I ran featurecounts from Galaxy GUI – it didn’t recognized genomic annotation UCSC from history. By no means! It runs on its own build in hg38, and it gave me (in table) not uc.3458…. ID but just numbers, which I cannot convert to Gene Symbols. What's the problem? How do you do it? Any suggestion?
Thanks
admin redacted email for privacy
Hello,
With FeatureCounts, when the option for Gene annotation file is set to "in your history", any datasets with a GTF or GFF datatype in the same history will be selectable as input. Maybe you need to change this option or the annotation dataset datatype assignment? https://galaxyproject.org/support/#getting-inputs-right
The FeatureCounts built-in annotations are described on the tool form (databases supported: hg38, hg19, mm10 and mm9) and in this post: https://biostar.usegalaxy.org/p/27461/
How to convert identifiers or add in gene symbols can vary by the tools available/installed on the Galaxy server you are working at. If at https://usegalaxy.org, the tool UniProt ID mapping and retrieval can convert IDs. Or you can upload any tabular mapping file to Galaxy that contains the IDs in your output and the IDs you want to convert to and merge the two with the tool Join two files or Join two Datasets side by side on a specified field.
Examples of joining data are in the Galaxy 101 tutorial: https://galaxyproject.org/learn/
All Support FAQs: https://galaxyproject.org/support/
Thanks! Jen, Galaxy team
Update
The Featurecounts tool now requires that the database metadata assignment is made to both the BAM and GTF inputs (when sourced from the history).
Please see this post for full details: https://biostar.usegalaxy.org/p/24154/#28027
Thanks! Jen, Galaxy team
