Related references

To retrieve the related reference IDs the following steps are performed:

• Get a summary of the reference id from the NCBI using the Eutils ESummary endpoint (for example LRG_303).

  • We consider the nucleotide database in the query.

  • If in the response the accessionversion is different than the input reference id, we include it within the related reference ids (NG_008376.4 for the previous LRG_303 example).

  • The assemblyacc value is also considered as related (AC254562.1 for LRG_303).

  • If a new version is available, the replacedby key is present in the response (see NG_012337.1). In this case we use the ESummary endpoint recursively to retrieve all the newer versions.

• Get the NCBI linked uids by using the Eutils ELink endpoint (for example LRG_303).

  • We use nucleotide in both database parameters (db and dbfrom).

  • We extract all the NCBI uids from the linksetdbs.

    • Note that for chromosomes we do not want all the transcript ids. For this reason, if genome equals chromosome in the previous ESummary response (for example NC_000022.11), we consider only those for which nuccore_nuccore_comp and nuccore_nuccore_rsgb appear as linknames in the ELink response.

• Use the ESummary endpoint with the NCBI uids extracted in the previous step to obtain their reference ids (for example LRG_303).

• If the reference id is from a transcript we make use of the NCBI Datasets REST API to obtain further related references (see NM_003002.2).

• We query the Ensembl REST API to obtain their related references (see LRG_303). Momentarily, we consider the entries for which dbname is either ENS_LRG_gene, LRG or`Ens_Hs_gene`.