction", "gel extraction", "none" ], "description": "Method used to select library fragments within the desired size range. SPRI bead subtraction allows size selection without gel; gel extraction provides precise size selection but is lower throughput" } } }, "library_preparation_notes": { "type": "string", "description": "Free-text field for additional notes about library preparation, including deviations from standard protocols, QC observations, batch effects, or troubleshooting steps taken" } } }, "sequencing": { "type": "object", "description": "Sequencing instrument configuration, run parameters, and metadata documenting the technical specifications of the sequencing experiment.", "required": [ "platform", "read_configuration", "sequencing_depth" ], "properties": { "platform": { "type": "string", "description": "Full name of the sequencing instrument model used. Not restricted to Illumina platforms to support emerging short-read and long-read sequencing technologies", "examples": [ "Illumina NovaSeq 6000", "Illumina NextSeq 2000", "MGI DNBSEQ-G400", "BGISEQ-500", "Element Biosciences AVITI", "Illumina HiSeq 4000" ] }, "flow_cell_type": { "type": "string", "description": "Flow cell type used for sequencing run, which determines cluster density, read length capacity, and total output. Relevant primarily for Illumina platforms", "examples": [ "S4", "S2", "SP", "P2", "High Output" ] }, "flow_cell_patterned": { "type": "boolean", "description": "Whether a patterned flow cell was used for sequencing. Patterned flow cells (e.g., Illumina NovaSeq, NextSeq 2000) use nanowells for higher cluster density and reduced optical duplicates compared to randomly distributed clusters" }, "read_configuration": { "type": "object", "description": "Sequencing read layout and length parameters defining the fundamental sequencing strategy", "required": [ "read_type", "read_length" ], "properties": { "read_type": { "type": "string", "enum": [ "single-end", "paired-end" ], "description": "Whether sequencing reads were generated from one end (single-end) or both ends (paired-end) of the library fragment. Paired-end sequencing improves alignment accuracy and enables detection of splice junctions and structural variants" }, "read_length": { "type": "object", "description": "Read lengths in base pairs — structured as a two-field object to support asymmetric paired-end runs where read 1 and read 2 may have different lengths", "required": [ "read1" ], "properties": { "read1": { "type": "integer", "minimum": 1, "description": "Length of the first sequencing read in base pairs (bp). Required for all library types", "examples": [ 75, 100, 150 ] }, "read2": { "type": "integer", "minimum": 1, "description": "Length of the second sequencing read in base pairs (bp). Only applicable for paired-end sequencing; may differ from read1 in asymmetric runs", "examples": [ 75, 100, 150 ] } } } } }, "sequencing_depth": { "type": "integer", "minimum": 1000000, "description": "Total number of reads generated per sample. Standard bulk RNA-seq for differential expression analysis requires 20–50 million reads; minimum of 1,000,000 enforced to exclude unrealistically low values", "examples": [ 20000000, 30000000, 50000000 ] }, "run_id": { "type": "string", "description": "Unique identifier for the sequencing run, either an internal run ID or a public repository accession number (e.g., SRA run accession)", "examples": [ "SRR123456", "RUN_2024_001", "240115_NB501234_0123_AHXXXXX" ] }, "multiplexing_pool_size": { "type": "integer", "minimum": 1, "description": "Number of samples pooled together and sequenced in a single sequencing run or lane. Relevant for understanding sequencing depth per sample and potential index hopping rates", "examples": [ 8, 12, 16, 24 ] }, "sequencing_date": { "type": "string", "format": "date", "description": "Date when sequencing was performed, in ISO 8601 format (YYYY-MM-DD). Required for temporal traceability and batch documentation", "examples": [ "2024-02-01" ] }, "sequencing_notes": { "type": "string", "description": "Free-text field for additional notes about the sequencing run, including cluster density, PhiX spike-in percentage, quality score distributions, or any technical issues encountered during the run" } } }, "primary_data_processing": { "type": "object", "description": "Computational analysis pipeline applied to raw sequencing data to produce gene-level expression estimates. All required fields are mandatory for FAIR computational reproducibility.", "required": [ "reference_genome", "gene_annotation", "trimming_tool", "alignment_tool", "quantification_tool", "normalization_method", "software_versions", "processing_date" ], "properties": { "reference_genome": { "type": "string", "description": "Reference genome assembly name and version used for read alignment. Should include assembly version to enable exact replication of alignment results", "examples": [ "GRCh38", "GRCm39", "GRCh38.p14", "T2T-CHM13v2.0" ] }, "gene_annotation": { "type": "string", "description": "Gene annotation file (GTF/GFF) name and version used for read quantification. Must match the reference genome assembly to ensure coordinate compatibility", "examples": [ "GENCODE v44", "Ensembl v109", "RefSeq GRCh38.p14" ] }, "trimming_tool": { "type": "string", "description": "Software tool and version used for adapter sequence removal and read quality trimming before alignment. Version number is required for reproducibility", "examples": [ "fastp v0.23.2", "Trimmomatic v0.39", "Cutadapt v4.0", "TrimGalore v0.6.7" ] }, "adapter_filtering_tool": { "type": "string", "description": "Separate tool used specifically for adapter sequence filtering if distinct from the general trimming tool. Used primarily in some DNBseq/MGI pipelines", "examples": [ "SOAPnuke v2.1.7", "Cutadapt v4.0" ] }, "alignment_tool": { "type": "string", "description": "Software tool and version used for aligning trimmed reads to the reference genome. Tool choice affects splice junction detection, multi-mapping handling, and alignment rate", "examples": [ "STAR v2.7.10a", "HISAT2 v2.2.1", "Bowtie2 v2.4.5" ] }, "quantification_tool": { "type": "string", "description": "Software tool and version used for counting aligned reads per gene or transcript to produce expression count matrices", "examples": [ "featureCounts v2.0.3", "Salmon v1.9.0", "RSEM v1.3.3", "HTSeq-count v0.13.5" ] }, "normalization_method": { "type": "string", "description": "Expression normalization strategy applied to raw counts to enable meaningful cross-sample comparisons. Method choice affects downstream statistical analysis", "examples": [ "TPM", "TMM (edgeR)", "DESeq2 VST", "FPKM", "CPM" ] }, "differential_expression_tool": { "type": "string", "description": "Statistical software and version used for differential expression analysis — optional field since not all RNA-seq experiments include a DE analysis step", "examples": [ "DESeq2 v1.38", "edgeR v3.40", "limma-voom v3.52" ] }, "software_versions": { "type": "array", "description": "Structured version tracking of all software tools used in the analysis pipeline. Each entry documents a tool-version pair to enable exact pipeline replication", "items": { "type": "object", "required": [ "tool", "version" ], "properties": { "tool": { "type": "string", "description": "Name of the software tool used in the analysis pipeline", "examples": [ "STAR", "Salmon", "DESeq2", "fastp" ] }, "version": { "type": "string", "description": "Version number of the software tool", "examples": [ "2.7.10a", "1.9.0", "1.38.0", "0.23.2" ] } } } }, "processing_date": { "type": "string", "format": "date", "description": "Date when primary data processing pipeline was completed, in ISO 8601 format (YYYY-MM-DD). Required for tracking computational reproducibility over time as tool versions and reference databases are updated", "examples": [ "2024-02-05" ] }, "processing_notes": { "type": "string", "description": "Free-text field for additional notes about computational processing decisions, including filtering thresholds applied, multi-mapping read handling strategy, known pipeline deviations, or QC flags raised during processing" } } }, "experimental_design": { "type": "object", "description": "Study-level experimental design parameters critical for statistical analysis model specification, batch effect correction, and accurate result interpretation.", "required": [ "replicates", "comparison_groups", "design_type", "batch_information" ], "properties": { "replicates": { "type": "integer", "minimum": 1, "description": "Number of independent biological replicates per experimental condition. Minimum of 3 biological replicates is recommended for reliable differential expression analysis", "examples": [ 3, 4, 6 ] }, "batch_information": { "type": "object", "description": "Batch processing metadata documenting grouping of samples processed together, required for downstream batch effect identification and correction", "required": [ "batch_id" ], "properties": { "batch_id": { "type": "string", "description": "Unique identifier for the processing batch — typically the date of library preparation or the sequencing run identifier used to group samples processed under identical conditions", "examples": [ "Batch_2024_01", "LibraryPrepLot_42", "SequencingRun_20240115" ] }, "batch_correction_method": { "type": "string", "description": "Statistical method used to correct for batch effects in the expression data, if applicable. Should be documented even if no correction was applied", "examples": [ "ComBat", "limma removeBatchEffect", "ComBat-seq", "none" ] } } }, "comparison_groups": { "type": "array", "description": "Specific pairwise condition contrasts used in differential expression analysis. Each entry should describe the exact biological comparison being made", "items": { "type": "string", "description": "A single pairwise comparison between two experimental conditions, written in the format 'condition A vs condition B'", "examples": [ "treated vs control", "tumor vs normal", "KO vs WT", "high dose vs low dose" ] } }, "design_type": { "type": "string", "enum": [ "paired", "unpaired", "time series", "dose response", "factorial" ], "description": "Experimental design structure that informs the appropriate statistical model for differential expression analysis. Paired designs require a blocking factor in DESeq2 or limma; factorial designs require interaction terms; time series designs require longitudinal modeling" } } }, "metadata": { "type": "object", "description": "Study-level metadata required for FAIR data compliance, public repository submission, and long-term data provenance tracking.", "required": [ "study_id", "experiment_date", "lab", "study_title", "funding_source" ], "properties": { "study_id": { "type": "string", "description": "Unique internal or external identifier for the study, used to link all samples and data files to the same project", "examples": [ "RNASEQ_2024_001", "PROJECT_MCF7_DOX" ] }, "study_title": { "type": "string", "description": "Full descriptive title of the research study or project, as it would appear in a manuscript or public repository submission", "examples": [ "Transcriptomic response of MCF7 cells to doxorubicin treatment" ] }, "experiment_date": { "type": "string", "format": "date", "description": "Date when the primary experiment was performed, in ISO 8601 format (YYYY-MM-DD). Refers to the date of sample collection or treatment, not library preparation or sequencing", "examples": [ "2024-01-10" ] }, "lab": { "type": "string", "description": "Name of the research laboratory or institution where the experiment was conducted, including department and institution name where possible", "examples": [ "Transcriptomics Lab, Department of Molecular Biology, Caltech" ] }, "researcher": { "type": "string", "description": "Name of the primary researcher or principal investigator responsible for the experiment", "examples": [ "Dr. Jane Smith", "Prof. John Doe" ] }, "funding_source": { "type": "string", "description": "Funding agency or grant number supporting the research. Required for FAIR compliance and manuscript acknowledgement", "examples": [ "NIH R01 GM123456", "ERC Starting Grant 987654", "NSF Award 1234567" ] }, "public_repository_accession": { "type": "string", "description": "Accession number in a public data repository where the raw and processed data have been deposited for open access", "examples": [ "GSE123456", "SRP123456", "PRJNA123456", "ERP123456" ] }, "submission_date": { "type": "string", "format": "date", "description": "Date when the data was submitted to a public repository, in ISO 8601 format (YYYY-MM-DD)", "examples": [ "2024-02-10" ] }, "project_id": { "type": "string", "description": "Consortium, collaborative project, or funding programme identifier linking this study to a broader research initiative", "examples": [ "ENCODE_Phase4", "GTEx_v9", "TCGA_BRCA" ] }, "notes": { "type": "string", "description": "Free-text field for any additional observations, caveats, or contextual information about the study that does not fit into structured fields" } } } }, "unevaluatedProperties": false }