NCBI Minute: New Version of E-utilities Supports Accession.version Identifiers

The webinar introduces a new functionality within the E-utilities API that supports accession.version identifiers. The presenter provides access to materials such as a PowerPoint and additional resources via an FTP site. The webinar is part of a series discussing changes in how NCBI handles sequence identifiers. The E-utilities API now fully supports accession.version identifiers, both as input and output. The presenter outlines the transition from using G.I. numbers to accession.version identifiers and introduces the 'idtype' parameter that controls this transition across various e-utilities.

The E-utilities are explained as tools that work across various Entrez databases, providing functions like ESearch, ESummary, EFetch, ELink, and EPost. These tools handle unique identifiers (UIDs) for records. The introduction of accession.version as UIDs is detailed, highlighting that while G.I. numbers are being phased out, they are not deleted and will remain accessible. The new functionality allows for the use of accession.version identifiers in place of G.I. numbers in ESummary and EFetch, and the ability for ESearch to output accession.version. The paragraph also covers the impact on UIDs and the upgrade in functionality for the APIs.

The behavior of ELink and ESearch with the new accession.version identifiers is discussed. ELink can now accept accession numbers as input and return accessions in the output. ESearch, when provided with the 'idtype=acc' parameter, will return accession.versions instead of GI numbers. The paragraph also addresses the issue of sequences without G.I. numbers, which are not part of the Entrez system and thus not accessible through ESearch, ESummary, or ELink, but can now be downloaded using EFetch.

The presenter discusses the new capability of EFetch to access GI-less sequences using their accession.version identifiers. This includes Whole Genome Shotgun (WGS) and Transcriptome Shotgun Assembly (TSA) datasets, many of which do not have GI numbers for individual contigs. The Norway Spruce project is highlighted as an example of a large dataset without GI numbers, where only the master record has a GI number. The presenter also explains how to check if a sequence has a GI number using ESummary and provides a Perl script for exploring the new functionalities.

The presenter demonstrates the use of EFetch to obtain data for GI-less sequences, such as the Norway Spruce WGS data. They also show how to use ESummary to check if a sequence has a GI number by including it in a mixed list of identifiers. The script provided on the FTP site is mentioned for attendees to experiment with the new functionalities. The presenter encourages questions and further engagement through social media and subscription to the utilities-announce@ncbi mailing list for updates.

The presenter addresses questions from the audience, including the best way to programmatically obtain the FTP location for entire WGS assemblies using BioProject or by searching through the nuccore database. They also discuss the current behavior of ESummary regarding the order of input UIDs and the presence of GI-less accession.versions. The webinar concludes with an invitation for further inquiries and an offer of assistance for any issues or suggestions regarding the new functionalities.