Does anyone know how to get a list of all protein aliases for human proteins from Uniprot? The aliases are described here: https://www.uniprot.org/help/different_protein_gene_names
These can be found individually for each protein but I wanted a table / list for all of them
2 answers
You could also download them from UniProt as a table using: https://www.uniprot.org/uniprot/?query=proteome:UP000005640+AND+proteomecomponent:%22Chromosome+1%22&sort=score
Actual query uses:proteomecomponent:"chromosome 1" AND organism:"Homo sapiens (Human) [9606]" AND proteome:up000005640 (change chromosome number to get tables for all chromosomes).
Configure columns as needed.
You can configure the output format of your search result table to include the columns you want, and then download the results in tab- delimited format. This can be done for any query, including the wildcard query "*" which allows you to retrieve all entries.
See https://www.uniprot.org/help/customize, (video, at about 1min25) or https://insideuniprot.blogspot.com/2015/03) and follow these instructions to include a column for the protein name (or gene names), and remove all columns that you do not wish to see or download.
For example for all reviewed entries, reviewed:yes
default columns, or the columns you had previously selected: https://www.uniprot.org/uniprot/?query=reviewed%3Ayes&sort=score
most columns removed: https://www.uniprot.org/uniprot/?query=reviewed%3Ayes&columns=id%2Centry%20name%2Creviewed%2Cprotein%20names&sort=score
tab-delimited, first 1000 entries: https://www.uniprot.org/uniprot/?query=reviewed:yes&format=tab&limit=1000&columns=id,entry%20name,reviewed,protein%20names&sort=score
(remove "&limit=1000" to get all). This URL can of course be used programmatically (see https://www.uniprot.org/help/api).
BTW we also provide this precomputed file, for all reviewed (i.e. UniProtKB/Swiss-Prot) entries https://ftp.uniprot.org/pub/databases/uniprot/current_release/knowledgebase/complete/docs/shortdes.txt
Log in to answer this question.
I think you may have to download the entire UniProtKB and run a custom query.
I was hoping that wasn't going to be necessary....
Not from UniProt (from NCBI) see my answer here : A: Easiest way to homogenize gene aliases across datasets?
I don't think these line up, GenoMax. The UniProt aliases are full descriptions where NCBI has symbol aliases.
Following should produce descriptions as well: