cdot vs UTA

Universal transcript archive is an excellent and ambitious project that:

• Performs its own mapping of transcript sequences to reference genomes
• Stores the transcript version data (exons etc) in a SQL database

This has some advantages, namely that you can resolve a GRCh37 coordinate for a transcript which was never officially released for that build.

However the complexity causes a few downsides:

• Alignments may not exactly match those in official Ensembl/RefSeq releases
• Local install requires a PostgreSQL installation
• Internet hosted UTA is a PostgreSQL server, so requires client Postgres libraries, is inaccessible behind firewalls. They have been planning on building a REST server since 2014
• High complexity manual process for releases means they do not support Ensembl and take a while to make RefSeq releases.