{
    "provenance_domain": {
        "name": "Human UniProtKB Xref GDLPA",
        "version": "v-2.0.3",
        "created": "2022-09-06T16:17:53.681025",
        "modified": "2022-09-26T09:08:12.669053",
        "contributors": [
            {
                "affiliation": "The George Washington University",
                "contribution": [
                    "createdBy"
                ],
                "name": "Jeet Vora",
                "email": "jeetvora@gwu.edu"
            },
            {
                "affiliation": "The George Washington University",
                "contribution": [
                    "createdBy"
                ],
                "name": "Robel Kahsay",
                "email": "rykahsay@gwu.edu"
            },
            {
                "affiliation": "The George Washington University",
                "contribution": [
                    "createdBy"
                ],
                "name": "Karina Martinez",
                "email": "karinamartinez@gwu.edu"
            }
        ],
        "license": "https: //creativecommons.org/licenses/by/4.0/"
    },
    "usability_domain": [
        "The dataset contains human [taxid:9606] UniProtKB canonical accessions cross-referenced to Gene and Drug Landing Page Aggregator resource. If you use this dataset please provide proper attribution to Gene and Drug Landing Page Aggregator and GlyGen",
        "Gene and Drug Landing Page Aggregator (GDLPA) has links to 53 gene, 18 variant and 19 drug repositories that provide direct links to gene and drug landing pages. You can search by gene or drug name and then choose the sites that contain knowledge about your gene or drug of interest. Resources supported by the NIH Common Fund are listed first and have the CFDE logo at their top right corner - https://cfde-gene-pages.cloud/."
    ],
    "extension_domain": [
        {
            "extension_schema": "http://www.w3id.org/biocompute/extension_domain/1.2.0/dataset/dataset_extension.json",
            "dataset_extension": {
                "additional_license": {
                    "data_license": "https://creativecommons.org/licenses/by/4.0/",
                    "script_license": "https://www.gnu.org/licenses/gpl-3.0.en.html"
                },
                "dataset_categories": [
                    {
                        "category_value": "Homo sapiens",
                        "category_name": "species"
                    },
                    {
                        "category_value": "Protein",
                        "category_name": "molecule"
                    },
                    {
                        "category_value": "GDLPA cross-references",
                        "category_name": "tags"
                    },
                    {
                        "category_value": "csv",
                        "category_name": "file_type"
                    },
                    {
                        "category_value": "Reviewed",
                        "category_name": "status"
                    },
                    {
                        "category_value": "internal",
                        "category_name": "scope"
                    }
                ]
            }
        }
    ],
    "description_domain": {
        "keywords": [
            "Gene and Drug Landing Page Aggregator",
            "GDLPA",
            "cross-references",
            "gene",
            "drug",
            "CFDE"
        ],
        "pipeline_steps": [
            {
                "step_number": 1,
                "name": "make-protein-dataset.py",
                "description": "The script extracts the gene names from the dataset and maps to the GlyGen's UniProtKB canonical accessions to generate a csv dataset.",
                "version": "",
                "prerequisite": [],
                "input_list": [
                    {
                        "uri": "https://data.glygen.org/ln2data/releases/data/v-2.0.3/reviewed/human_protein_masterlist.csv",
                        "filename": "human_protein_masterlist.csv"
                    },
                    {
                        "uri": "https://data.glygen.org/ln2data/releases/data/v-2.0.3/compiled/human_protein_xref_cfde_gdlpa.csv",
                        "filename": "human_protein_xref_cfde_gdlpa.csv"
                    }
                ],
                "output_list": [
                    {
                        "uri": "https://data.glygen.org/ln2data/releases/data/v-2.0.3/reviewed/human_protein_xref_cfde_gdlpa.csv",
                        "filename": "human_protein_xref_cfde_gdlpa.csv"
                    }
                ]
            }
        ]
    },
    "execution_domain": {
        "script": [
            {
                "uri": {
                    "uri": "https://github.com/glygener/glygen-backend-integration/blob/master/dataset-maker/make-protein-dataset.py",
                    "filename": "make-protein-dataset.py"
                }
            }
        ],
        "script_driver": "manual",
        "software_prerequisites": [
            {
                "version": "2.7.5",
                "name": "Python",
                "uri": {
                    "uri": "https://www.python.org/download/releases/2.7.5/"
                }
            }
        ],
        "external_data_endpoints": [],
        "environment_variables": {}
    },
    "parametric_domain": [
        {
            "param": "",
            "value": "",
            "step": ""
        }
    ],
    "io_domain": {
        "input_subdomain": [
            {
                "uri": {
                    "filename": "human_protein_xref_cfde_gdlpa.csv",
                    "uri": "https:/data.glygen.org/ln2releases/data/v-2.0.3/compiled/human_protein_xref_cfde_gdlpa.csv"
                }
            }
        ],
        "output_subdomain": [
            {
                "mediatype": "csv",
                "uri": {
                    "uri": "https:/data.glygen.org/ln2releases/data/v-2.0.3/reviewed/human_protein_xref_cfde_gdlpa.csv",
                    "filename": "human_protein_xref_cfde_gdlpa.csv"
                }
            },
            {
                "mediatype": "csv",
                "uri": {
                    "uri": "https:/data.glygen.org/ln2releases/data/v-2.0.3/reviewed/human_protein_xref_cfde_gdlpa.stat.csv",
                    "filename": "human_protein_xref_cfde_gdlpa.stat.csv"
                }
            }
        ]
    },
    "error_domain": {
        "empirical_error": {},
        "algorithmic_error": {}
    },
    "object_id": "https://biocomputeobject.org/GLY_000693/v-2.0.3",
    "spec_version": "1.3.0",
    "etag": "05f7e77fd0b36dd54a6d9f9b0c82c2a3"
}