-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
7 changed files
with
1,653 additions
and
0 deletions.
There are no files selected for viewing
Binary file modified
BIN
+0 Bytes
(100%)
NCBImeta/config/__pycache__/paper-phylogeography_config.cpython-35.pyc
Binary file not shown.
Binary file modified
BIN
+8 Bytes
(100%)
NCBImeta/config/__pycache__/paper-phylogeography_config.cpython-37.pyc
Binary file not shown.
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,191 @@ | ||
| OUTPUT_DIR = "NCBImeta/output" | ||
| EMAIL = "eatonk3@mcmaster.ca" | ||
| DATABASE = "yersinia_pestis_db.sqlite" | ||
| TABLES = ["Nucleotide"] | ||
| SEARCH_TERMS = {"Nucleotide": "(plague OR pestis OR Yersinia pestis) AND WGS_MASTER[KYWD] AND nucleotide assembly[Filter]"} | ||
|
|
||
| TABLE_COLUMNS = { | ||
| "Assembly" : [ | ||
| {"AssemblyAccession" : "AssemblyAccession"}, | ||
| {"AssemblyBioSampleAccession" : 'BioSampleAccn'}, | ||
| {"AssemblyBioSampleID" : 'BioSampleId'}, | ||
| {"AssemblyGenbankBioprojectAccession" : ["GB_BioProjects","BioprojectAccn"]}, | ||
| {"AssemblyGenbankID" : 'GbUid'}, | ||
| {"AssemblyRefseqBioprojectAccession" : ["RS_BioProjects","BioprojectAccn"]}, | ||
| {"AssemblyRefSeqCategory" : 'RefSeq_category'}, | ||
| {"AssemblyRefSeqID" : 'RsUid'}, | ||
| {"AssemblyWGSAccession" : 'WGS'}, | ||
| {"AssemblyInfraspecies" : ["InfraspeciesList","Biosource","Sub_value"]}, | ||
| {"AssemblyIsolate" : "Isolate"}, | ||
| {"AssemblyOrganism" : 'Organism'}, | ||
| {"AssemblySpeciesTaxonomicID" : 'SpeciesTaxid'}, | ||
| {"AssemblySpeciesName" : 'SpeciesName'}, | ||
| {"AssemblyTaxonomicID" : 'Taxid'}, | ||
| {"AssemblyName" : "AssemblyName"}, | ||
| {"AssemblyStatus" : 'AssemblyStatus'}, | ||
| {"AssemblyType" : 'AssemblyType'}, | ||
| {"AssemblyCoverage" : 'Coverage'}, | ||
| {"AssemblyChromosomes" : ["Stat", "chromosome_count", "category"]}, | ||
| {"AssemblyContigCount": ["Stat", "contig_count", "category"]}, | ||
| {"AssemblyContigN50" : ["Stat", "contig_n50", "category"]}, | ||
| {"AssemblyContigL50" : ["Stat", "contig_l50", "category"]}, | ||
| {"AssemblyNonChromosomalReplicons" : ["Stat", "non_chromosome_replicon_count", "category"]}, | ||
| {"AssemblyReplicons" : ["Stat", "replicon_count", "category"]}, | ||
| {"AssemblyScaffolds" : ["Stat", "scaffold_count", "category"]}, | ||
| {"AssemblyScaffoldN50" : ["Stat", "scaffold_n50", "category"]}, | ||
| {"AssemblyScaffoldL50" : ["Stat", "scaffold_l50", "category"]}, | ||
| {"AssemblyTotalLength" : ["Stat", "total_length", "category"]}, | ||
| {"AssemblyUngappedLength" : ["Stat", "ungapped_length", "category"]}, | ||
| {"AssemblySubmitterOrganization" : 'SubmitterOrganization'}, | ||
| {"AssemblySubmissionDate" : 'SubmissionDate'}, | ||
| {"AssemblyReleaseDate" : 'SeqReleaseDate'}, | ||
| {"AssemblyFTPAssemblyReport" : 'FtpPath_Assembly_rpt'}, | ||
| {"AssemblyFTPGenbank" : 'FtpPath_GenBank'}, | ||
| {"AssemblyFTPRefSeq" : 'FtpPath_RefSeq'}, | ||
| {"AssemblyFTPStatsReport" : 'FtpPath_Stats_rpt'}, | ||
| {"AssemblyComment" : 'NullValue'} | ||
| ], | ||
| "BioSample" : [ | ||
| {"BioSampleAccession": "Accession"}, | ||
| {"BioSampleAccessionSecondary" : "NullValue"}, | ||
| {"BioSampleBioProjectAccession": ["Link","label"]}, | ||
| {"BioSampleSRAAccession": ["Id","SRA","db"]}, | ||
| {"BioSampleTitle": "Title"}, | ||
| {"BioSampleName": ["Id","Sample name","db_label"]}, | ||
| {"BioSampleType": ["Attribute","sample_type","harmonized_name"]}, | ||
| {"BioSamplePackage": "Package"}, | ||
| {"BioSampleInfraspecies": "Infraspecies"}, | ||
| {"BioSampleOrganism": "OrganismName"}, | ||
| {"BioSampleOrganismAlt": ["Organism","taxonomy_name"]}, | ||
| {"BioSampleSubSpecies": ["Attribute","sub_species","harmonized_name"]}, | ||
| {"BioSampleStrain": ["Attribute","strain","harmonized_name"]}, | ||
| {"BioSampleTaxonomyID": "Taxonomy"}, | ||
| {"BioSampleBiovar": ["Attribute","biovar","harmonized_name"]}, | ||
| {"BioSampleCollectionDate": ["Attribute","collection_date","harmonized_name"]}, | ||
| {"BioSampleGeographicLocation": ["Attribute","geo_loc_name","harmonized_name"]}, | ||
| {"BioSampleHost": ["Attribute","host","harmonized_name"]}, | ||
| {"BioSampleHostDisease" : ["Attribute", "host_disease", "harmonized_name"]}, | ||
| {"BioSampleIsolateNameAlias": ["Attribute","isolate_name_alias","harmonized_name"]}, | ||
| {"BioSampleIsolationSource": ["Attribute","isolation_source","harmonized_name"]}, | ||
| {"BioSampleLat" : ["Attribute", "latitude", "harmonized_name"]}, | ||
| {"BioSampleLatLon" : ["Attribute", "lat_lon", "harmonized_name"]}, | ||
| {"BioSampleLon" : ["Attribute", "longitude", "harmonized_name"]}, | ||
| {"BioSampleDate": "Date"}, | ||
| {"BioSampleModificationDate": "ModificationDate"}, | ||
| {"BioSamplePublicationDate": "PublicationDate"}, | ||
| {"BioSampleOrganization": "Organization"}, | ||
| {"BioSampleComment" : "NullValue"}, | ||
| ], | ||
| "BioProject" : [ | ||
| {"BioProjectAccession" : "Project_Acc"}, | ||
| {"BioProjectDataType" : "Project_Data_Type"}, | ||
| {"BioProjectDescription" : "Project_Description"}, | ||
| {"BioProjectMethodType" : "Project_MethodType"}, | ||
| {"BioProjectName" : "Project_Name"}, | ||
| {"BioProjectTargetCapture" : "Project_Target_Capture"}, | ||
| {"BioProjectTargetMaterial" : "Project_Target_Material"}, | ||
| {"BioProjectTargetScope" : "Project_Target_Scope"}, | ||
| {"BioProjectTitle" : "Project_Title"}, | ||
| {"BioProjectType" : "Project_Type"}, | ||
| {"BioProjectOrganismLabel" : "Organism_Label"}, | ||
| {"BioProjectOrganismStrain" : "Organism_Strain"}, | ||
| {"BioProjectSupergroup" : "Supergroup"}, | ||
| {"BioProjectTaxonomicID" : "TaxId"}, | ||
| {"BioProjectRegistrationDate" : "Registration_Date"}, | ||
| {"BioProjectRelevanceMedical" : "Relevance_Medical"}, | ||
| {"BioProjectSequencingStatus" : "Sequencing_Status"}, | ||
| {"BioProjectSubmitterOrganization" : "Submitter_Organization"}, | ||
| {"BioProjectComment" : "NullValue"} | ||
| ], | ||
| "Nucleotide" : [ | ||
| {"NucleotideAccession" : 'GBSeq_primary-accession'}, | ||
| {"NucleotideAccessionVersion" : 'GBSeq_accession-version'}, | ||
| {"NucleotideBioSampleAccession": "NucleotideBioSample"}, | ||
| {"NucleotideBioProjectAccession" : "GBSeq_project"}, | ||
| {"NucleotideFirstAccession" : ['GBAltSeqData_items', 'GBSeq_alt-seq', 'GBAltSeqItem_first-accn']}, | ||
| {"NucleotideLastAccession" : ['GBAltSeqData_items', 'GBSeq_alt-seq', 'GBAltSeqItem_last-accn']}, | ||
| {"NucleotideOrganism" : 'GBSeq_organism'}, | ||
| {"NucleotideTaxonomy" : "GBSeq_taxonomy"}, | ||
| {"NucleotideDefinition" : "GBSeq_definition"}, | ||
| {"NucleotideDivision" : "GBSeq_division"}, | ||
| {"NucleotideJournal": ['GBSeq_references', 'GBReference_journal']}, | ||
| {"NucleotideLength" : 'GBSeq_length'}, | ||
| {"NucleotideMoleculeType" : "GBSeq_moltype"}, | ||
| {"NucleotideReferenceTitle": ['GBSeq_references', u'GBReference_title']}, | ||
| {"NucleotideSeqDataName" : ['GBSeq_alt-seq', 'GBAltSeqData_name']}, | ||
| {"NucleotideSource" : 'GBSeq_source'}, | ||
| {"NucleotideStrandedness" : "GBSeq_strandedness"}, | ||
| {"NucleotideTopology" : "GBSeq_topology"}, | ||
| {"NucleotideCreateDate" : "GBSeq_create-date"}, | ||
| {"NucleotideUpdateDate" : "GBSeq_update-date"}, | ||
| {"NucleotideGenBankComment" : "GBSeq_comment"}, | ||
| {'NucleotideAnnotationDate': 'Annotation Date'}, | ||
| {'NucleotideAnnotationMethod': 'Annotation Method'}, | ||
| {'NucleotideAnnotationPipeline': 'Annotation Pipeline'}, | ||
| {'NucleotideAnnotationProvider': 'Annotation Provider'}, | ||
| {'NucleotideAnnotationSoftwarerevision': 'Annotation Software revision'}, | ||
| {'NucleotideAssemblyDate': 'Assembly Date'}, | ||
| {'NucleotideAssemblyMethod': 'Assembly Method'}, | ||
| {'NucleotideAssemblyName': 'Assembly Name'}, | ||
| {'NucleotidetCDS': 'CDS (total)'}, | ||
| {'NucleotidetCDSCoding': 'CDS (coding)'}, | ||
| {'NucleotidetCRISPRArrays': 'CRISPR Arrays'}, | ||
| {'NucleotidetExpectedFinalVersion': 'Expected Final Version'}, | ||
| {'NucleotidetFeaturesAnnotated': 'Features Annotated'}, | ||
| {'NucleotidetGenes': 'Genes (total)'}, | ||
| {'NucleotidetGenesCoding': 'Genes (coding)'}, | ||
| {'NucleotidetGenesRNA': 'Genes (RNA)'}, | ||
| {'NucleotidetGenomeCoverage': 'Genome Coverage'}, | ||
| {'NucleotidetGenomeRepresentation': 'Genome Representation'}, | ||
| {'NucleotidetncRNAs': 'ncRNAs'}, | ||
| {'NucleotidetPseudoGenes': 'Pseudo Genes (total)'}, | ||
| {'NucleotidetPseudoGenesAmbResidues': 'Pseudo Genes (ambiguous residues)'}, | ||
| {'NucleotidetPseudoGenesFrameshifted': 'Pseudo Genes (frameshifted)'}, | ||
| {'NucleotidetPseudoGenesIncomplete': 'Pseudo Genes (incomplete)'}, | ||
| {'NucleotidetPseudoGenesInternalStop': 'Pseudo Genes (internal stop)'}, | ||
| {'NucleotidetPseudoGenesMultipleProblems': 'Pseudo Genes (multiple problems)'}, | ||
| {'NucleotidetrRNAs': 'rRNAs'}, | ||
| {'NucleotidetrRNAsComplete': 'complete rRNAs'}, | ||
| {'NucleotidetrRNAsPartial': 'partial rRNAs'}, | ||
| {'NucleotideSequencingTechnology': 'Sequencing Technology'}, | ||
| {'NucleotidetRNAs': 'tRNAs'}, | ||
| {'NucleotideComment' : 'NullValue'} | ||
| ], | ||
| "SRA" : [ | ||
| {"SRABioProjectAccession" : "Bioproject"}, | ||
| {"SRABioSampleAccession" : "Biosample"}, | ||
| {"SRAExperimentAccession" : ["Experiment", "acc"]}, | ||
| {"SRARunAccession" : ["Run", "acc"]}, | ||
| {"SRASampleAccession" : ["Sample", "acc"]}, | ||
| {"SRAExperimentName" : ["Experiment", "name"]}, | ||
| {"SRAExperimentStatus" : ["Experiment", "status"]}, | ||
| {"SRAExperimentVersion" : ["Experiment", "ver"]}, | ||
| {"SRAIsPublic" : ["Run", "is_public"]}, | ||
| {"SRASampleName" : ["Sample", "name"]}, | ||
| {"SRAStaticDataAvailable" : ["Run", "static_data_available"]}, | ||
| {"SRAStudyAcc" : ["Study", "acc"]}, | ||
| {"SRAStudName" : ["Study", "name"]}, | ||
| {"SRATitle" : "Title"}, | ||
| {"SRAOrganismName" : ["Organism", "CommonName"]}, | ||
| {"SRAOrganismTaxID" : ["Organism", "taxid"]}, | ||
| {"SRAClusterName" : ["Statistics", "cluster_name"]}, | ||
| {"SRAInstrumentModel" : ["Platform", "instrument_model"]}, | ||
| {"SRALibraryName" : "LIBRARY_NAME"}, | ||
| {"SRALibraryLayout" : "LIBRARY_LAYOUT"}, | ||
| {"SRALibrarySelection" : "LIBRARY_SELECTION"}, | ||
| {"SRALibrarySource" : "LIBRARY_SOURCE"}, | ||
| {"SRALibraryStrategy" : "LIBRARY_STRATEGY"}, | ||
| {"SRAPlatform" : "Platform"}, | ||
| {"SRATotalBases" : ["Statistics", "total_bases"]}, | ||
| {"SRATotalSize" : ["Statistics", "total_size"]}, | ||
| {"SRATotalSpots" : ["Statistics", "total_spots"]}, | ||
| {"SRATotalRuns" : ["Statistics", "total_runs"]}, | ||
| {"SRACreateDate" : "CreateDate"}, | ||
| {"SRAUpdateDate" : "UpdateDate"}, | ||
| {"SRACenterName" : ["Submitter", "center_name"]}, | ||
| {"SRAContactName" : ["Submitter", "contact_name"]}, | ||
| {"SRALabName" : ["Submitter", "lab_name"]}, | ||
| {"SRASubmitterAccession" : ["Submitter", "acc"]}, | ||
| {'SRAComment' : 'NullValue'} | ||
| ] | ||
| } |
Binary file not shown.
Binary file not shown.
Oops, something went wrong.