We are using previously published RNA-seq data for a new project, which was mapped in 2018 to an older GRCh38 version. Is it worth re-mapping to GRCh38.p14 so we have the most up-to-date build at the onset?
At least so far to my knowledge, I don't think it would be worth re-mapping to GRCh38.p14, if it is too much trouble or with what you're looking at is unaffected by the updates in the minor releases. The minor releases of the assembly (i.e. indicate by the *.p14) are patches to certain parts of the assembly without changing the actual chromosome coordinates (so assuming it would be more related to small changes in annotation). There is some more information about what these patches contain here: NCBI Patches; and a slightly different but somewhat similar question here
We get this question like clockwork every 6 weeks.
I'll say it one more time:
A freeze is a freeze.
Patches are not a thing that 99% of people will need to worry about or can even use in any meaningful manner.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
If you have a one-click workflow ready and the necessary computation power to casually run it, then go for it. If it is any greater effort or costs money then no. It is just a minor update within an existing genome. Still, if data are published and you uploaded e.g. raw counts at GEO then you could simply use that. Somewhat helps with reproducibility as you can just pull an existing file and go. Also easy to publish your code based on publically-available files without the need to do any more processing.
Major build of the genome remains GRCh38. What you are looking at is the latest patch release for that build. Only changes are likely to be in annotations as noted below.