search

andrewrech / antigen.garnish

Other
46 stars 13 forks source link

Nomenclature MHC II #143

Closed nickhir closed 3 years ago

nickhir commented 3 years ago

I was wondering if you could elaborate how we should define the name of our MHC II alleles. If I take a look at the output of list_mhc() it seems like the MHCII alleles are defined like this:

...
[11703] "HLA-DPA10103-DPB10101"  "HLA-DPA10103-DPB10201"
[11705] "HLA-DPA10103-DPB10202"  "HLA-DPA10103-DPB10301"
[11707] "HLA-DPA10103-DPB10401"  "HLA-DPA10103-DPB10402"
[11709] "HLA-DPA10103-DPB10501"  "HLA-DPA10103-DPB10601"
[11711] "HLA-DPA10103-DPB10801"  "HLA-DPA10103-DPB10901"
[11713] "HLA-DPA10103-DPB110001" "HLA-DPA10103-DPB11001"
[11715] "HLA-DPA10103-DPB110101" "HLA-DPA10103-DPB110201"
[11717] "HLA-DPA10103-DPB110301" "HLA-DPA10103-DPB110401"
[11719] "HLA-DPA10103-DPB110501" "HLA-DPA10103-DPB110601"
[11721] "HLA-DPA10103-DPB110701" "HLA-DPA10103-DPB110801"
[11723] "HLA-DPA10103-DPB110901" "HLA-DPA10103-DPB111001"
[11725] "HLA-DPA10103-DPB11101"  "HLA-DPA10103-DPB111101"
[11727] "HLA-DPA10103-DPB111201" "HLA-DPA10103-DPB111301"
[11729] "HLA-DPA10103-DPB111401" "HLA-DPA10103-DPB111501"
[11731] "HLA-DPA10103-DPB111601" "HLA-DPA10103-DPB111701"
[11733] "HLA-DPA10103-DPB111801" "HLA-DPA10103-DPB111901"
[11735] "HLA-DPA10103-DPB112101" "HLA-DPA10103-DPB112201"
...

This confused me a little, because I cant find this nomenclature anywhere else on the internet, and the MHC I alleles are defined as usual. Furthermore, it seems like even common MHC II alleles like HLA-DQA1*01:01 are not supported. The MHC II allele that you listed in the example (HLA-DRB1*14:67) is also not supported. Can you maybe tell me how to "translate" the nomenclature HLA-DRB1*14:67 to the one that is required by antigen.garnish?

andrewrech commented 3 years ago

These are taken directly from the NetMHC prediction tools, e.g. from Docker

~/antigen.garnish/netMHC/netMHCIIpan-4.0/netMHCIIpan -list

It can be tricky to figure out what needs to be passed because the nomenclature is not consistent over alleles, time, or tools. In your case, try DRB1_1467.

~/antigen.garnish/netMHC/netMHCIIpan-4.0/netMHCIIpan -list | grep 1467

DRB1_1467

Once you have your allele list you can figure out how to modify the strings, which generally involves removing the * and possible prepending HLA.