{
  "schema_version": 1,
  "updated": "2026-10-07",
  "feature_scopes": {
    "all": {"label": "Complete Daphnia Browser bibliography", "keys": ["nies_assembly", "nies_genome_paper", "lrv_assembly", "lrv_genome_paper", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly", "refseq", "geo", "gse150820", "gse264123", "gse264123_zenodo", "anatomy_asset", "kallisto", "tximport", "deseq2", "bh_fdr", "efp", "uniprot", "diamond", "famsa", "fasttree", "interproscan", "interpro", "interpro_members", "pfam", "panther", "orthofinder2019", "orthofinder2026", "mcl", "blastplus", "doench2016", "azimuth_software", "biopython", "crisprscore"]},
    "genome": {"label": "Genome search, annotation map, and sequence retrieval", "keys": ["nies_assembly", "nies_genome_paper", "lrv_assembly", "lrv_genome_paper", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly", "refseq"]},
    "nies_annotation": {"label": "Precomputed NIES correspondence and supplementary annotation", "description": "DIAMOND 2.2.4, --sensitive, up to 10 targets, search E-value 1e-3; nies-rbh-v2. Strict gene-level RBH: both directions E-value <=1e-10, identity >=30%, both coverages >=70%, aligned length >=50 aa. Distinct NIES genes within 98% of the top bit score are ambiguous. Partial/one-way/weak hits do not transfer a function. Generic NIES products remain ID-only. Gene rows summarize the strongest isoform; inspect gene_mapping_consistency. Correspondence is not proof of orthology. Only the seven indexed Daphnia annotations are supported; no new sequence search is run.", "keys": ["diamond", "nies_assembly", "nies_genome_paper", "lrv_assembly", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly", "refseq"]},
    "atlas": {"label": "Bulk anatomical-compartment RNA-seq atlas and Excel export", "keys": ["gse264123", "gse264123_zenodo", "geo", "efp", "diamond", "anatomy_asset"]},
    "embryo_expression": {"label": "Embryo RNA-seq expression and differential-expression evidence", "keys": ["gse150820", "geo", "kallisto", "tximport", "deseq2", "bh_fdr"]},
    "quick_annotation": {"label": "Reviewed-homolog annotation search", "keys": ["diamond", "uniprot"]},
    "phylogeny": {"label": "Exploratory protein homolog placement", "keys": ["diamond", "uniprot", "famsa", "fasttree"]},
    "within_species_phylogeny": {"label": "Exploratory target-proteome top-hit tree (within-genome or cross-dataset query placement)", "keys": ["diamond", "famsa", "fasttree", "nies_assembly", "nies_genome_paper", "lrv_assembly", "lrv_genome_paper", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly", "refseq"]},
    "interpro": {"label": "Protein signatures, domains, GO and pathway mappings", "keys": ["interproscan", "interpro", "interpro_members", "pfam", "panther"]},
    "orthofinder": {"label": "Cross-assembly correspondence and multi-species orthology evidence", "keys": ["orthofinder2019", "orthofinder2026", "diamond", "mcl", "famsa", "fasttree"]},
    "sequence_export": {"label": "FASTA, GenBank and protein sequence export", "keys": ["nies_assembly", "nies_genome_paper", "lrv_assembly", "lrv_genome_paper", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly", "refseq"]},
    "primer": {"label": "PCR primer design and whole-genome specificity screen", "keys": ["blastplus", "nies_assembly", "lrv_assembly", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly"]},
    "crispr": {"label": "SpCas9 guide ranking with validated Rule Set 2 and CFD calculations", "keys": ["doench2016", "azimuth_software", "biopython", "crisprscore", "blastplus", "nies_assembly", "lrv_assembly", "magna_xinb3_assembly", "carinata_assembly", "sinensis_assembly", "pulicaria_assembly", "pulex_kap4_assembly"]}
  },
  "entries": {
    "nies_assembly": {"type": "dataset", "short": "NCBI RefSeq assembly GCF_020631705.1", "citation": "Daphnia magna NIES, assembly ASM2063170v1.1, RefSeq accession GCF_020631705.1; NCBI Daphnia magna Annotation Release 101.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_020631705.1/", "accessed": "2026-08-11"},
    "nies_genome_paper": {"type": "genome publication", "short": "NIES reference genome publication", "citation": "Byeon E et al. (2022) The freshwater water flea Daphnia magna NIES strain genome as a resource for CRISPR/Cas9 gene targeting: The glutathione S-transferase omega 2 gene. Aquatic Toxicology 242:106021.", "doi": "10.1016/j.aquatox.2021.106021"},
    "lrv_assembly": {"type": "dataset", "short": "NCBI GenBank assembly GCA_030254905.1", "citation": "Daphnia magna LRV0_1, assembly UOB_LRV0_1, GenBank accession GCA_030254905.1, with the bundled original Dmagna-ID annotation crosswalk.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCA_030254905.1/", "accessed": "2026-08-11", "note": "The legacy Dmagna-ID crosswalk is a local derived resource; cite the original annotation source if used in a publication."},
    "lrv_genome_paper": {"type": "genome publication", "short": "LRV0_1 reference genome publication", "citation": "Chaturvedi A et al. (2023) The hologenome of Daphnia magna reveals possible DNA methylation and microbiome-mediated evolution of the host genome. Nucleic Acids Research 51:9785-9803.", "doi": "10.1093/nar/gkad685"},
    "magna_xinb3_assembly": {"type": "dataset", "short": "NCBI GenBank assembly GCA_001632505.1", "citation": "Daphnia magna strain Xinb3, assembly daphmag2.4, GenBank accession GCA_001632505.1; Don Gilbert Daphnia magna Annotation 1.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCA_001632505.1/", "accessed": "2026-08-21", "note": "APZ42 is the locus-tag prefix in this annotation, not the strain name. The assembly is scaffold-level."},
    "carinata_assembly": {"type": "dataset", "short": "NCBI RefSeq assembly GCF_022539665.2", "citation": "Daphnia carinata strain CSIRO-1, assembly CSIRO_AGI_Dcar_HiC_V3, RefSeq accession GCF_022539665.2; annotation GCF_022539665.2-RS_2023_09.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_022539665.2/", "accessed": "2026-08-21"},
    "sinensis_assembly": {"type": "dataset", "short": "NCBI GenBank assembly GCA_013167095.2", "citation": "Daphnia sinensis strain WSL, assembly Dsi, GenBank accession GCA_013167095.2; annotation submitted by Huazhong Agricultural University.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCA_013167095.2/", "accessed": "2026-08-21"},
    "pulicaria_assembly": {"type": "dataset", "short": "NCBI RefSeq assembly GCF_021234035.1", "citation": "Daphnia pulicaria isolate SC F1-1A, assembly SC_F0-13Bv2, RefSeq accession GCF_021234035.1; NCBI Daphnia pulicaria Annotation Release 100.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_021234035.1/", "accessed": "2026-08-21"},
    "pulex_kap4_assembly": {"type": "dataset", "short": "NCBI RefSeq assembly GCF_021134715.1", "citation": "Daphnia pulex isolate KAP4, assembly ASM2113471v1, RefSeq accession GCF_021134715.1; NCBI Daphnia pulex Annotation Release 100.", "url": "https://www.ncbi.nlm.nih.gov/datasets/genome/GCF_021134715.1/", "accessed": "2026-08-21"},
    "refseq": {"type": "database", "short": "NCBI RefSeq", "citation": "O'Leary NA et al. (2016) Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res 44:D733-D745.", "doi": "10.1093/nar/gkv1189"},
    "geo": {"type": "database", "short": "NCBI Gene Expression Omnibus", "citation": "Edgar R, Domrachev M, Lash AE (2002) Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res 30:207-210.", "doi": "10.1093/nar/30.1.207"},
    "gse150820": {"type": "dataset", "short": "GSE150820 / Nong et al. 2020", "citation": "Nong QD, Matsuura T, Kato Y, Watanabe H (2020) Two Doublesex1 mutants revealed a tunable gene network underlying intersexuality in Daphnia magna. PLOS ONE 15:e0238256. GEO: GSE150820.", "doi": "10.1371/journal.pone.0238256", "url": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE150820"},
    "gse264123": {"type": "dataset", "short": "GSE264123 / Dua & Yampolsky 2025", "citation": "Dua I, Yampolsky LY (2025) Transcriptional atlas of Daphnia magna. Comp Biochem Physiol Part D 55:101504. GEO: GSE264123.", "doi": "10.1016/j.cbd.2025.101504", "url": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE264123"},
    "gse264123_zenodo": {"type": "reference dataset", "short": "XINB3 v3.0 reference used for atlas mapping", "citation": "Fields P, Ebert D (2024) Daphnia magna XINB3 v.3.0 genome assembly (FI-XINB3). Zenodo record 11283641.", "doi": "10.5281/zenodo.11283641", "note": "This is the reference assembly/annotation used to connect GSE264123 atlas transcripts; it is not the RNA-seq study itself."},
    "anatomy_asset": {"type": "asset provenance", "short": "Author-created Daphnia anatomy schematic", "citation": "Anatomical schematic of adult Daphnia magna generated with AI-assisted image generation using ChatGPT (OpenAI, 2026) and manually refined by the authors based on D. magna anatomy.", "year": "2026", "note": "Author-created atlas schematic. ChatGPT was used as an image-generation aid; the authors manually refined the anatomy and retain responsibility for the final figure. No separate open-content license is asserted."},
    "kallisto": {"type": "software", "short": "kallisto", "citation": "Bray NL, Pimentel H, Melsted P, Pachter L (2016) Near-optimal probabilistic RNA-seq quantification. Nat Biotechnol 34:525-527.", "doi": "10.1038/nbt.3519", "version": "0.46.1"},
    "tximport": {"type": "software", "short": "tximport", "citation": "Soneson C, Love MI, Robinson MD (2015) Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Research 4:1521.", "doi": "10.12688/f1000research.7563.1"},
    "deseq2": {"type": "software", "short": "DESeq2", "citation": "Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550.", "doi": "10.1186/s13059-014-0550-8"},
    "bh_fdr": {"type": "method", "short": "Benjamini-Hochberg FDR", "citation": "Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 57:289-300.", "doi": "10.1111/j.2517-6161.1995.tb02031.x"},
    "efp": {"type": "visualization method", "short": "eFP visualization concept", "citation": "Winter D, Vinegar B, Nahal H, Ammar R, Wilson GV, Provart NJ (2007) An Electronic Fluorescent Pictograph Browser for Exploring and Analyzing Large-Scale Biological Data Sets. PLOS ONE 2:e718.", "doi": "10.1371/journal.pone.0000718", "note": "Daphnia Browser uses a custom implementation inspired by the eFP visual concept; it does not reuse the BAR eFP software."},
    "uniprot": {"type": "database", "short": "UniProtKB/Swiss-Prot", "citation": "The UniProt Consortium (2025) UniProt: the Universal Protein Knowledgebase in 2025. Nucleic Acids Res 53:D609-D617.", "doi": "10.1093/nar/gkae1010", "note": "Report the local Swiss-Prot snapshot/release and each stable primary accession used."},
    "diamond": {"type": "software", "short": "DIAMOND", "citation": "Buchfink B, Reuter K, Drost HG (2021) Sensitive protein alignments at tree-of-life scale using DIAMOND. Nat Methods 18:366-368.", "doi": "10.1038/s41592-021-01101-x", "version": "2.2.4 on Jetstream2"},
    "famsa": {"type": "software", "short": "FAMSA", "citation": "Deorowicz S, Debudaj-Grabysz A, Gudys A (2016) FAMSA: Fast and accurate multiple sequence alignment of huge protein families. Sci Rep 6:33964.", "doi": "10.1038/srep33964"},
    "fasttree": {"type": "software", "short": "FastTree 2", "citation": "Price MN, Dehal PS, Arkin AP (2010) FastTree 2 - Approximately Maximum-Likelihood Trees for Large Alignments. PLOS ONE 5:e9490.", "doi": "10.1371/journal.pone.0009490"},
    "interproscan": {"type": "software", "short": "InterProScan 5", "citation": "Jones P et al. (2014) InterProScan 5: genome-scale protein function classification. Bioinformatics 30:1236-1240.", "doi": "10.1093/bioinformatics/btu031", "version": "5.78-109.0"},
    "interpro": {"type": "database", "short": "InterPro", "citation": "Blum M et al. (2025) InterPro: the protein sequence classification resource in 2025. Nucleic Acids Res 53:D444-D456.", "doi": "10.1093/nar/gkae1082"},
    "interpro_members": {"type": "database guidance", "short": "InterPro member databases", "citation": "InterProScan integrates signatures from member databases such as Pfam, PANTHER, Gene3D and MobiDB-lite. When a specific signature is interpreted, also cite that member database using the official InterPro citation guidance.", "url": "https://interproscan-docs.readthedocs.io/en/v5/Citing.html"},
    "pfam": {"type": "member database", "short": "Pfam", "citation": "Mistry J et al. (2021) Pfam: The protein families database in 2021. Nucleic Acids Res 49:D412-D419.", "doi": "10.1093/nar/gkaa913", "note": "Relevant when a displayed InterProScan result contains a Pfam signature."},
    "panther": {"type": "member database", "short": "PANTHER", "citation": "Mi H et al. (2021) PANTHER version 16: a revised family classification, tree-based classification tool, enhancer regions and extensive API. Nucleic Acids Res 49:D394-D403.", "doi": "10.1093/nar/gkaa1106", "note": "Relevant when a displayed InterProScan result contains a PANTHER signature."},
    "orthofinder2019": {"type": "software", "short": "OrthoFinder method", "citation": "Emms DM, Kelly S (2019) OrthoFinder: phylogenetic orthology inference for comparative genomics. Genome Biol 20:238.", "doi": "10.1186/s13059-019-1832-y"},
    "orthofinder2026": {"type": "software", "short": "OrthoFinder 3", "citation": "Emms DM, Liu Y, Belcher L, Holmes J, Kelly S (2026) OrthoFinder: improved phylogenetic orthology inference with enhanced accuracy and scalability. Nat Methods 23:1327-1333.", "doi": "10.1038/s41592-026-03126-6", "version": "3.1.5"},
    "mcl": {"type": "software/method", "short": "Markov Cluster algorithm", "citation": "Enright AJ, Van Dongen S, Ouzounis CA (2002) An efficient algorithm for large-scale detection of protein families. Nucleic Acids Res 30:1575-1584.", "doi": "10.1093/nar/30.7.1575"},
    "blastplus": {"type": "software", "short": "NCBI BLAST+", "citation": "Camacho C et al. (2009) BLAST+: architecture and applications. BMC Bioinformatics 10:421.", "doi": "10.1186/1471-2105-10-421", "version": "2.17.0 on Jetstream2", "note": "Primer selection/ranking and amplicon pairing are custom Daphnia Browser heuristics, not NCBI Primer-BLAST or Primer3."},
    "doench2016": {"type": "method source", "short": "Doench Rule Set 2 / CFD", "citation": "Doench JG et al. (2016) Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPR-Cas9. Nat Biotechnol 34:184-191.", "doi": "10.1038/nbt.3437", "note": "Daphnia Browser reproduces the released Azimuth V3 no-position Rule Set 2 model and the published CFD mismatch-position/PAM weights. Scores were trained outside Daphnia and are not experimental probabilities."},
    "azimuth_software": {"type": "software", "short": "Microsoft Research Azimuth 2.0", "citation": "Microsoft Research (2016) Azimuth: CRISPR/Cas9 on-target activity prediction software, release 2.0.", "url": "https://github.com/MicrosoftResearch/Azimuth", "version": "2.0; V3_model_nopos", "accessed": "2026-08-13", "license": "BSD-3-Clause", "note": "The bundled static model reproduces all 947 official no-position fixture predictions within 1e-3 (maximum absolute error below 6e-10)."},
    "biopython": {"type": "software", "short": "Biopython", "citation": "Cock PJA et al. (2009) Biopython: freely available Python tools for computational molecular biology and bioinformatics. Bioinformatics 25:1422-1423.", "doi": "10.1093/bioinformatics/btp163", "version": "Azimuth-compatible Tm_staluc implementation"},
    "crisprscore": {"type": "software/data source", "short": "crisprScore CFD reference tables", "citation": "Fortin JP et al. crisprScore: on-target and off-target scoring algorithms for CRISPR guide RNAs. Bioconductor.", "doi": "10.18129/B9.bioc.crisprScore", "url": "https://bioconductor.org/packages/crisprScore", "accessed": "2026-08-13", "note": "The bundled Doench 2016 CFD mismatch and PAM tables were imported from the crisprScore source distribution and verified against its seven Cas9 CFD test cases."}
  }
}