Restructure Annotation Procedure

TCP-Lab / bioTEA-box

The BioTEA's Docker container code

MIT License

0 stars 0 forks source link

Restructure Annotation Procedure #2

Open MrHedmad opened 1 year ago

MrHedmad commented 1 year ago

This issue was raised by @Feat-FeAR :

The internal annotations for human chips do not have linked scripts for re-generation. Additionally, there are other considerations to make:

Many Agilent microarrays do not have annotations as bioconductor packages, but only as GEO GPL accessions. This should be addressed by adding annotations for all Agilent arrays.
The annotated columns as gene symbols should be surpassed by ENSG when possible. This has all the benefits of using ENSGs instead of gene names, plus it better supports RNA-Seq, which by default is annotated with ENSGs in most pipelines.

MrHedmad commented 1 year ago

I want to add that the annotation module could be split from BioTEA if we make a small tool to interact with BioMart. Something simple and from the command line that could add annotations from BioMart given a .csv file with ENSGs. This would allow the user to easily add a variety of annotations (and not just those given by biotea annotate), and it would not be limited to BioTEA.

If we move to ENSGs, then the gene symbols should be hot-downloaded from BioMart at runtime.

MrHedmad commented 1 year ago

Looking at the issues again, this was already mentioned in a bioTEA issue: https://github.com/CMA-Lab/bioTEA/issues/1