A tool for collections-specific searches in genetic databases

Abstract

It is becoming increasingly important for museums and other scientific collections to quantify the amount of genetic resources being derived from their holdings. Genetic database records, such as GenBank and Barcode of Life (BOLD), have an optional field for indicating the specimen that it derived from, and, on the other side, specimen databases, such as GBIF (gbif.org) and iDigBio (idigbio.org), have an optional field for indicating sequence records that were derived from it. Making connections between the two types of records should be easy, but unfortunately they are made difficult by inconsistent standards. For example, GenBank has a catch-all "country" term that holds all geographic locality data for a specimen, whereas in Darwin Core (DwC) there are 12 atomized levels of locality names. The software tool described here was originally created for Smithsonian data managers to search genetic databases in a targeted manner for DNA sequences generated from Smithsonian specimens. It is being made open source to be utilized by other scientific institutions to quantify and document the genetic impact of their collections. Other potential uses include checking for data inconsistencies between sequence records and specimen records, and enforcing specimen loan agreements.