Provide NCBI API keys when possible (#203)

* Provide NCBI API keys when possible

* Fix requirements.txt

bin/insert_new_assembly.py

@@ -53,7 +53,8 @@ def main():
             insert_new_assembly_and_taxonomy(
                 metadata_connection_handle=conn,
                 assembly_accession=assembly_accession,
-                taxonomy_id=taxon_id
+                taxonomy_id=taxon_id,
+                ncbi_api_key=cfg['eutils_api_key']
             )
         elif taxon_id:
             ensure_taxonomy_is_in_evapro(conn, taxon_id)

eva_submission/eload_ingestion.py

@@ -206,6 +206,7 @@ def check_variant_db(self):
                     metadata_connection_handle=conn,
                     assembly_accession=assembly,
                     taxonomy_id=self.taxonomy,
+                    ncbi_api_key=cfg['eutils_api_key']
                 )
 
         for db_info in assembly_to_db_name.values():

eva_submission/eload_utils.py

@@ -54,7 +54,7 @@ def resolve_accession_from_text(reference_text):
     if NCBIAssembly.is_assembly_accession_format(reference_text):
         return [reference_text]
     # Search for a reference genome that resolve this text
-    accession = retrieve_genbank_assembly_accessions_from_ncbi(reference_text)
+    accession = retrieve_genbank_assembly_accessions_from_ncbi(reference_text, api_key=cfg['eutils_api_key'])
     if accession:
         return accession
 

eva_submission/vep_utils.py

@@ -139,7 +139,7 @@ def get_species_and_assembly(assembly_acc):
     Returns None if the taxonomy is not known.
     """
     # We first need to search for the species associated with the assembly
-    assembly_dicts = get_ncbi_assembly_dicts_from_term(assembly_acc)
+    assembly_dicts = get_ncbi_assembly_dicts_from_term(assembly_acc, api_key=cfg['eutils_api_key'])
     taxid_and_assembly_name = set([
         (assembly_dict.get('taxid'), assembly_dict.get('assemblyname'))
         for assembly_dict in assembly_dicts
@@ -253,7 +253,7 @@ def recursive_nlst(ftp, root, pattern):
 
 @retry(tries=4, delay=2, backoff=1.2, jitter=(1, 3), logger=logger)
 def download_and_extract_vep_cache(ftp, vep_cache_file, taxonomy_id):
-    scientific_name = retrieve_species_scientific_name_from_tax_id_ncbi(taxonomy_id)
+    scientific_name = retrieve_species_scientific_name_from_tax_id_ncbi(taxonomy_id, api_key=cfg['eutils_api_key'])
     species_name = scientific_name.replace(' ', '_').lower()
 
     tmp_dir = tempfile.TemporaryDirectory()

eva_submission/xlsx/xlsx_validation.py

@@ -107,7 +107,7 @@ def check_reference_genome(self):
         """Check if the references can be retrieved"""
         references = set([row['Reference'] for row in self.metadata['Analysis'] if row['Reference']])
         for reference in references:
-            accessions = retrieve_genbank_assembly_accessions_from_ncbi(reference)
+            accessions = retrieve_genbank_assembly_accessions_from_ncbi(reference, api_key=cfg['eutils_api_key'])
             if len(accessions) == 0:
                 self.error_list.append(f'In Analysis, Reference {reference} did not resolve to any accession')
             elif len(accessions) > 1:

requirements.txt

@@ -1,6 +1,6 @@
 cached-property
 cerberus
-ebi-eva-common-pyutils[eva-internal]==0.6.4
+ebi-eva-common-pyutils[eva-internal]==0.6.6
 eva-vcf-merge==0.0.8
 humanize
 lxml